KiCad 6 is coming! (has arrived!!)

[Karel]

I couldn't resist 

https://techexplorations.com/blog/kicad/kicad-6-review-new-and-improved-features/

[thinkfat]

Manually exporting the net list was never a thing with KiCad 5 either. I cannot remember ever having to do it. I always used "Update PCB from Schematic".

[Whales]

Kicad 6?  Chrome is already up to version 88 and Firefox 84, Kicad really needs to catch up.

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.


Thinkfat is right about the update PCB from schematic stuff:

To be fair, in KiCad 5, the “Update PCB from Schematic” button did exist, under the Tools menu. But in KiCad 6, the transition to a netlist file-less workflow is complete.

What was incomplete about it previously?  It works fine in 5.1.8 and it's on the toolbar, not just hidden in a menu.  Maybe an error/oversight in this article (it's a really big article)?

[janoc]

Manually exporting the net list was never a thing with KiCad 5 either. I cannot remember ever having to do it. I always used "Update PCB from Schematic".

You still could do it and it was sometimes useful - e.g. when you didn't want to delete some components that weren't in the schematic on netlist update. And before version 5 it was the only way to do things. The "Update PCB" button only automated the workflows of exporting the netlist and reloading it in the PCB editor.

[SiliconWizard]

Kicad 6?  Chrome is already up to version 88 and Firefox 84, Kicad really needs to catch up.

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.


Thinkfat is right about the update PCB from schematic stuff:

To be fair, in KiCad 5, the “Update PCB from Schematic” button did exist, under the Tools menu. But in KiCad 6, the transition to a netlist file-less workflow is complete.

What was incomplete about it previously?  It works fine in 5.1.8 and it's on the toolbar, not just hidden in a menu.  Maybe an error/oversight in this article (it's a really big article)?

I'm not quite sure yet how far they have gone with this "change". Yes, it was already available before. BUT... I think behind the scene, they were still using netlists. I suspect the change is that now they are using some kind of internal data instead of netlists, or something.

I remember it was discussed on the Kicad forum last year - particularly, I was making the point that netlist -> PCB was a useful feature for people willing to use the PCB editor without using the schematic editor (I was working on a schematic-less design tool that would export to a Kicad netlist).

Some people on the forum assured me they would just make the schematic -> PCB more "seamless" (still not completely sure what they really added in v6), but that they wouldn't remove the "netlist import" feature in PCBnew. But I was still concerned that they would now have little incentive to maintain the netlist import feature, and that in time, it might become not up-to-date and eventually would disappear. Which would frankly SUCK.

Anyone knows for sure if the netlist import in PCBnew will still be supported in v6?

[janoc]

I doubt that they removed the netlists completely - netlist is not only for the PCB design but also for things like circuit simulation (which, BTW, KiCAD supports as well).

In the new version the manual netlist generation is available under File->Export->Export Netlist from the schematic editor.

[gnuarm]

I haven't checked the version 6 release but I believe it now contains the Altium importer.  We had to use that on a project recently by using the nightly builds and it worked... not perfectly, but better than naught.  I expect a few more of the issues have been resolved with the release.

[james_s]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

[phil from seattle]

In reading the article, I get the sense that this is more an internal change (S expressions uber alles) and some visual candy.  Now that I've got all the obscure aspects memorized, time to change! Hopefully it will be easy.

It doesn't look like they put effort into simplifying the libraries - it took me a while to figure it all out but is way more convoluted than it needs to be.  Especially if you want to distribute a kicad design. There really needs to be a button that packages up all the lib elements for distribution. Something that Eagle does automatically since like 2005. I find missing symbols and footprints in all sorts of Kicad files out there. Annoying to have to chase them down, if you even can. Still haven't found the Pico's USB connector symbol.

I also hope they made managing design revisions easier but it doesn't look like they did. Again, something that is trivially easy in Eagle.

[james_s]

IIRC there is a library cache file you can package up that does more or less that. It's certainly not the most convenient or intuitive thing though. KiCad has warts for sure, although I'm used to most of them by now and Eagle has one much bigger and nastier wart that I can't work around, that is the licensing.

[pidcon]

This is very good news. I hope in the future when the software is as mature as other established open source software, they would offer a Long Term Support (LTS) version.

[gsocker]

It doesn't look like they put effort into simplifying the libraries - it took me a while to figure it all out but is way more convoluted than it needs to be.  Especially if you want to distribute a kicad design. There really needs to be a button that packages up all the lib elements for distribution. Something that Eagle does automatically since like 2005. I find missing symbols and footprints in all sorts of Kicad files out there. Annoying to have to chase them down, if you even can. Still haven't found the Pico's USB connector symbol.

The new schematic file format embeds the symbols used. This should completely eliminate the missing symbols issue. From the PCB side, footprints have been embedded in the PCBnew file since at least V5.

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

Color schemes are now JSON files and can be selected in the preferences. The EESchema default background color (if an older configuration wasn't imported) is now something along the lines of the one in "Solarized Light" themes.
There is a collection at https://github.com/pointhi/kicad-color-schemes

[janoc]

This is very good news. I hope in the future when the software is as mature as other established open source software, they would offer a Long Term Support (LTS) version.

You mean a software that has been around for over 30 years (not kidding - KiCAD was started in 1992) and has a major and still backwards compatible release only every 2-3 years (e.g. Kicad 4 series started in 2015, Kicad 5 in 2018 ...) is not mature "as other established open source software" and needs an LTS version? Seriously? 

What for, exactly? And if you still do need one (even though you can always keep an installer of an old version around - it is not like an OS which if you use an old version you could get instantly hacked) - are you willing to pay for it?

Maintaining an old version of the software costs developer time and money. KiCAD isn't something like Ubuntu where there is a corporate sponsor with a millionaire owner behind it financing it and where the long term support is needed because you can't change your server OS every 3 years. And even the Ubuntu LTS versions are supported only for 5 years from the date of first release.

Another case in the point - Qt toolkit developers TrollTech have recently announced that their LTS versions will be available only to paying clients. Guess why. It does cost a non-trivial amount of resources to maintain.

[Yansi]

My main concern is, whether the new 6th version of KiCAD can:

a) SELECT MULTIPLE OBJECTS and CHANGE a COMMON PROPERTY  (like a text size, via diameter,  trace width, etc.). Without this, it is still a toy PCB CAD. Having to click hundreds of objects one by one, making 5 clicks or so to change a trace width or whatever, is beyond annoying.

b) MAKE CURVED TRACES.  My first task in KiCAD at my former employer was to draw a curved flex-PCB and it was almost an impossible job to do. I needed to hack around it by drawing the stuff in a graphical editor, export in DXF, import in some other (not Cu, mind you) layer and then manually switch the imported lines to a F.Cu / B.Cu layer piece by piece, arc by arc.  God I hated the KiCAD since! 

c) So does it finally allow to import DXF into a copper layer?  Someone ingenious decided that the import DXF dialog does not contain a copper layer as a selection. 

d) Allow a graphical object to become an electrical one too. Like for example, how is one supposed to connect a PCB antenna pattern (or any other distributed element structure) to an electrical net? Even if you manage to go through b and c steps above, you still can not connect it electrically and need to use hacks.

e) Allow to switch the annoying quadrant IV operation into a Quadrant I, such as almost all CADs use. Yes, I am talking you, the flipped Y axis.

These are some of the stuff why I have not considered using the KiCAD so far personally, but hell I would like to!

So do you know if any of the stuff above got any improvement or at least an option is available?

[thinkfat]

My main concern is, whether the new 6th version of KiCAD can:

a) SELECT MULTIPLE OBJECTS and CHANGE a COMMON PROPERTY  (like a text size, via diameter,  trace width, etc.). Without this, it is still a toy PCB CAD. Having to click hundreds of objects one by one, making 5 clicks or so to change a trace width or whatever, is beyond annoying.

b) MAKE CURVED TRACES.  My first task in KiCAD at my former employer was to draw a curved flex-PCB and it was almost an impossible job to do. I needed to hack around it by drawing the stuff in a graphical editor, export in DXF, import in some other (not Cu, mind you) layer and then manually switch the imported lines to a F.Cu / B.Cu layer piece by piece, arc by arc.  God I hated the KiCAD since! 

c) So does it finally allow to import DXF into a copper layer?  Someone ingenious decided that the import DXF dialog does not contain a copper layer as a selection. 

d) Allow a graphical object to become an electrical one too. Like for example, how is one supposed to connect a PCB antenna pattern (or any other distributed element structure) to an electrical net? Even if you manage to go through b and c steps above, you still can not connect it electrically and need to use hacks.

e) Allow to switch the annoying quadrant IV operation into a Quadrant I, such as almost all CADs use. Yes, I am talking you, the flipped Y axis.

These are some of the stuff why I have not considered using the KiCAD so far personally, but hell I would like to!

So do you know if any of the stuff above got any improvement or at least an option is available?

a) yes, but only for traces (edit: also vias), it doesn't seem to work for text items.
b) yes, curved traces are available now.
c) yes
d) no. while you can draw e.g. arcs now on a copper layer (and import graphics items into copper layers), you cannot assign a net to them. you can connect traces, but it might fail DRC then. The items are recognized as obstacles by the interactive router, though.
e) yes

What I'm still missing is via fences. It's the primary reason I didn't yet ditch KiCAD 5.

[gsocker]

My main concern is, whether the new 6th version of KiCAD can:

d) Allow a graphical object to become an electrical one too. Like for example, how is one supposed to connect a PCB antenna pattern (or any other distributed element structure) to an electrical net? Even if you manage to go through b and c steps above, you still can not connect it electrically and need to use hacks.

So do you know if any of the stuff above got any improvement or at least an option is available?

You can't import copper and assign a net in PCBnew, but I suspect the following will work (don't have anything to test with it, though):

• Insert symbol into schematic with pins for connection points and connect nets
• Create footprint
• Import DXF into appropriate copper layer in footprint
• Create custom pads in the appropriate spots, numbered to match the schematic symbol
• Assign created footprint to the schematic symbol

At that point you should be able to route tracks to the custom pad points.

[janoc]

d) no. while you can draw e.g. arcs now on a copper layer (and import graphics items into copper layers), you cannot assign a net to them. you can connect traces, but it might fail DRC then. The items are recognized as obstacles by the interactive router, though.

A not so dumb workaround for something like this (e.g. a PCB antenna) is to first create a footprint/component for your graphic item and design the copper parts as pads. Then it will work fine.

[Yansi]

Well, it does not work as easy as you  say. Try it!

If you for example need to connect the pad to a ground pour, it won't work. The ground pour will avoid filling to it due to seeing the graphical line at/next to the pad as an obstacle.  Same with the interactive router. You will be fighting with it to get it connected.

Even if you create it as a footprint. You can try for example the 2.4 GHz IFA PCB antenna, that is included somewhere in the KiCAD libs.

[madires]

Interesting new features! However, I hope the limited naming scheme for bus signals will be changed into a more practical one. The current bus[0..n] is quite cumbersome in most cases. No way to create an SPI bus with comprehensible signal names, for example.

[Yansi]

My main concern is, whether the new 6th version of KiCAD can:

a) SELECT MULTIPLE OBJECTS and CHANGE a COMMON PROPERTY  (like a text size, via diameter,  trace width, etc.). Without this, it is still a toy PCB CAD. Having to click hundreds of objects one by one, making 5 clicks or so to change a trace width or whatever, is beyond annoying.

b) MAKE CURVED TRACES.  My first task in KiCAD at my former employer was to draw a curved flex-PCB and it was almost an impossible job to do. I needed to hack around it by drawing the stuff in a graphical editor, export in DXF, import in some other (not Cu, mind you) layer and then manually switch the imported lines to a F.Cu / B.Cu layer piece by piece, arc by arc.  God I hated the KiCAD since! 

c) So does it finally allow to import DXF into a copper layer?  Someone ingenious decided that the import DXF dialog does not contain a copper layer as a selection. 

d) Allow a graphical object to become an electrical one too. Like for example, how is one supposed to connect a PCB antenna pattern (or any other distributed element structure) to an electrical net? Even if you manage to go through b and c steps above, you still can not connect it electrically and need to use hacks.

e) Allow to switch the annoying quadrant IV operation into a Quadrant I, such as almost all CADs use. Yes, I am talking you, the flipped Y axis.

These are some of the stuff why I have not considered using the KiCAD so far personally, but hell I would like to!

So do you know if any of the stuff above got any improvement or at least an option is available?

a) yes, but only for traces (edit: also vias), it doesn't seem to work for text items.
b) yes, curved traces are available now.
c) yes
d) no. while you can draw e.g. arcs now on a copper layer (and import graphics items into copper layers), you cannot assign a net to them. you can connect traces, but it might fail DRC then. The items are recognized as obstacles by the interactive router, though.
e) yes

What I'm still missing is via fences. It's the primary reason I didn't yet ditch KiCAD 5.

Thank you for answering, seems there indeed is some improvement. Even though the d) is not solved yet, it is a rather niche problem and there are some haccs that can be utilized for that.

I may give it a try. I would like to use that tool, it is overall a very capable one, but those little things are always the most irritating ones.

[gsocker]

Interesting new features! However, I hope the limited naming scheme for bus signals will be changed into a more practical one. The current bus[0..n] is quite cumbersome in most cases. No way to create an SPI bus with comprehensible signal names, for example.

This is supported, but I don't think the documentation has been updated yet. The old syntax is still supported, and it's also possible to do something like SRAM{RW CS ACK A[0..16] D[0..16]}

[julian1]

The move to s-expressions inspires confidence! (better than json or yaml) Should help a lot for readability, scripting and manual text editing.

I also like that netclasses can be managed from eeschema, not just pcbnew. Identifying and tagging nets with associated netclasses is a big kicad painpoint. 

[futureshocked]

I have updated the article to clarify this, thank you.

[futureshocked]

In KiCad 6, you can still use Netlist the netlist file. You can find it under the File menu (I have edited the article to mention this).

[thinkfat]

Well, it does not work as easy as you  say. Try it!

If you for example need to connect the pad to a ground pour, it won't work. The ground pour will avoid filling to it due to seeing the graphical line at/next to the pad as an obstacle.  Same with the interactive router. You will be fighting with it to get it connected.

Even if you create it as a footprint. You can try for example the 2.4 GHz IFA PCB antenna, that is included somewhere in the KiCAD libs.

It should work if you create a footprint with a custom-shaped pad. If you import the antenna structure into the footprint, place a regular pad e.g. on one leg, then select the whole and then there's an option somewhere to make the whole thing into a pad with arbitrary shape. I remember I made a spark gap with this trick. That's not even new, it was already available in KiCAD 5, only you couldn't import graphics onto the copper layer, IIRC.

[james_s]

The difficulty with putting graphics on copper in KiCad is something I've never understood. I see no logical reason to allow it in silkscreen and not copper, the layers are handled the same way.

[futureshocked]

Hi all,

I have spent the last two days updating the article.

There’s still work to do, but want to:

    Let you know there’s a lot of new content as suggested by the contributors in this discussion topic.
    To thank you for helping me make the article as complete as possible.

More to come in a few days.

[futureshocked]

James, you can set colors (including the background) to whatever you want. I went with the default.

[2N3055]

Is object selection in EEschema fixed so it doesn't move selection automatically when you drag select?

[Karel]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

You can set whatever color you want in V5.
I don't like white but I also don't like a dark background for the schematic editor. So I set it to the same as Eagle: RGB 238, 238, 206.
It looks old fashioned but it's perfect for me, no eyestrain anymore.

[thinkfat]

Is object selection in EEschema fixed so it doesn't move selection automatically when you drag select?

Not sure what you're referring to - there is in V5 eeschema a specific behavior, that when you select an area, it automatically goes into drag/move mode. Is that what you're referring to? If yes, that's been changed (fixed?). It's now slightly less convenient to move parts of the schematic around during cleanup. You need to hit "m" to start moving the selection.

[2N3055]

Is object selection in EEschema fixed so it doesn't move selection automatically when you drag select?

Not sure what you're referring to - there is in V5 eeschema a specific behavior, that when you select an area, it automatically goes into drag/move mode. Is that what you're referring to? If yes, that's been changed (fixed?). It's now slightly less convenient to move parts of the schematic around during cleanup. You need to hit "m" to start moving the selection.

Yes ,that is what I was referring to. Thank you for the answer!!

I understand what you are saying, but that also made it inconsistent with all other tools in the world and PCBnew itself in how it treats mouse behaviour. It was easier to move and infuriating for everything else..

Anyways, that is, for me, good news. Once 6.0 is released,I might try doing some simpler projects in Kicad to see how am I going to get along with it..

[futureshocked]

Hmm, I am not sure what you mean. Could you clarify so I can do a test for you?

[futureshocked]

Interesting new features! However, I hope the limited naming scheme for bus signals will be changed into a more practical one. The current bus[0..n] is quite cumbersome in most cases. No way to create an SPI bus with comprehensible signal names, for example.

This is supported, but I don't think the documentation has been updated yet. The old syntax is still supported, and it's also possible to do something like SRAM{RW CS ACK A[0..16] D[0..16]}
[/quote]

This is something that I find difficult to find information about. Is there any information you can provide to help me fill in this stub in the article? (https://techexplorations.com/blog/kicad/kicad-6-review-new-and-improved-features/#t-1611032693203)

[Warhawk]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

You can set whatever color you want in V5.
I don't like white but I also don't like a dark background for the schematic editor. So I set it to the same as Eagle: RGB 238, 238, 206.
It looks old fashioned but it's perfect for me, no eyestrain anymore.

Just for inspiration below is my color scheme. I downloaded it somewhere from github.

[phil from seattle]

Impressive review.  Thanks to Peter for doing it.  Kicad 6 looks to be far better than I initially thought.

So, the million dollar question - is it stable enough to switch to full time?  Guess I need to get it and see for myself.

[gsocker]

Interesting new features! However, I hope the limited naming scheme for bus signals will be changed into a more practical one. The current bus[0..n] is quite cumbersome in most cases. No way to create an SPI bus with comprehensible signal names, for example.

This is supported, but I don't think the documentation has been updated yet. The old syntax is still supported, and it's also possible to do something like SRAM{RW CS ACK A[0..16] D[0..16]}

This is something that I find difficult to find information about. Is there any information you can provide to help me fill in this stub in the article? (https://techexplorations.com/blog/kicad/kicad-6-review-new-and-improved-features/#t-1611032693203)
[/quote]
I had to digging for this one, they were introduced quite a while ago : https://github.com/KiCad/kicad-doc/blob/d27c4009275d2801da932cf0d0dfdccd4448ed58/src/eeschema/eeschema_schematic_creation_and_editing.adoc#wires-buses-labels-power-ports

Edit:Fix link

[futureshocked]

Interesting new features! However, I hope the limited naming scheme for bus signals will be changed into a more practical one. The current bus[0..n] is quite cumbersome in most cases. No way to create an SPI bus with comprehensible signal names, for example.

This is supported, but I don't think the documentation has been updated yet. The old syntax is still supported, and it's also possible to do something like SRAM{RW CS ACK A[0..16] D[0..16]}

I have just updated the article with information on the new bus capabilities.

[futureshocked]

Some bugs found right after I downloaded the latest nightly PPA release:

1. Footprint editor will not display grid correctly.
2. Schematic library conversion gets the library lost.

I didn't test further before removing it. I'll wait for its final release.

Would be really cool if you could raise issues here: https://gitlab.com/kicad/code/kicad/-/issues

[futureshocked]

In reading the article, I get the sense that this is more an internal change (S expressions uber alles) and some visual candy.  Now that I've got all the obscure aspects memorized, time to change! Hopefully it will be easy.

It doesn't look like they put effort into simplifying the libraries - it took me a while to figure it all out but is way more convoluted than it needs to be.  Especially if you want to distribute a kicad design. There really needs to be a button that packages up all the lib elements for distribution. Something that Eagle does automatically since like 2005. I find missing symbols and footprints in all sorts of Kicad files out there. Annoying to have to chase them down, if you even can. Still haven't found the Pico's USB connector symbol.

I also hope they made managing design revisions easier but it doesn't look like they did. Again, something that is trivially easy in Eagle.

It seems that in KiCad 6, the schematic and layout files will contain all symbols and footprints. No dependencies on external libraries. When you share the project, all symbols and footprints will travel along. Also, the main KiCad window has a function under the File menu to create a ZIP archive of the project (which you can share with others).

[futureshocked]

Impressive review.  Thanks to Peter for doing it.  Kicad 6 looks to be far better than I initially thought.

So, the million dollar question - is it stable enough to switch to full time?  Guess I need to get it and see for myself.

Well... I haven't had any work lost while playing with the nightly build. Only various annoying issues that I have reported them as issues for fixing. The file management and libraries look solid.

I am mindful that the program changes daily, so I work on copies of my projects.

I would wait a bit longer for serious work though. KiCad devs are careful, but it only takes one bad commit to break something important.

[TheNewLab]

It HERE!!!
At least with my update on Mint Linux OS. It just downloaded it, after asking to change the direct repo link.

 Thought I was just getting some bug fixes, although I have not yet looked at it.

[bson]

Another excellent new feature is the ability to drag a footprint - and have the P&S router drag the traces connected to it.  Makes it much less work to scoot a footprint over a little.

[SiliconWizard]

Another excellent new feature is the ability to drag a footprint - and have the P&S router drag the traces connected to it.  Makes it much less work to scoot a footprint over a little.

Sounds great!

[nigelwright7557]

While kicad is generally ok and has good libraries I did spot a huge clanger in their software.
If I try to layout a pcb it just puts the pcb parts in a line across the top of the pcb.
This is a real pain.
I want the parts laid out as they were in schematic with associated components next to each other !

With kicads way I have to move all the parts onto the pcb then work out what should be close to what !
What a useless pile of rubbish.
Still you pay your money (or not) and take your choice.

kicads way is maybe ok for half a dozen components but for something bigger just plain stupid.

[retiredfeline]

I want the parts laid out as they were in schematic with associated components next to each other !

That's what happens now, the positions mirror those in the schematic. Your observation is out of date.

[thinkfat]

[nigelwright7557]

I want the parts laid out as they were in schematic with associated components next to each other !

That's what happens now, the positions mirror those in the schematic. Your observation is out of date.

I just downloaded latest version and thats how it did it.

Nearly as sensible as the dark red sheet layout on black !

[retiredfeline]

Install the schematic locations plugin and use it. It'll probably be standard behaviour in V6.

Nobody pays any attention to the sheet header in Pcbnew anyway but you can change the colour.

[Monkeh]

What a useless pile of rubbish.
Still you pay your money (or not) and take your choice.

Yes, we've all already seen that any software you didn't write is crap. How dare anyone attempt to write something else and not have the same priorities or preferences as you.

[phil from seattle]

While kicad is generally ok and has good libraries I did spot a huge clanger in their software.
If I try to layout a pcb it just puts the pcb parts in a line across the top of the pcb.
This is a real pain.
I want the parts laid out as they were in schematic with associated components next to each other !

With kicads way I have to move all the parts onto the pcb then work out what should be close to what !
What a useless pile of rubbish.
Still you pay your money (or not) and take your choice.

kicads way is maybe ok for half a dozen components but for something bigger just plain stupid.

Better in 6 but even in 5.x, it isn't that hard to do. I use a dual monitor set up so YMMV. If you don't have at least 2 monitors, you are missing out.

Open schema on one screen, PCB New on the other with an empty pcb.
Update PCB from schematic to get the components.
Turn off ratsnest (aka airwires) to make it clearer.
Click on a part in the schematic and it highlights in pcbnew, move it where you want it and repeat until every thing is in place.

I just did a 250 part board and took about 30 minutes to get the major groups set up.  Once I have that, I start juggling them to find the best fit. Schematic order isn't always that helpful, especially if you are using hierarchical sheets and labels.

It is a lot easier than Eagle.

[madires]

... or you can use the search function. Search for "C117" and you'll get your C117. Footprint auto-placement can be quite annoying when you have your own idea on how to place them. They get spread out all over the place, sometimes blocking the area where you want to place something else or they are at the wrong location. As more components you have as cluttered it becomes. However, auto-placement of a group of components would be helpful. Maybe something like: "auto-place the components of the power supply section here". That would give you a good starting point to place the components of that group. It also would follow the workflow of most users, since they work component by component, and not all components at once.

[Karel]

Even better:

Put the schematic on the first monitor and put the pcb layout on the second monitor.

In Pcbnew, press 't'. Enter "c413" and, voila, the component is immediately attached to your mouse cursor. No need to pan/zoom.
While it is attached to your mouse cursor, press 'r' to rotate 90 degrees and press 'f' to flip the compononent to the other side of the board.
At least this works in V5.

[janoc]

Even better:

Put the schematic on the first monitor and put the pcb layout on the second monitor.

In Pcbnew, press 't'. Enter "c413" and, voila, the component is immediately attached to your mouse cursor. No need to pan/zoom.
While it is attached to your mouse cursor, press 'r' to rotate 90 degrees and press 'f' to flip the compononent to the other side of the board.
At least this works in V5.

Don't bother arguing with Nigel. This thread should tell you everything you need about how a "proper" PCB tool is supposed to operate:

https://www.eevblog.com/forum/buysellwanted/pcb-design-software-with-components-autoplacers/

The board examples made with it were telling too but sadly they don't seem to be available anymore.

[phil from seattle]

@Janoc, thanks for the link.  ugh.  I find "trash your competition" as the worst kind of marketing. And, not explaining that one is the author of a competing product is borderline unethical.

[janoc]

@Janoc, thanks for the link.  ugh.  I find "trash your competition" as the worst kind of marketing. And, not explaining that one is the author of a competing product is borderline unethical.

It's not even the trash talk (everyone seems to love to slag KiCAD) but Nigel has some very "special" ideas how modern software should work (like two pages of argument on why only idiots need an undo function), so getting into a heated debate with him is going to be a waste of time. Oh and calling free_electron an "amateur" after some constructive criticism of the software by him was priceless too. 

[Uky]

I grew up using Mentor BoardStation (Unix on SUN/Sparc) and have been using Expedition/VX and Cadence 15.xx to 17.40. I wonder why all low cost (or free) tools are so in love with bit-mapped graphics? KiCAD, which seems to be a nice free alternative is no exception. I have been poking around to see if there were any setting which would provide a "single-pixel-width" schematic editor with thinnest possible lines for symbols as well as connection lines. At least I found the "skeletal view" setting in the layout editor but the schematic editor seems to have no such feature. If wanting to make symbol lines thinner, if not mistaken I have to edit every symbol in the library...

[nfmax]

On high DPI displays, a single pixel wide line would be invisible. And a lot of us still have low DPI displays. You have to choose a line width somehow

[bson]

I can't think of anything in KiCad that's pixel based, except specifically images imported as graphics.  Everything else is vectored, based on actual dimensions.

What KiCad *could* use is dimensional calibration and the ability to display something at its actual size.  Including 3D renderings.  The only way to currently do that is to print at 1:1, save as a PDF, and open that in e.g. MacOS Preview.  That'll yield the exact size, just as if it were printed (which on even a cheap a laser printer will be quite accurate).

Also, like F360 it would be very useful to be able to import an image, calibrate for dimensions, and use as a backdrop (or in the case of F360, sketch canvas).  This would make it very easy to create an outline to fit a specific chassis or enclosure, and get mounting holes, tabs, and other mechanicals right - without leaving KiCad and creating a DXF in some other tool.

[nfmax]

Also, like F360 it would be very useful to be able to import an image, calibrate for dimensions, and use as a backdrop (or in the case of F360, sketch canvas).  This would make it very easy to create an outline to fit a specific chassis or enclosure, and get mounting holes, tabs, and other mechanicals right - without leaving KiCad and creating a DXF in some other tool.

Yes please! This would make copying an existing PCB from an image so much easier, especially now that 'schematicless' workflow will be supported. It needs a general affine transform of the backdrop, to cope with geometrical image distortion. Supporting front and back images would be handy, too

[Anders Petersson]

"In KiCad 6, copy and paste works across Pcbnew and Eeschema instances."

This is huge, guys.

[emece67]

.

[tbavcevic]

sorta long time eagle user, have been forcing myself to try kicad every major release and now and then, i want to use it and like it but it is still unfinished software, lots of bugs, lots of changes from version to version that seems to backfire as old bugs stay and new are introduced, waste of time since sometimes you loose your data, very unstable overall. tried pointing out bugs on their forum yet posts get removed, targeted moderation. they are using people as beta testers and i know it is open source but feels like at first sign of somewhat bug free release they will probably somehow try to sell it.

rant off: do anyone know how to setup kicad libraries on ubuntu 20.10, nothing works. thanks.

[Karel]

sorta long time eagle user, ...
... they are using people as beta testers.

Cadsoft Eagle was indeed very stable. The autodesk Eagle is terrible.
Nowadays all software uses their users as betatesters, commercially or not. Welcome in the age of agile programming.

... and i know it is open source but feels like at first sign of somewhat bug free release they will probably somehow try to sell it.

Everybody is allowed to sell open source (GPL) software.
What you can't do is sell GPL'ed software without the source code.
So, you will sell one copy and the buyer will distribute if for free to the whole world.
And no, you cannot magically change GPL'ed software into closed source.

[tbavcevic]

yes autodesk f*ked up eagle, thats why i am rooting for kicad.but they threatening me on official kicad forums that they will ban me for saying on their forums that damned thing dont work for me, well this is interesting, maybe they are for real forcing only positive feedback posts.

thing is, lets say eagle freezes up or i i have brownout and my pc restarts in middle of routing some pcb and i forgot to save for few hours, with eagle there are autosaves, happened many times that those saved me few hours of re-work, and i am used to that "protective" approach, safety first. with kicad everything ok, saved many times, open board next time there is no components?! or making some custom part waste whole morning on it, save it and tomorow its gone, like i cant understand is it that i am so dumb that i cant use it or is it realy not well designed.

[thinkfat]

If your posts to the KiCad forum were all like the post above, I need not ask why you were threatened with a lifelong ban.

[phil from seattle]

What part of Free don't you understand?

Yes, software should do auto saves to protect us boneheads that forget to.  Except, when auto save causes crashes. Sketchup had a notorious bug where autosave caused crashes. Took them a long time to issue a fix. Most people just turned it off. Problem solved, heh heh heh.

[MitjaN]

sorta long time eagle user, have been forcing myself to try kicad every major release and now and then, i want to use it and like it but it is still unfinished software, lots of bugs, lots of changes from version to version that seems to backfire as old bugs stay and new are introduced, waste of time since sometimes you loose your data, very unstable overall. tried pointing out bugs on their forum yet posts get removed, targeted moderation. they are using people as beta testers and i know it is open source but feels like at first sign of somewhat bug free release they will probably somehow try to sell it.

rant off: do anyone know how to setup kicad libraries on ubuntu 20.10, nothing works. thanks.

As you've already figured out, nothing works, therefore you'll not be able to setup Kicad libraries on ubuntu 20.10

[janoc]

sorta long time eagle user, ...
... they are using people as beta testers.

Cadsoft Eagle was indeed very stable. The autodesk Eagle is terrible.
Nowadays all software uses their users as betatesters, commercially or not. Welcome in the age of agile programming.

Now I am no fan of agile but this has nothing to do with agile programming.

If you have used any of Autodesk products (AutoCAD, Inventor, 3DS Max, Fusion360 ... maybe with an exception of Maya that still seems to be resisting the "autodeskization" somewhat), you would have known that this is their way of working. Even products that they have got by acquisitions (most of them, in fact - Maya, Max, Eagle, Revit ...) get "molded" into this mess over time.

The UIs are designed by monkeys, the code APIs are an inconsistent buggy mess, there are extremely weird design decisions in the APIs and UIs that don't make any sense except to that outsourced programmer somewhere in India that was hired to code it, etc.

And don't get me started on how slow, resource hungry and buggy their products are ...

With the Fusion360 suite it is even worse than usual because that is literally a test bed for features that will be rolled into their more expensive products like Inventor later. Which was ok while they weren't asking money for it but now you are literally paying to be a beta tester, with stuff that worked one week breaking horribly the week later.
 
Unfortunately, I have the "pleasure" to have to deal with their stuff on a regular basis because we need it at work.

[gnuarm]

If your posts to the KiCad forum were all like the post above, I need not ask why you were threatened with a lifelong ban.

Jeez, this place is starting to look like usenet!

 

[JohnnyMalaria]

yes autodesk f*ked up eagle, thats why i am rooting for kicad.but they threatening me on official kicad forums that they will ban me for saying on their forums that damned thing dont work for me, well this is interesting, maybe they are for real forcing only positive feedback posts.

thing is, lets say eagle freezes up or i i have brownout and my pc restarts in middle of routing some pcb and i forgot to save for few hours, with eagle there are autosaves, happened many times that those saved me few hours of re-work, and i am used to that "protective" approach, safety first. with kicad everything ok, saved many times, open board next time there is no components?! or making some custom part waste whole morning on it, save it and tomorow its gone, like i cant understand is it that i am so dumb that i cant use it or is it realy not well designed.

Well, let's think about it for a nanosecond...how's that their fault? Save your work frequently. 

[gnuarm]

yes autodesk f*ked up eagle, thats why i am rooting for kicad.but they threatening me on official kicad forums that they will ban me for saying on their forums that damned thing dont work for me, well this is interesting, maybe they are for real forcing only positive feedback posts.

thing is, lets say eagle freezes up or i i have brownout and my pc restarts in middle of routing some pcb and i forgot to save for few hours, with eagle there are autosaves, happened many times that those saved me few hours of re-work, and i am used to that "protective" approach, safety first. with kicad everything ok, saved many times, open board next time there is no components?! or making some custom part waste whole morning on it, save it and tomorow its gone, like i cant understand is it that i am so dumb that i cant use it or is it realy not well designed.

Well, let's think about it for a nanosecond...how's that their fault? Save your work frequently. 

Auto save is a very useful feature and only the poorest of software designs don't include it.  Blaming users for not being machines is a bit silly.  Yeah, I'm supposed to check the air in my tires every month, get a physical checkup every year and never, ever cross the beams.  But we are human, not machines and in particular, when performing work that requires intense focus on the not so trivial details find it hard to set a biological timer for 10 minutes without fail. 

Two things in particular make it hard to remember to save the work every few minutes.  One is that computers don't screw up as often as they used to providing less reinforcement for the frequent saves.  The other is that the vast majority of software these days provides auto saving making the need for manual saves much less and so hard to remember with these fewer and further between programs.

It just strikes me as absurd for blaming a user for the problems in software.   

[phil from seattle]

It just strikes me as absurd for blaming a user for the problems in software.   

Uh, so this is what you appear to be saying - if I know that a given program I am relying on doesn't autosave and I don't save as I am going along, it isn't my fault if it crashes and I lose my work. You don't see anything slightly absurd with that?

What ever happened to personal responsibility???

[gnuarm]

It just strikes me as absurd for blaming a user for the problems in software.   

Uh, so this is what you appear to be saying - if I know that a given program I am relying on doesn't autosave and I don't save as I am going along, it isn't my fault if it crashes and I lose my work. You don't see anything slightly absurd with that?

What ever happened to personal responsibility???

What ever happened to not putting works in other peoples' mouths? 

Should I invent something for how I interpret your post and question that?

[phil from seattle]

It just strikes me as absurd for blaming a user for the problems in software.   

Uh, so this is what you appear to be saying - if I know that a given program I am relying on doesn't autosave and I don't save as I am going along, it isn't my fault if it crashes and I lose my work. You don't see anything slightly absurd with that?

What ever happened to personal responsibility???

What ever happened to not putting works in other peoples' mouths? 

Should I invent something for how I interpret your post and question that?

OK, show me how I misinterpreted your comment.  I'm willing to learn.

[gnuarm]

It just strikes me as absurd for blaming a user for the problems in software.   

Uh, so this is what you appear to be saying - if I know that a given program I am relying on doesn't autosave and I don't save as I am going along, it isn't my fault if it crashes and I lose my work. You don't see anything slightly absurd with that?

What ever happened to personal responsibility???

What ever happened to not putting works in other peoples' mouths? 

Should I invent something for how I interpret your post and question that?

OK, show me how I misinterpreted your comment.  I'm willing to learn.

Not going to engage in a pointless debate.  English is not an overly complex language.  What I wrote was clear.  What you wrote was your spin on it.  For starters you cut off most of what I posted. 

I'm not going around with you on this.  That would be of no value to either of us and not appropriate here.

[tbavcevic]

lets be clear i want kicad to work and be replacement for all people that have no money to buy altium or similar. but with behavior surounding it i am afraid it will never happen.

If your posts to the KiCad forum were all like the post above, I need not ask why you were threatened with a lifelong ban.

how is sincere feedback on bug report forums bad? should not they expect only negative feedback i n order to make it better, well i give only negative feedback on bug report forums (pun intended).

Well, let's think about it for a nanosecond...how's that their fault? Save your work frequently. 

i am paranoid and have double backups on different drives and sometimes triple for important stuff, save regularly and using ups for brownout protections but sometimes i waste few hours on something and finaly make board work and routable and then in second i would like to save pc freezes/eagle freezes or something else happens, but eagle leaves me with .bXX and .sXX autosaves that are somewhat recent and i need not waste more than few minutes and i am back in bussines exaclty where i left of, this is realy nice. with kicad if all is ok and saved on 4 places that is not guarantee that all data will be there! its absurd!

What part of Free don't you understand?
....

if some software is free does that mean that i am not allowed to say if it does not work for me? interesting concept, repeated on kicad forums each time someone asks why something does not work.

[Karel]

Here's how I deal(t) with it, both Eagle and KiCad.
I disabled the auto-save functionality.
I press ctl-s after making a (small) modification.
Every hour or so, depending on the kind of modification, I enter "git commit -a" and "git push" in the terminal.
I never forget it because I trained myself to do this and it became a habit and now I do it without thinking.
I never lost a (part of a) project.

You can create and setup free, private git repo's on Gitlab.

p.s. I have been using Alium Designer for a year (on windows) and it was much easier to end up with a broken,
non-functional schematic or pcb file.

[thinkfat]

lets be clear i want kicad to work and be replacement for all people that have no money to buy altium or similar. but with behavior surounding it i am afraid it will never happen.

If your posts to the KiCad forum were all like the post above, I need not ask why you were threatened with a lifelong ban.

how is sincere feedback on bug report forums bad? should not they expect only negative feedback i n order to make it better, well i give only negative feedback on bug report forums (pun intended).

Let's say your attitude towards the people who dedicate their time to improving it is a little strange. "It doesn't work for me so I'm either stupid (implied NO!) or it's badly designed!".

[tbavcevic]

but i am dedicating my time each major release with trying to get it work and fail each time, hence implied that i am either stupid or something aint right, maybe both.
can you answer how to get it working on ubuntu 20.10? or am i not allowed to say it does not work?

[gsocker]

but i am dedicating my time each major release with trying to get it work and fail each time, hence implied that i am either stupid or something aint right, maybe both.
can you answer how to get it working on ubuntu 20.10? or am i not allowed to say it does not work?

Probably, but we need more information from you. "Libraries don't work" doesn't provide much to go on.
Is the "Add Symbol" dialog empty (no libraries) when you try to insert components in a new project? Or is something else wrong?

[gsocker]

V5 has schematic/pcb autosave at a configurable time interval. I've used this before when KiCad or the computer crashes (or lost power)

In addition, V6 has configurable whole-project backup when saving. Optionally, it can also do so when the autosave takes place.
They are  stored in a 'backups' directory in the project directory as zipped files.

[tbavcevic]

both manualy added libraries downloaded from git and installed trough apt dont show up in schematic editor when pressing button to add part, i cant explain it any other way.
also there is no kicad folder in root/.config.

[Monkeh]

both manualy added libraries downloaded from git and installed trough apt dont show up in schematic editor when pressing button to add part, i cant explain it any other way.
also there is no kicad folder in root/.config.

Then you may need to configure the paths properly. Bear in mind they're not responsible for the mess your distro may make.

Also, why on earth are you running as root?

[gsocker]

both manualy added libraries downloaded from git and installed trough apt dont show up in schematic editor when pressing button to add part, i cant explain it any other way.
also there is no kicad folder in root/.config.

Then you may need to configure the paths properly. Bear in mind they're not responsible for the mess your distro may make.

Also, why on earth are you running as root?

Just tried spinning up an Ubuntu 20.10 VM and installing KiCad using the distribution packages. It's on 5.1.6 when the latest is 5.1.9, but otherwise started up file and did the normal auto-population of the symbol libraries when starting a project for the first time.

Running it as root using sudo also works fine (though I'd never actually use it like this).

[Bassman59]

Here's how I deal(t) with it, both Eagle and KiCad.
I disabled the auto-save functionality.
I press ctl-s after making a (small) modification.
Every hour or so, depending on the kind of modification, I enter "git commit -a" and "git push" in the terminal.
I never forget it because I trained myself to do this and it became a habit and now I do it without thinking.
I never lost a (part of a) project.

You can create and setup free, private git repo's on Gitlab.

p.s. I have been using Alium Designer for a year (on windows) and it was much easier to end up with a broken,
non-functional schematic or pcb file.

Really, how hard is it to type Ctrl-S (or Cmd-S on a Mac) after making a change, even a minor one? I’ve been in the habit of “save often” since the days of DOS.

There’s also a good reason for not using auto save: say you open the file to look at it for reference as you’re debugging. When you select things or change visibility, some programs “touch” the open file. You don’t want autosave to update a released file. And maybe you don’t want TortoiseSVN (or whatever) claiming that the file was modified and you should commit it.

[PlainName]

Really, how hard is it to type Ctrl-S (or Cmd-S on a Mac) after making a change, even a minor one? I’ve been in the habit of “save often” since the days of DOS.

Not difficult but a pain in the ass. I once found myself doing a save even after not changing anything merely because I couldn't remember the previous time I'd done it (a few mins before, likely). But this is all victim-blaming. You are saying people are at fault for not hitting save, rather than accepting that the software is at fault for obliging the user to allow for its brokenness.

Personally, I don't use auto-save ever (for the good reason you mention). I save after doing some reasonable amount of work that I am happy with - it is protecting against me screwing that up with my next mods. But sometimes I am 'in the flow' and make considerable changes over quite some time without a save. That's just how people do, or fail to do, things when they are focused.

If I use some software that requires me to make manual saves all the time because it is likely to crash (or otherwise lose data), I won't be using it anymore. Simple as that - it is shit software. The only exception is if something external to it (like the OS or another app) is causing the loss. The only reason y'all accept fault if you don't do regular saves is because you've allowed that mindset to take root.

[Karel]

You are saying people are at fault for not hitting save, rather than accepting that the software is at fault for obliging the user to allow for its brokenness.

I dare to say that an electronics engineer, who believes it's possible to write bugfree software, is naive.

No matter how hard you try, no matter how much money you throw at it, no matter how good programmers are,
it's simply not possible. Period.
Ofcourse, that doesn't mean that you are not allowed to criticize software. My point is, don't take any risks, save often and use git.

[janoc]

Running it as root using sudo also works fine (though I'd never actually use it like this).

Running as root using sudo and as an "regular" root (either by directly logging in or using su) is not quite the same - sudo will still keep all your environment, so e.g. $HOME is set to your regular home directory, only your effective UID changes. If you run as "regular" root, then your $HOME is typically /root, etc.

But yes, I do wonder why the heck is he trying to run this as root? I guess he likes playing with fire. Given this I come to think that his messed up config likely isn't a KiCAD issue ...

[PlainName]

Ofcourse, that doesn't mean that you are not allowed to criticize software.

That is the issue in this thread - the chap is being ostracised for perfectly valid criticism.

[tbavcevic]

...
But yes, I do wonder why the heck is he trying to run this as root? I guess he likes playing with fire. Given this I come to think that his messed up config likely isn't a KiCAD issue ...

ok lets start from begining,

first try
1. i go to kicad and follow instructions for ubuntu and add PPA
2. install kicad from software manager

no libraries

try x2, remove everything

1. reload ppa from terminal
2. use terminal to fetch all needed packages

in short this:
_{sudo add-apt-repository --yes ppa:kicad/kicad-5.1-releases
sudo apt update
sudo apt install --install-recommends kicad}

no libraries

searching around net, some people say go to root/ folder and remove kicad folder for help with this issue, there is no kicad folder there.

try x3

reinstall everything trough terminal commands like on x2, manualy download libraries from git and add them to my libraries folder, point to location of libraries in kicad settings(path config).

no libraries


as someone else said here do not put words in my mouth, and dont try to put this issue on me as there is a lot of users complaining about same issues over the nets and over time, even quick google search confirms that, there is one thread on this forum also even from 2017.


sorry if my engrish is not up to standard as it is not my native tongue and could lead to some misunderstandings.

only thing i have not tried yet is to manualy replace automatic downloaded libraries with ones from github in their original folders, i have not tried it yet as i use my secondary drive for projects and libraries as automatic backup is made on external drive, and in case i must use folders on os drive i would have to manualy copy that each time which is not realy fun but if it works who cares.

[thinkfat]

I don't quite remember how my first KiCAD install went, but I think you need to just add the libraries manually. I'm not sure if the Ubuntu install will do that automatically or not.
Adding the symbol libraries can be done in bulk, all of them at once. Same for the footprint libraries.

[woofy]

I'm running kicad 5.1.9 on ubuntu 20.04.1 (actually Mint 20.1). I don't recall having any problems with libraries, so I just downloaded a random one from github (an ESP32 module) and extracted the .zip.
In kicad Eeshema, I went to Preferences - Manage Symbol Libraries and clicked plus (+) to add a new library. Entered the Nickname, Path and Description and pressed ok.
Now when I try to add a symbol, I find the ESP32 library and can select the module.

So far (only a handful of projects) I've not seen a problem with libraries.

[tbavcevic]

I don't quite remember how my first KiCAD install went, but I think you need to just add the libraries manually. I'm not sure if the Ubuntu install will do that automatically or not.
Adding the symbol libraries can be done in bulk, all of them at once. Same for the footprint libraries.

this is help while i have been adding libs manualy you made me realise i was adding folder path like in eagle while kicad expects each library to be added separate, now it works and it even uses libraries on secondary drive without issue, now i can finaly test damned thing. thank you.

so it was dumb user error 

[julian1]

In debian default (kicad version is a bit behind), footprints, libraries, 3d models are separate packages, and can be installed individually. Not saying to use debian, but maybe another datapoint that helps..

$ aptitude search kicad | grep -v doc
i  kicad - Electronic schematic and PCB design software
p  kicad-common - Old common files used by kicad - Transitional Package
i A kicad-demos - Demo projects for kicad
i A kicad-footprints - Footprint symbols for KiCad's Pcbnew
i A kicad-libraries - Virtual package providing common used libraries by kicad
p  kicad-packages3d - 3D models for 3D viewer in KiCad's Pcbnew and Footprint Editor
i A kicad-symbols - Schematic symbols for KiCad's Eeschema
i A kicad-templates - Project templates for KiCad

[gsocker]

As a side note, you said you were looking for a config file under /root, so we all assumed you were running it as root. However, if you're running it as a normal user (the one you log in as), then the config file would be under your home directory, not /root.

[tbavcevic]

i just followed some online troubleshooting, and since i never used kicad on ubuntu before i do not know where config files would be.

i am dumb for this kicad i can even find out how to place component in schematics from parts browser, something changed ther from what i remember, this exercise is already taking too much time and effort for stuff that should be simple and normal.

[julian1]

I found this video really good, for a complete end-to-end walkthrough for a non-trivial board design.



The trick to know with Kicad is that when people say it is 'good', they mean that it is powerful and modular, works great with source-control/git, and has tons of community assets.

But (at least for me) I had to learn it, rather than just intuitively play around to get the hang of it.

[Monkeh]

I've never found any CAD tool to be intuitive. It's all complex software to achieve complex goals and requires a time investment to use effectively. Those unwilling to make a time investment will inevitably fail and complain loudly about how hard it all is.

[Karel]

https://www.udemy.com/course/kicad-like-a-pro-2e/

(No, I'm not affiliated to them.)

[gnuarm]

I've never found any CAD tool to be intuitive. It's all complex software to achieve complex goals and requires a time investment to use effectively. Those unwilling to make a time investment will inevitably fail and complain loudly about how hard it all is.

There was never an affordable electric vehicle... until there was. 

Those who are unwilling to ask for better products will not see them improved unless others do the asking for them.

[Monkeh]

I've never found any CAD tool to be intuitive. It's all complex software to achieve complex goals and requires a time investment to use effectively. Those unwilling to make a time investment will inevitably fail and complain loudly about how hard it all is.

There was never an affordable electric vehicle... until there was.

Well, there still isn't, so..

You seem to confuse my point with "the software need not improve" - that's on you. Using tools is a skill, a skill one must learn and practice. This applies whether it's a CAD tool or a bloody hammer.

[CatalinaWOW]

A lot of bashing here in this thread that doesn't make sense to me. 

Any complex tool requires a lot of learning.  But I find that some user interfaces/workflows click with me.  Others don't.  Unfortunately KiCAD is one that doesn't.  In an ideal world there would be versions that did things in ways that please everybody, but that isn't going to happen for a variety of good reasons.  The net result is that it is much harder for me than for some others to come up the curve, and I am sure that applies to others.  The roadblock is often something trivial once you figure it out, but it can be quite hard to find the appropriate documentation to point you in the right direction.  Several posts in this thread have demonstrated that process.  The treatment of libraries is an example. 

I do find it interesting how many people are ready to jump on a bash Windows or Office when they change their interface in ways that aren't obvious, while the response in KiCAD's case is often to leap to their defense.  I'm not a fan of MS changing the interface all the time, but given the huge user base it usually only takes a couple of minutes on Google to find the solution to your frustration.  The much smaller user base of specialist software like KiCAD means that those resources are far more sparse.  A little venting after hours or days fighting what seems like an impossible problem is not unexpected, and probably should be greeted with some tolerance.

[JohnnyMalaria]

But I find that some user interfaces/workflows click with me.  Others don't.  Unfortunately KiCAD is one that doesn't.

So true. I spent a couple of weeks trying to get my head around Eagle/Fusion 360. In the end, I decided to give KiCAD a try. It took me less than a day to learn and successfully created schematics, PCBs and new 3D models.

[gnuarm]

I've never found any CAD tool to be intuitive. It's all complex software to achieve complex goals and requires a time investment to use effectively. Those unwilling to make a time investment will inevitably fail and complain loudly about how hard it all is.

There was never an affordable electric vehicle... until there was.

Well, there still isn't, so..

You seem to confuse my point with "the software need not improve" - that's on you. Using tools is a skill, a skill one must learn and practice. This applies whether it's a CAD tool or a bloody hammer.

Not going to go on an on about this with you, but the fact that a tool has to be learned is not justification for a user interface that needs to be "improved".  It is that simple in spite of your protestations. 

Schematic capture and layout are inherently simple processes that were done for a very long time with pencil, paper and tape.  That's how simple it is.  We use computers because it makes the process less eraser intensive.  In the meant time the UI has become enormously complex because the programming under the hood is complex.  The goal of any UI is to hide that complexity and present a simple interface to the user. 

A perfect example is the lack of multi-page schematics in KiCAD.  They "emulate" pages by using a hierarchy where each drawing in the hierarchy is a page.  But you still can't move between pages, you can't go to the next page or the previous page or even know what page you are on easily.  The present handling of pages is clumsy because we have to deal with the complexity of the hierarchical structure KiCAD uses.  No doubt this will be fixed in a future release.  In the mean time people will complain about it just as they complain about other aspects of the tool that are needlessly messy or complex. 

The fact that you are satisfied with the tool is not the issue, just as your claim there are no affordable electric cars is wrong.  Currently there are nearly 2 million EV owners in the US who say you are wrong.

[thinkfat]

I felt the same when I tried KiCAD V4. It just didn't click. It was just quirky. I watched Daves video about KiCAD V4 when pcbnew didn't have an interactive push-and-shove router and footprints were called "modules". Decided it was not for me. Started to look at it again only after KiCAD V5 was released, which was vastly improved over V4.

There's an important BUT, though: back then when KiCAD V4 was new, I didn't have enough incentive to really dig into it. The whole "schematic and pcb layout" thing I mean. That only changed when services like OSHpark, SeeedStudio, DirtyPCBs, JLCPCB, PCBWay, Aisler etc came along. Before that time, it was nigh impossible for a beginner or electronics hobbyist to collect experience on a budget (save etching@home, which is largely wife-incompatible). Nowadays, I can get 4 layer PCBs in decent quality with impedance control at almost negligible cost.

[thinkfat]

KiCAD's handling of hierarchical sheets feels similar to how Altium works. Also, meanwhile you can copy components between sheets, even without having their annotation reset.

Main advantage of using an integrated CAD tool for schematics and PCB is not in saving on pencil eraser but in keeping track of nets to be routed.

[Monkeh]

Not going to go on an on about this with you, but the fact that a tool has to be learned is not justification for a user interface that needs to be "improved".  It is that simple in spite of your protestations.

I have genuinely no idea what point you're trying to make. No tool is perfect, from the outset or after years of development, and all tools require a time investment to operate effectively.

just as your claim there are no affordable electric cars is wrong.  Currently there are nearly 2 million EV owners in the US who say you are wrong.

And they've all got bigger budgets than me or different requirements to me. Therefore your claim that there are affordable electric cars is wrong.. in my context. What a wonderful concept.

[phil from seattle]

Affordable is a relative term. Though, in the case of EVs, it won't be long before it is the only kind sold. Not sure why that got dragged into the argument, though.

I have yet to find a CAD program that is intuitive. Perhaps it is a lack of focus on UI by the developers.  Or, perhaps it just the nature of the task at hand. Eagle was a fight to learn. F360 was too (and no they aren't even close despite being pushed as a package). AutoCAD at every version was obscure and strange. Sketchup, too. FreeCAD, same. Blender, ditto. and yes, Kicad.

To be fair, some are easier to learn and having learned one makes the next a lot easier. I actually found Kicad 5 relatively easy to learn. The trick there is memorizing all the keyboard commands. And stuff is still spread out (like where is minimum trace width again? Not in board setup. Oh yeah, in DRC.)  Also, the developers managed to make libraries really complicated.  Once I mastered them, I appreciated the power but that pool does not have a shallow end.  I think 6 is cleaning up a lot of issues so it is getting better.

[gnuarm]

I've used a layout tool that is great in terms of being intuitive.  There are a very few things that need to be learned and the rest is pretty obvious. 

But this conversation is getting old.  People have their opinions and seem to not be discussing so much as asserting their opinions are right.  So everyone enjoy what you enjoy.  Sorry to those who can't afford a good car. 

[Bassman59]

I do find it interesting how many people are ready to jump on a bash Windows or Office when they change their interface in ways that aren't obvious, while the response in KiCAD's case is often to leap to their defense.  I'm not a fan of MS changing the interface all the time, but given the huge user base it usually only takes a couple of minutes on Google to find the solution to your frustration.  The much smaller user base of specialist software like KiCAD means that those resources are far more sparse.  A little venting after hours or days fighting what seems like an impossible problem is not unexpected, and probably should be greeted with some tolerance.

See, the problems for both Kicad and Microsoft are that:

(i) some people don't like the interfaces and some people do, but the users get on with it and do the task at hand. 

(ii) if the user interface is changed to appease the people who didn't like the old interface, the other users complain, even if the new interface is better.

(iii) users who a few years ago tried the program and found the interface lacking in whatever manner still use that obsolete experience to complain about the interface -- even if the interface has changed to what might be more their liking.

The problem with CAD software, whether for mechanical design or for electronics design, is that there really is no "best" way to do it. If there was, every program would follow the same paradigm.

gnuarm's assertion that "schematic capture and layout are inherently simple processes that were done for a very long time with pencil, paper and tape.  That's how simple it is." is reductive. "Inherently" simple, perhaps, but the devil is in the details. Paper-based schematics and rubylith-based layouts were used when there was no other option, but for obvious reasons those processes have been abandoned. Can a paper-based schematic output a BOM for purchasing? Can you absolutely guarantee correspondence between all nets on the paper schematic and all traces on the rubylith layout? Can you manage the correspondence between a resistor symbol on your paper schematic and the footprint on the PCB and the part you order? The point here is that the tools have grown to manage that complexity, but of course that means that the tools themselves have gotten more complex.

[MadScientist]

Having not had any exposure to Kicad , ( being a long time DipTrace ) user. I sat down with my first Kicad V5 design and produced an acceptable SMD based design first go , this include designing new symbols , footprints and a 3D image or two. Of course I know about PCBs etc but I had no issues with Kicad and there’s plenty of support on the net.

Now that’s not bad for a piece of software , I would not say its " complicated " , certainly nothing like 3D or even Mech CAD etc

Are there things that could improve , of course.

But the package isn’t complex to use in the main. Fusions learning curve was far greater

By the way.  I learned with paper pcb layout. I never want to go back to that

[bdunham7]

just as your claim there are no affordable electric cars is wrong.  Currently there are nearly 2 million EV owners in the US who say you are wrong.

Logical fallacy alert!  You have no way of knowing whether those 2 million people agree or disagree with you just because they happen to own an EV.  And as an EV owner, if you will allow me to read practicality into the concept of affordability,  I would contend that for a very large majority of Americans (can't say elsewhere) an electric vehicle is not 'affordable', in the sense that they can't reasonably afford the extra expense of owning and operating an EV over a conventional gas car or hybrid.  Especially the hybrid.

b/t/w, glad to read your honest renditions of your experience with Tesla.  I know of people who have had repeated issues that more than qualify them for lemon-law protection but they won't even admit publicly that their cars have broken, let alone return them.

[gnuarm]

just as your claim there are no affordable electric cars is wrong.  Currently there are nearly 2 million EV owners in the US who say you are wrong.

Logical fallacy alert!  You have no way of knowing whether those 2 million people agree or disagree with you just because they happen to own an EV.  And as an EV owner, if you will allow me to read practicality into the concept of affordability,  I would contend that for a very large majority of Americans (can't say elsewhere) an electric vehicle is not 'affordable', in the sense that they can't reasonably afford the extra expense of owning and operating an EV over a conventional gas car or hybrid.  Especially the hybrid.

You ain't out here for the huntin', are yuh? 

I'll ignore the issue of 2 million car owners thinking their cars are not affordable.  The issue of the "extra" expense of an EV is a fallacy.  EVs have lower operating costs including lower repair costs, lower maintenance costs and lower fuel costs making the total cost of ownership lower than an equivalent ICE.  That's not me talking, that is many sources including government agencies that have them and track the costs as they wear them out including police.

b/t/w, glad to read your honest renditions of your experience with Tesla.  I know of people who have had repeated issues that more than qualify them for lemon-law protection but they won't even admit publicly that their cars have broken, let alone return them.

I only wish I could get them to fix the many problems or give all of my money back.  My car was no where near "affordable", lol.  It was my toy.  But it's not nearly the joy toy people make them out to be.  I expect by 2025 when nearly every car maker will have many models available Tesla won't be looking so good.  I read recently (maybe here) they are at the bottom of the list, even below Cadillac, for reliability.  That is quite the statement.

[Monkeh]

I'll ignore the issue of 2 million car owners thinking their cars are not affordable.  The issue of the "extra" expense of an EV is a fallacy.  EVs have lower operating costs including lower repair costs, lower maintenance costs and lower fuel costs making the total cost of ownership lower than an equivalent ICE.  That's not me talking, that is many sources including government agencies that have them and track the costs as they wear them out including police.

An EV to meet my needs would cost three to four times what my car cost, and would not in my current (excluding pandemic-induced changes) usage pattern recoup this in the next 10 years. This is not 'affordable', this is 'waste of money'. My needs are not your needs, my means are not your means. Your truth is not the only truth.

[nigelwright7557]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

I found a bug today in kicad in regards to keeping schematic layout when converting to a pcb.
At least on my version it just puts components along top of pcb to place.
I prefer components already placed where they were on schematic.
Associated components then stay next to each other which makes laying out better on medium to large sized pcb's.
After a lot of to-ing and fro-ing they eventually accepted it was a bug.
Bug ticket number 5480 !!!!!!

A few hours later they came back with a statement saying that due to hierarchical sheets they couldn't find a way around it.

[Monkeh]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

I found a bug today in kicad in regards to keeping schematic layout when converting to a pcb.
At least on my version it just puts components along top of pcb to place.
I prefer components already placed where they were on schematic.
Associated components then stay next to each other which makes laying out better on medium to large sized pcb's.
After a lot of to-ing and fro-ing they eventually accepted it was a bug.
Bug ticket number 5480 !!!!!!

A few hours later they came back with a statement saying that due to hierarchical sheets they couldn't find a way around it.

That they did no such thing is a matter of public record.. in issue 7480, for those wondering what a five month old bug about multi-line comments has to do with component autoplacement.

[nigelwright7557]

Well, there still isn't, so..

You seem to confuse my point with "the software need not improve" - that's on you. Using tools is a skill, a skill one must learn and practice. This applies whether it's a CAD tool or a bloody hammer.

I sell PCB design software (well a little bit) and software can quickly get out of control if you arent careful and become a monster.
Seems to be a happy medium between too simple with not enough functions to millions of icons with too many functions that you cant see the wood for the tree's.

I suffer from 2 bad types of software buyers.
1/ Those who spend £5 on mine but are expecting Eagle ($300) functionality for the same price.
2/ Old fart syndrome where they cant get their head around it. This is despite a demo video on how to do it and a manual.

I get many more complaints now my software has much more functionality than I ever did when it was very simple.

Just no longer worth the hassle so took up hardware sales instead.

[nigelwright7557]

That they did no such thing is a matter of public record.. in issue 7480, for those wondering what a five month old bug about multi-line comments has to do with component autoplacement.

Sorry quoted wrong message !

[gnuarm]

I do find it interesting how many people are ready to jump on a bash Windows or Office when they change their interface in ways that aren't obvious, while the response in KiCAD's case is often to leap to their defense.  I'm not a fan of MS changing the interface all the time, but given the huge user base it usually only takes a couple of minutes on Google to find the solution to your frustration.  The much smaller user base of specialist software like KiCAD means that those resources are far more sparse.  A little venting after hours or days fighting what seems like an impossible problem is not unexpected, and probably should be greeted with some tolerance.

See, the problems for both Kicad and Microsoft are that:

(i) some people don't like the interfaces and some people do, but the users get on with it and do the task at hand. 

(ii) if the user interface is changed to appease the people who didn't like the old interface, the other users complain, even if the new interface is better.

(iii) users who a few years ago tried the program and found the interface lacking in whatever manner still use that obsolete experience to complain about the interface -- even if the interface has changed to what might be more their liking.

The problem with CAD software, whether for mechanical design or for electronics design, is that there really is no "best" way to do it. If there was, every program would follow the same paradigm.

gnuarm's assertion that "schematic capture and layout are inherently simple processes that were done for a very long time with pencil, paper and tape.  That's how simple it is." is reductive. "Inherently" simple, perhaps, but the devil is in the details. Paper-based schematics and rubylith-based layouts were used when there was no other option, but for obvious reasons those processes have been abandoned. Can a paper-based schematic output a BOM for purchasing? Can you absolutely guarantee correspondence between all nets on the paper schematic and all traces on the rubylith layout? Can you manage the correspondence between a resistor symbol on your paper schematic and the footprint on the PCB and the part you order? The point here is that the tools have grown to manage that complexity, but of course that means that the tools themselves have gotten more complex.

You seem to think the process of design is complex because the tool can provide a BoM or track connectivity.  None of that is complex.  What is complex is constructing an internal data base that accurately reflects every aspect of the design while allowing users to make changes in ways that make intuitive sense.  It is the INTERNAL complexity that needs to be hidden from the user, because the mental image of what the user wants to construct is not complex.  It is the tool that is complex and the tool must shield the user from all of that as a good tool does.  A poor tool demands that the user see behind the curtain and work behind the curtain where he needs to be the Wizard of Oz rather than a Dorothy user.

It's no different from the automobile.  100 years ago cars were complicated, difficult to use machines.  You had to know how to set the choke, how to retard the spark, how to manage the gears, the throttle, the steering, the brakes because none of that was simple back then.  Now cars are even more complex, but to the user they are trivial.  Any 16 year old can operate one.  So much in a car is there to prevent the user from having to even know the car is shifting gears or applying the choke or controlling ignition timing.  You just point the steering wheel and step on the gas or the brake.... or use the accelerator to do both! 

That's the kind of simplicity that should be the goal in designing a PCB, not making excuses for how inherently complex a schematic capture and layout tool *has* to be.  pffft!   

[gnuarm]

I'll ignore the issue of 2 million car owners thinking their cars are not affordable.  The issue of the "extra" expense of an EV is a fallacy.  EVs have lower operating costs including lower repair costs, lower maintenance costs and lower fuel costs making the total cost of ownership lower than an equivalent ICE.  That's not me talking, that is many sources including government agencies that have them and track the costs as they wear them out including police.

An EV to meet my needs would cost three to four times what my car cost, and would not in my current (excluding pandemic-induced changes) usage pattern recoup this in the next 10 years. This is not 'affordable', this is 'waste of money'. My needs are not your needs, my means are not your means. Your truth is not the only truth.

I don't want to keep discussing EVs as it is off topic in this thread.  If someone starts another thread somewhere I'll jump in.

[emece67]

.

[gnuarm]

Wouldn't be possible to have a single thread in this forum where EVs are not mentioned?

What do you expect in the EVDA group?  ;-)

[Bassman59]

On a more serious note: it looks like a lot of interface changes have occurred, I'll have to see how they feel in-person.  I really hope eeschema's background colour hasn't changed to grey as the screenshots suggest.

I would assume it's configurable. The default in the older versions is blinding white, I always change it to black first thing after installing, I'd be shocked (and really annoyed) if they took away the ability to customize colors.

I found a bug today in kicad in regards to keeping schematic layout when converting to a pcb.
At least on my version it just puts components along top of pcb to place.
I prefer components already placed where they were on schematic.
Associated components then stay next to each other which makes laying out better on medium to large sized pcb's.
After a lot of to-ing and fro-ing they eventually accepted it was a bug.
Bug ticket number 5480 !!!!!!

A few hours later they came back with a statement saying that due to hierarchical sheets they couldn't find a way around it.

OK, I shouldn't bother, but ...

Your preference for a specific way of doing something doesn't mean it's a bug that when the software does it another way.

It's just that. A preference.

I actually don't care how the tool plops down the parts on first import into the layout. Generally, once they're there, I have the schematic open on one monitor and the layout on the other, and I start selecting the "big stuff" (connectors and other things whose locations must meet mechanical constraints, the big chips, etc) and place them. Then I get the support stuff from the schematic and place it.

In any case, you're still selecting and moving parts, so whether you do it as a group and fiddle, or whether you do it one by one and fiddle, there's not much time difference in the end.

[Karel]

In any case, you're still selecting and moving parts, so whether you do it as a group and fiddle, or whether you do it one by one and fiddle, there's not much time difference in the end.

Pcbnew's shortcut 't' ("Get and Move Footprint) comes to mind...

[nfmax]

In any case, you're still selecting and moving parts, so whether you do it as a group and fiddle, or whether you do it one by one and fiddle, there's not much time difference in the end.

Pcbnew's shortcut 't' ("Get and Move Footprint) comes to mind...

My way of working is to have all the part footprints outside the board area, out of sight & off screen, then use t to select and place each. I usually keep the ratsnest hidden for initial placement. I also take some care to work out roughly where everything will go before I start placing, but I mainly only route analogue PCBs. For boards like these, placement is much the harder problem, routing is pretty easy, and I don't want any automation messing the placement up for me!

[twospoons]

I still recall the bad old days (30 years ago) where importing a netlist resulted in a pile of footprints at (0,0). Having them spread on a grid is a vast improvement.
Schematic layout should be for readability, which is not necessarily the same as a functional board placement, especially for complex schematics. So initial placement based on schematic is not much of an advantage.

[gnuarm]

Schematic layout should be for readability, which is not necessarily the same as a functional board placement, especially for complex schematics. So initial placement based on schematic is not much of an advantage.

That depends on how the schematic is done.  I find many plop down circuits in little pieces with scarce structure above that.  I prefer to show the flow, both electrical and functional, in the schematic.  This often translates well to a good layout. 

[twospoons]

  I find many plop down circuits in little pieces with scarce structure above that.  I prefer to show the flow, both electrical and functional, in the schematic. 

Thats precisely what I mean by "readability" - nothing is worse that trying to follow a fragmented schematic connected by nothing but net tags.  I still find the board often does not look that much like the schematic, and that final placement isn't really assisted by component grouping when you can pick parts to move using designators, which is easier and faster than visual hunting.

[woofy]

In any case, you're still selecting and moving parts, so whether you do it as a group and fiddle, or whether you do it one by one and fiddle, there's not much time difference in the end.

Pcbnew's shortcut 't' ("Get and Move Footprint) comes to mind...

My way of working is to have all the part footprints outside the board area, out of sight & off screen, then use t to select and place each. I usually keep the ratsnest hidden for initial placement. I also take some care to work out roughly where everything will go before I start placing, but I mainly only route analogue PCBs. For boards like these, placement is much the harder problem, routing is pretty easy, and I don't want any automation messing the placement up for me!

I find the 't' command to be a bit tedious. I much prefer to have the schematic up on the second monitor. When I click a part on the schematic, it's highlighted on the layout and I can place it as required.
It's 2 clicks vs 5 clicks.

[Karel]

In any case, you're still selecting and moving parts, so whether you do it as a group and fiddle, or whether you do it one by one and fiddle, there's not much time difference in the end.

Pcbnew's shortcut 't' ("Get and Move Footprint) comes to mind...

My way of working is to have all the part footprints outside the board area, out of sight & off screen, then use t to select and place each. I usually keep the ratsnest hidden for initial placement. I also take some care to work out roughly where everything will go before I start placing, but I mainly only route analogue PCBs. For boards like these, placement is much the harder problem, routing is pretty easy, and I don't want any automation messing the placement up for me!

I find the 't' command to be a bit tedious. I much prefer to have the schematic up on the second monitor. When I click a part on the schematic, it's highlighted on the layout and I can place it as required.
It's 2 clicks vs 5 clicks.

But no mouseclicks! I prefer to type 't' and then type "c813" because typing something on a keyboard does not require to use the mouse (apart from placing the componnet obviously) and
no panning zooming. This avoids RSI.
The more you can do by typing instead of mouseclicks, the better! Eagle was superior in that. There was some kind of "console" in Eagle with a command history.
You typed "move c601"<enter> and the component was immediately attached to your mousecursor, ready for rotating/flipping/placing (like the 't" command in KiCad).
But, after that, you hit the "up-arrow" and the same command was there and you changed it quickly in c602 and hit enter again.
That workflow was superior to whatever I have seen in my life.

[thinkfat]

If you were able to memorize the commands, sure. But some things were only available through this CLI. Some functions became totally obscure. I remember erasing all traces was something along "net ripup all". Not accessible otherwise.

[Karel]

Imho, if pcb layout is part of your job & you have to layout a pcb often & you want to avoid RSI, memorizing the most used shortcuts is a must and well worth the energy invested.
Nice side effect, you will be faster.

[twospoons]

I think it was an early version of PADS where you could click the move button then there'd be a dialog box with a list of parts and a text box. You could put "R*" in the text box and sequentially place every part starting with "R" - no additional keys or mouse clicks needed. Or maybe that was Altium ? Not sure now.

Eagle's command line input is too verbose for me - I much prefer single key hotkeys. Though sequential hotkeys aren't too bad either (eg Altium).

[riyadh144]

It is pads, honestly taht is one feature that I actually miss from pads. Other than that I love my KiCAD and wouldn't want to go back to that peace of shit software.

[nfmax]

[…] especially now that 'schematicless' workflow will be supported […]

I recently ―less than a month ago― tried KiCAD for the first time (no nightly builds, but stable 5.1.9), coming from Eagle (and before of it, from OrCAD). Although I have found some oddities, it is really usable, and much more user friendly, and I decided to definitely switch to it. I'm not a heavy user of PCB design tools (I only design PCB as hobby, not work, and usually try to avoid it) so I'm not really aware of such schematicless workflow in KiCAD or in any other tool, but designing now some μstrip and GCPW RF circuits such approach seems quite desirable. Do you have more information about the way it is supported in KiCAD?

Thanks & regards.

Just last night some kind person posted a tutorial on schematicless design on the Kicad forum:

https://forum.kicad.info/t/board-layout-without-schematic/27992/4?u=maxwellhadley

[nigelwright7557]

It is pads, honestly taht is one feature that I actually miss from pads. Other than that I love my KiCAD and wouldn't want to go back to that peace of shit software.

When comparing software you need to compare like with like.
Pads is paid for software.
But so is kicad, it was funded to the tune of $20,000 last year (quoted from their website)
I would suggest kicad compares to software around the £300 mark.
You pay for kicad whether you buy it or not through taxes via grants.

The only true free PCBCAD software I know is GNU.

Even the so called free PCBCAD software Design Spark is paid for by RS Components.

[nigelwright7557]

I still recall the bad old days (30 years ago) where importing a netlist resulted in a pile of footprints at (0,0). Having them spread on a grid is a vast improvement.
Schematic layout should be for readability, which is not necessarily the same as a functional board placement, especially for complex schematics. So initial placement based on schematic is not much of an advantage.

kicad lines up components along top of pcb. Or at least my version does.
I prefer having pcb components in same layout as schematic so associated components are already next to each other.

[nigelwright7557]

Some bugs found right after I downloaded the latest nightly PPA release:

1. Footprint editor will not display grid correctly.
2. Schematic library conversion gets the library lost.

I didn't test further before removing it. I'll wait for its final release.

Your a brave man saying that about kicad.
kicad is the best thing since sliced bread and has NO bugs or issues.

However when you consider they had $20,000 funding last year they should be doing a better job........
Shock horror ! no kicads not free, you pay for it through taxes and grants.

[JohnnyMalaria]

I still recall the bad old days (30 years ago) where importing a netlist resulted in a pile of footprints at (0,0). Having them spread on a grid is a vast improvement.
Schematic layout should be for readability, which is not necessarily the same as a functional board placement, especially for complex schematics. So initial placement based on schematic is not much of an advantage.

kicad lines up components along top of pcb. Or at least my version does.

Either upgrade or, if using the latest, RTFM.

[Bassman59]

Some bugs found right after I downloaded the latest nightly PPA release:

1. Footprint editor will not display grid correctly.
2. Schematic library conversion gets the library lost.

I didn't test further before removing it. I'll wait for its final release.

Your a brave man saying that about kicad.
kicad is the best thing since sliced bread and has NO bugs or issues.

All due respect, but get off of this. He was running a nightly build of the development branch, so bugs are to be expected. The reason for the nightly builds is to get the latest code into the hands of anyone who wants to test the code.

We know you don't have anyone test your code. We know you don't do nightly releases of a development branch. We know you don't publish your source code, probably because it's embarrassingly shitty.

I notice that you ignored my post from last week about a bug I found that the developers fixed within two hours of my report.

Your fucking problem is that you think people say that "Kicad has no bugs or issues," when you are told repeatedly that yes there are bugs, yes there are issues, yes they get handled, and also your preference for how footprints are ordered upon import into the layout is not a bug.

However when you consider they had $20,000 funding last year they should be doing a better job........
Shock horror ! no kicads not free, you pay for it through taxes and grants.

Why do you keep harping on the fact that Kicad's developers get some money from CERN? Does it matter? Does it matter if Linux kernel developers get paid?

And honestly, if you took the stick out of your ass and actually did some work using Kicad 5.1.9, and then did another board using the 5.99 nightly, you'd know that there's a ton of improvements in all areas: in the user interface, in the libraries, in design rules, in routing, in board/design set-up, in all of it.

But you won't do that, because the parts aren't organized on initial import into the board the way you prefer.

[MadScientist]

Some bugs found right after I downloaded the latest nightly PPA release:

1. Footprint editor will not display grid correctly.
2. Schematic library conversion gets the library lost.

I didn't test further before removing it. I'll wait for its final release.

Your a brave man saying that about kicad.
kicad is the best thing since sliced bread and has NO bugs or issues.

However when you consider they had $20,000 funding last year they should be doing a better job........
Shock horror ! no kicads not free, you pay for it through taxes and grants.

I pay no extra money to use Kicad. That’s “free”  as far as I’m concerned

[phil from seattle]

I pay no extra money to use Kicad. That’s “free”  as far as I’m concerned

Please don't feed the troll.

[Miti]

I love Kicad and I appreciate and respect the effort that is put in it by the developers! I just want to ask if Kicad 6 will fix the ERC? It is pretty bad in Kicad 5.

Edit: Ok, I found out that my FPGA symbol have a lot of undefined pins, so in part the errors that I'm getting may be justified.
Edit2: My bad. I withdraw my statement.

[vad]

However when you consider they had $20,000 funding last year they should be doing a better job........
Shock horror ! no kicads not free, you pay for it through taxes and grants.

$20,000 funding in the last year. Wow. That must be a fortune for for-profit software company. Shock and horror for Altium. Software giants like Microsoft and Adobe are jealous.

... oh, wait a second. Altium’s revenue was US$189M in the same year. The “M” here stands for millions. That’s 4 orders of magnitude difference.

https://www.altium.com/resources/altium_announces_financial_results_for_the_full_year_to_30_june_2020.pdf

[voltsandjolts]

However when you consider they had $20,000 funding last year they should be doing a better job........
Shock horror ! no kicads not free, you pay for it through taxes and grants.

I never wanted Tony Blair to take the UK to war in Iraq and Afghanistan, but I paid my share regardless. Welcome to society.
If only that GBP30B had gone to KiCAD instead.

...and v. sorry for feeding the troll.

[PlainName]

If only that GBP30B had gone to KiCAD instead.

It would be spaffed up the wall now that we are out of the club and unable to benefit from it.

[wilfred]

I found this YT video by Robert Feranec where he interviews the Kicad project leader on the future course planned for Kicad. He seemed mostly interested in how future plans will negatively impact someone thinking of committing their project to Kicad. I did look a bit for previous posts linking this video but hopefully it interests someone here.

It is a long video so grab a coffee.

[mstevens]

Shock horror ! no kicads not free, you pay for it through taxes and grants.

I've watched you say this several times. By your argument nothing is free; after all you do pay taxes for the air you breath and the hair that grows on your body. Dude, get a life!

[phil from seattle]

Please don't feed the Trolls.

[nigelwright7557]

Please don't feed the Trolls.

If I could have got it to do what I want I would have used it but I cant.
I couldnt even complete a track it wouldnt put down last segment.

I think it needs a bit longer to mature before it is usable.

[Monkeh]

Who let Nigel out of his room again?

[LazyJack]

Well. Many of us succeeded in drawing the last segment in a track, or even so, the last segment of a huge number of tracks and that resulted in having a manufactured PCB.
So I'm not sure now where to look for the problem.

[madires]

The problem is that KiCad is Nigel's competitor.

[benst]

I think I need a bit longer to mature before I am usable.

There, fixed it.

[phil from seattle]

Chris Gammell did an interview with Kicad developer Jon Evans about KiCad 6 development a couple days ago.  It isn't on youtube but can be accessed here.  You may need to download it to watch. Interesting insight into the design philosophy.  It's a bit "chit-chatty" but still has a decent SNR.

[phil from seattle]

I think I need a bit longer to mature before I am usable.

There, fixed it.

Feedin' the Troll, again...   

[twospoons]

So I'm not sure now where to look for the problem.

I would start by looking between the keyboard and the chair.

[ebastler]

Things have gone quiet regarding the KiCad 6 release. I am aware of the team's "it' will be ready when it's ready" approach and appreciate that it is not a commercial product with a committed schedule. But weren't they hoping to release 6.0 sometime during the first half of the year?

A quick web search did not find any recent updates or discussions. Does anyone have further insights into the current status and plans?

[retiredfeline]

There are threads on the KiCad forum on this, e.g.: https://forum.kicad.info/t/version-6-release-plan/30013/62

Some people feel 5.99 is stable enough for daily use. Myself I prefer to wait, as you cannot go back to 5.1 after the files have been converted.

[Karel]

I must admit I'm a bit disappointed as well but I don't blame the developers.
I prefer they polish it as much as possible and release it when they feel confident.
I hate agile software development...

[thinkfat]

I must admit I'm a bit disappointed as well but I don't blame the developers.
I prefer they polish it as much as possible and release it when they feel confident.
I hate agile software development...

"It's ready when it's ready" is not agile. Agile means that you have a backlog of tasks that will get you to the release. Each sprint the backlog gets shorter and by the speed of the burn-down you can usually tell when it'll be ready.

But if the backlog doesn't get shorter or even grows, it's not agile, it's a death march.

[I wanted a rude username]

Some people feel 5.99 is stable enough for daily use.

Most of the KiCad users I know have switched to 5.99 nightlies for new boards. I've completed two projects using it ... really enjoying the mitred tracks, layer opacity control, and less recalcitrant routing. Will probably keep old projects in KiCad 5 format indefinitely though (or unless a major change is required).

[rf-messkopf]

Things have gone quiet regarding the KiCad 6 release. I am aware of the team's "it' will be ready when it's ready" approach and appreciate that it is not a commercial product with a committed schedule. But weren't they hoping to release 6.0 sometime during the first half of the year?

A quick web search did not find any recent updates or discussions. Does anyone have further insights into the current status and plans?

On GitLab the target date for the 6.0.0 release candidate 1 is shown as Nov. 30, 2021. Recently I submitted a couple of bug reports and they were fixed promptly in the nightlies.

Currently there are 49 open issues with the 6.0.0-rc1 milestone.

[bitwelder]

Fresh announcement from Wayne:

https://lists.launchpad.net/kicad-developers/msg45375.html

The lead development team has agreed on a version 6 release schedule as follows:

- String freeze November 1.
- Tag RC1 on November 15.
- All repos frozen on December 14.
- Tag all repos with 6.0.0 on December 15.
- 6.0.0 release announcement December 31 (maybe the day before Christmas
  just for fun).
- Branch V6 and open master for new feature development on January 1.

[thinkfat]

That is excellent news!

[ebastler]

Good news indeed. Although if one reads the remaining lines of that announcement, it becomes clear that we might not want to hold our breath for a Jan 1st release: 

Our goal is to stick to that as closely as possible [...]
Hopefully the version 6 release will go reasonably smoothly and we can get started on version 7 in 2022.

[james_s]

I'll be curious to have a look when it's ready. I'm still using 2013-07-07, I tried 5.1.0 a while back but experienced a lot of problems with the rendering and trying to convert existing designs just made a big mess so I rolled back.

[Bassman59]

I'll be curious to have a look when it's ready. I'm still using 2013-07-07, I tried 5.1.0 a while back but experienced a lot of problems with the rendering and trying to convert existing designs just made a big mess so I rolled back.

Jeez, they're up to 5.10 now -- and that's much improved over 5.1.0. And the old 2013? Yikes. Upgrade, man, upgrade!

[retiredfeline]

Jeez, they're up to 5.10 now -- and that's much improved over 5.1.0. And the old 2013? Yikes. Upgrade, man, upgrade!

Actually it's 5.1.10, i.e. the same 5.1 series as 5.1.0. And 5.1.11 has just been released.

[Bassman59]

Jeez, they're up to 5.10 now -- and that's much improved over 5.1.0. And the old 2013? Yikes. Upgrade, man, upgrade!

Actually it's 5.1.10, i.e. the same 5.1 series as 5.1.0. And 5.1.11 has just been released.

You are correct! Dunno why I screwed that up, especially since I could've just launched Kicad and looked at the "About" dialog.

The advice about "upgrade, man, upgrade!" remains!

[retiredfeline]

It seems 5.1.11 was short lived as they found a regression so didn't even appear for some platforms so 5.1.12 is imminent.

[Cerebus]

As I've said before, it's worth at least trying the current nightlies (5.99, the work in progress for 6).

Many of the things that were show stoppers for me in earlier versions (mostly really dumb UI issues) have gone away. To the point that whereas I was quite a detractor of KiCad pre 5.99 I'm quite happy to use it now and recommend it to people who don't have access to one of the big money packages. Yes, there are still things that could be better done, there are still some egregious UI issues (e.g. grossly inconsistent tabbing orders) but it's useable whereas before I didn't believe that it was. I've cranked out two 4 layer boards and two 2 layer boards with it without problems.

It still lacks features that some of the big packages have, but on the other hand it crashes less often than some of the big packages. The UI could still use a whole lot of work, but it at least now has some consistency rather than being the ragbag it was before. If you're using a older version I would strongly recommend giving it a try.

[fourfathom]

I'm using v5.1.8 and just finished a couple of 4-layer boards.  These had been 2-layer but some circuit changes convinced me to try going 4-layer.  It was ridiculously easy, and once I figured out how to have a zone within a zone (see "priority") I ended up being quite happy with the result.  A much cleaner layout, done in one evening. (See attachment.  The SMA connector 3D models aren't exactly right, but the pads on the board do fit the connectors I am using.  I added the breakout grids surrounding the two big chips because I had room and figured I might want to use the board for other things.)

I downloaded the v5.99 version (or whatever it was a couple of months ago), and it looked good, but I'm afraid to transition my work in progress to it as it apparently uses a different library structure and you can't revert back.  Since I'm pretty happy with v5.1.8 I was planning to hold out for the official 6 release. 

[ebastler]

I am running 5.1.10. Had to set up a new computer recently and went with the then-current production version of KiCad.

My reason for being reluctant to try 5.99 and then update to 6.x hopefully very soon is that KiCad's update process has not worked well for me in the past. After I updated KiCad on my previous machine I found myself with a second set of libraries and a lot of confusion which ones were used where. I don't recall the details, but it took a long time to clean up.

Have you in general been able to update (or try 5.99 in parallel with your prior version?) without problems?

[retiredfeline]

Well I was lucky because I never ran 4, so I never suffered upgrade trauma. Although I probably could figure it out if I encountered a problem. I suspect 90% of the problems boil down to library paths.

The reason I'm holding off is simply because I only design hobby boards for myself not for a living so I can live without the new features a bit longer. Also I want any unfortunates who download my designs to be able to open them with a production version. I don't want to have to say you have to install 5.99 to modify my design.

If you are trying it for the first time, or haven't used any version since 4, I'd recommend making a fresh start with 5.99. Make sure you clear all vestiges of any 4 installs though. Many professional users have already given the thumbs up to 5.99 so it will be good. In any case, within a couple of months time it will be moot.

[Cerebus]

I downloaded the v5.99 version (or whatever it was a couple of months ago), and it looked good, but I'm afraid to transition my work in progress to it as it apparently uses a different library structure and you can't revert back.  Since I'm pretty happy with v5.1.8 I was planning to hold out for the official 6 release.

I went through a similar thinking process about the non-reversibility. After playing with 5.99 (All the nightlies are '5.99', you need to use a date or the git hash to distinguish between them) in a parallel setup I switched to using the nightlies as daily driver. For the last six months or so I've found almost all the nightlies as stable as production versions - only one have I had to grab the next day's version because of a problem. I don't update nightly, only every few weeks.

The glacial progress of many FOSS projects means that often the ostensible alpha/beta version is the one to use once it's basically stable rather than wait a year or two for the pukka 'release' version. KiCad is no different in that regard.

[Cerebus]

Have you in general been able to update (or try 5.99 in parallel with your prior version?) without problems?

There are instructions (somewhere, don't ask where, it's more than six months ago I used it) on setting up in parallel and they worked without a hitch for me. I soon however migrated off that setup because 5.99 is surprisingly trouble free. As always with these things, I find it best to regard the secondary set-up as "throw away" if need be, and the main original setup as sacrosanct. That way you don't mess up your workflow if the parallel setup goes horribly wrong. As I say, I probably migrated within a month of first trying 5.99.

[djsb]

Plugin and content manager now active in the latest nightly version. Only one so far is an INTERACTIVE HTML BOM plugin.

[fourfathom]

I just installed the latest 5.99 (after archiving all my existing jobs and libraries) and it is really nice.  Setting up the library path was easy  (I use my own custom library), and as a test I opened up an existing test project.  Schematic, layout, 3D render -- all look good.  I tried a little board editing, and while I haven't checked all the functions i's sure nice to see the cut/paste/move/drag command keyboard shortcuts normalized between the schematic and PCB tools.  I added a couple of layers to a 2-layer test board and it all went smoothly.  The PCB layer opacity features are nice.  Thanks, Cerebus, for nudging me to upgrade.

[edit] And it has hatched area fills!  I don't know if I need them, but it's a nice feature.

[Cerebus]

i's sure nice to see the cut/paste/move/drag command keyboard shortcuts normalized between the schematic and PCB tools

That is one of the things that was driving me crazy before it was fixed.

[Harjit]

I just installed the latest 5.99 (after archiving all my existing jobs and libraries) and it is really nice.  Setting up the library path was easy  (I use my own custom library), and as a test I opened up an existing test project.  Schematic, layout, 3D render -- all look good.  I tried a little board editing, and while I haven't checked all the functions i's sure nice to see the cut/paste/move/drag command keyboard shortcuts normalized between the schematic and PCB tools.  I added a couple of layers to a 2-layer test board and it all went smoothly.  The PCB layer opacity features are nice.  Thanks, Cerebus, for nudging me to upgrade.

[edit] And it has hatched area fills!  I don't know if I need them, but it's a nice feature.

Can you elaborate on the steps for setting up your own library path? I'm in exactly the same situation and have been putting it off because I didn't want to mess around.

Thanks in advance.

[Cerebus]

[edit] And it has hatched area fills!  I don't know if I need them, but it's a nice feature.

Not something I ever use, or at least haven't since PCB processes improved and it wasn't necessary to accommodate a board house's requirements. Then the other day I read a recommendation in an Analog data sheet (ADR1399) to use a hatched ground plane to reduce thermal mass when dealing with a device where thermal emfs may be an issue. Might find a use for them yet (I have an ADR1399 sample on order).

[fourfathom]

Can you elaborate on the steps for setting up your own library path? I'm in exactly the same situation and have been putting it off because I didn't want to mess around.

If you mean for V5.x, I have a directory(folder) where I keep all my design stuff, completely separate from KiCad installation directory.  In this I have folders for each rev of a project (schematic, PCB, and related files).  I also have a personal library folder there, and that contains three more folders: 3D, footprints, and symbols.  I started out by opening up the default KiCad libraries and saving what I was using from that in my personal libraries.  From this point on any design uses parts from my personal library, and when I need a new part I build it myself (I prefer my own schematic symbols to what I usually find through SnapEDA or UltraLibrarian.  I do grab the 3D file (if available) from those sources, or elsewhere, I still haven't learned how to do my own 3D models.

I set up the path in the schematic-capture using  "Preferences / Configure Paths" and "Preferences / Manage Symbol Libraries"
In PCB layout it's "Preferences / Manage Footprint Libraries"

There may be more to it than that, I just sort of tried stuff until it worked.  Somewhere along the line I removed the paths to the default libraries so I didn't have to scroll through those every time I wanted to place a component.

[Harjit]

@fourfathom - thank you for the info. I have exactly the same structure.

[Mark19960]

I figured I would toss my hat into the version 6 ring....

I like what I see. There are numerous improvements to functionality and "quality of life".
I can't see going backwards to 5.x from a nightly....
(and yeah, I know that can bite me)

Just liking the new experience.

[Smokey]

I'm a little confused about what's going on with this. 

Here are their GitLab Milestones:
https://gitlab.com/kicad/code/kicad/-/milestones?sort=due_date_desc&state=all

They show version 6.0.0-rc1 as 100% complete with no open issues, but there isn't either an RC2, or a full release milestone. 

Does this mean it's ready and they are just working on logistics of pushing out the official v6 release? 
Is there some other place where they are keeping track of what's left to do for the 6 release?

[retiredfeline]

They show version 6.0.0-rc1 as 100% complete with no open issues, but there isn't either an RC2, or a full release milestone. 

Does this mean it's ready and they are just working on logistics of pushing out the official v6 release? 
Is there some other place where they are keeping track of what's left to do for the 6 release?

Complete means that task (RC1) was done. Other tasks remain to be done before release.

[Smokey]

They show version 6.0.0-rc1 as 100% complete with no open issues, but there isn't either an RC2, or a full release milestone. 

Does this mean it's ready and they are just working on logistics of pushing out the official v6 release? 
Is there some other place where they are keeping track of what's left to do for the 6 release?

Complete means that task (RC1) was done. Other tasks remain to be done before release.

I get that.  I've worked with milestones before.

My confusion is coming from the fact that when you complete one milestone in a big long term project you typically move on to another one. 
If there was more work to do for v6, it would typically be included in another v6 related milestone.
I was expecting them to be working on something like "v6-RC2" or  "Official v6" or something like that.
The only obviously open milestone is "7.0".

I really don't care what they call their releases, or how they manage their project.  I'm much more interested in what bugs they still think need fixing in the nightlines for 6.0, and that's not obvious from their gitlab.

[retiredfeline]

Maybe they haven't decided on the remaining milestones yet. Not every project is super-organised, small ones manage the best they can. Just sit back and it will be a pleasant surprise when it's released.

[SiliconWizard]

I like what I see. There are numerous improvements to functionality and "quality of life".
I can't see going backwards to 5.x from a nightly....

What would be the most significant improvement in your opinion?

[nigelwright7557]

I like what I see. There are numerous improvements to functionality and "quality of life".
Just liking the new experience.

Bugs should get less as time goes on.
But having said that adding new code creates new bugs.
Test test test is the answer or just put it out to people to debug it for them.
My preferred method is to just be very careful with the code you write in the first place.

[e0ne199]

hello, i have a question... when can i download the latest KICAD (v6.0.0 rc1 or rc2)? does anyone here have the link? i don't really want to build that myself since i don't have the compiler 

[woofy]

This gets you RC2

sudo add-apt-repository --yes ppa:js-reynaud/kicad-dev-nightly
sudo apt update
sudo apt install kicad-nightly

Or equivalent for your distro. That is the legacy way that I used.

Latest:
sudo add-apt-repository --yes ppa:kicad/kicad-dev-nightly
sudo apt update
sudo apt install kicad-nightly

But I've not tested that.

[chickenHeadKnob]

Has anyone here achieved a successful  install of the reynaud nightlies on Ubuntu or Mint virtual machine lately, as in  the last few days? I tried on both LTS 20.04 based ubuntu minimum and Mint (mate edition). The result is the download appeared successful and I can get the top level running but the moment I try to launch eeschema, or PCB ect. it crashes silently.

I fear I am doing something really stupid and obviously wrong.

[Dagon]

Has anyone here achieved a successful  install of the reynaud nightlies on Ubuntu or Mint virtual machine lately, as in  the last few days?

I installed RC2 on mint 20.2 from kicad-dev-nightly that @woofy posted above.
No issues, everything work fine and I didn't need to do anything special. I will say it's a physical desktop rather then a VM.

Try running one of the programs directly from the cli and see if you get an error message e.g /usr/bin/eeschema-nightly

[retiredfeline]

The jsreynaud PPA says:

This legacy PPA should not be used anymore for new configuration. Please use official PPA instead:

ppa:kicad/kicad-dev-nightly

See https://www.kicad.org/download/ubuntu/ for details

[woofy]

With V6 due on Xmas eve , It might be worth waiting for the release at this stage.
https://www.kicad.org/img/donations/2021_flyer.pdf

[Quarlo Klobrigney]

My Mint 20.2 updated Kicad to 6.00 rc2 to my surprise a few days ago and works (my limited usage) just fine.

[thinkfat]

Looks like 6.0.0 has been tagged in the source repository, I'm compiling it as I write this post. The master branch has already been switched to version "6.99.0". Libraries and 3D models were also tagged, I've already downloaded and installed them.

I'm guessing all that's left is writing the announcement

[Cerebus]

Looks like 6.0.0 has been tagged in the source repository, I'm compiling it as I write this post. The master branch has already been switched to version "6.99.0". Libraries and 3D models were also tagged, I've already downloaded and installed them.

I'm guessing all that's left is writing the announcement

I'm not sure whether that's good or not. It all depends whether it's being done because there's been a lot of progress in a short time, or because people are rushing to reach an arbitrary "by Christmas" milestone. I used RC1 in ernest to do a couple of boards last week (neither still quite finished, I gave up on my arbitrary "before Christmas" milestone  ) and found a lot of bugs, including crashes, that had reappeared or freshly appeared. To my shame, I didn't get the time to report them. I grabbed RC2 1 to 2 days ago and haven't had time to give it a workout yet so it might have improved greatly; then again it may not have. I do hope that the 6.0 tagged version is good and not a victim of deadlineitis and that we don't find ourselves waiting for 6.0.1 with bated breath.

[Doctorandus_P]

KiCad V6 is not going to be here before crhistmas.
Expectancy currently is "in the first few weeks of 2022.
A few days ago the "kicad-nighly" repository for my Linux box changed from V5.99 to V6.9 and at the moment this means for me that I can not run V6.0.0-rc2 without hacking into the package management. I could focus on this as a (temporary) annoyance, but it's also a clear sign that a stable V6.0 is real close now.

Good things come to those who wait.

[rolycat]

You can install an unannounced and potentially still under construction V6 on Ubuntu/Mint Linux as follows:

Code: [Select]

sudo add-apt-repository --yes ppa:kicad/kicad-6.0-releases
sudo apt update
sudo apt install --install-recommends kicad


[nigelwright7557]

Using freshly produced code isnt always sensible.

I am an insider with Microsoft and have had Win 11 for many months now.
Loads of crashes and freezes to start with.
Its got better recently though.

Same for Visual Studio, the buggers took out winforms for c++, had to fight tooth and nail to get it put back in.
Then they took out publishing single/multiple files from solution explorer FFS !
We have been around the loop 3 times now with it being removed, put back in, removed again !

Then got into python websites. After 2 weeks of trying it wont publish the python project to AWS.
Have to zip up files instead and deploy them myself.

If you value your projects/sanity stick to older well tested software.

[PlainName]

Using freshly produced code isnt always sensible.

Seems like using freshly produced Microsoft code is never sensible. Which is not news, and is rather source-specific.

[ataradov]

That nightly (v6-rc2) PPA is pretty stable at this point. Most of the updates that are happening now there are to the libraries.

I'd say if you are waiting for V6 to start using KiCad, it is perfectly fine to start with the nightly at this point.

[rolycat]

As thinkfat pointed out, the nightlies are now from the new master branch, V6.99, and will increasingly diverge from a stable version 6.00.

The version installed from kicad-6.0-releases as I described earlier is reporting the following:

Version: 6.0.0-d3dd2cf0fa~116~ubuntu20.04.1, release build

Looks like KiCad 6 to me...

[ataradov]

The latest nightly from http://ppa.launchpad.net/kicad/kicad-dev-nightly/ubuntu is "6.0.0-rc2-unknown-160328abc7~144~ubuntu21.10.1, release build" built on Dec 20 2021 21:03:28.

[rolycat]

The latest nightly from http://ppa.launchpad.net/kicad/kicad-dev-nightly/ubuntu is "6.0.0-rc2-unknown-160328abc7~144~ubuntu21.10.1, release build" built on Dec 20 2021 21:03:28.

No, it isn't. I've just installed the latest nightly using the following commands:

Code: [Select]

sudo add-apt-repository --yes ppa:kicad/kicad-dev-nightly
sudo apt update
sudo apt install kicad-nightly
The version reported is:

Version: 6.99.0-a9ca49a600~144~ubuntu20.04.1, release build
Date: Dec 23 2021 15:10:40

[ataradov]

This is strange. I had it installed for a while, but now "apt-get update" / "apt-get upgrade" gives me the version I copied earlier. And the most recent couple updates only contained libraries.

Not sure what is going on.

But also, I'll just install the release one anyway. Nightly served its purpose for me.

[nuclearcat]

The latest nightly from http://ppa.launchpad.net/kicad/kicad-dev-nightly/ubuntu is "6.0.0-rc2-unknown-160328abc7~144~ubuntu21.10.1, release build" built on Dec 20 2021 21:03:28.

No, it isn't. I've just installed the latest nightly using the following commands:

Code: [Select]

sudo add-apt-repository --yes ppa:kicad/kicad-dev-nightly
sudo apt update
sudo apt install kicad-nightly
The version reported is:

Version: 6.99.0-a9ca49a600~144~ubuntu20.04.1, release build
Date: Dec 23 2021 15:10:40

I guess better to stick to "sudo add-apt-repository ppa:kicad/kicad-6.0-releases" now
Several people reported "6.99 problem" on nightly.

[djsb]

Version 6.00 is being released next week, and so the nightly version has been changed to version 6.99 (version 7). This is the case with both the windows and linux flavours (I download both). There is a freeze on the V6.0 version at the moment, so any bug fixes happen in version 6.99. There will be a 6.01 release later.

[Quarlo Klobrigney]

Mine comes up as 6.99.0-a9ca49a600~1344~ubuntu20.04.1.release build

[JohanH]

I've been using 5.99/6.0 from Fedora nightly copr repository for several months. I noticed that it changed to 6.99 three days ago and had to downgrade to latest 6.0, because I don't want to mess up my system again (had some issues with library locations etc. moving from 5.0 to 6.0 earlier, but not sure if it could happen again). And doing like this I guess shouldn't be hard to move to the official 6.0 version later when it becomes available. I will probably not move to next version before late in its development stage.

So on Fedora the last 6.0 is kicad-nightly-6.0.0-1.20211220git160328a.fc35.x86_64. After that it changed to 6.99.

I've had some occasional crashes when manually adjusting/moving tracks (but it was some weeks ago). Other than that, love the new 6.0 version.

[MadScientist]

I’ve been using V6 RC2 for a few days.  Very nice. Found a few funnies ( a cap only showing one rat nest line but when you routed it appeared )

[delfinom]

Fair warning, anyone on the "nightly" branch (any kicad that displays 6.99 now) is already getting 7.0 changes, the file format has already changed and more work is coming down the pipe.


6.0 releases are available for windows, mac and flatpak. The ubuntu ppa think has packages too. Every other linux distro isn't maintained by kicad so it'll land when the distro maintainers get to it.

[djsb]

Just downloaded V6.00
Now I have V6 and V6.99 nightly so I can compare new features. I'll only be using V6 for any manufactured boards though.

[PlainName]

already getting 7.0 changes, the file format has already changed

Why do the file formats keep changing? One of the nice things about Altium is the backwards and forwards compatibility - I've used the '09 viewer to view '21 files and it works. Complains about new things it doesn't know about and outputs a list of them, but shows the essentials just fine. And, of course, the latest version opens the most ancient without quibbling.

I could understand it a format change happened once a decade or so, but every version? Suggests someone has not just dropped the ball but is dribbling with it.

[ataradov]

Why do the file formats keep changing?

File format includes generating KiCad version. Even if nothing changes each major version will have a "unique" format. This might change in the release if no actual changes to the format happen.

Edit: Although it looks like it does not include the version of the creating software, but rather a version of the file format as a date. So, current v6 format for PCB is "20211014" and for the schematics "20211123". So if version changes, there must be some changes in the data that will not be understood by older versions.

New versions import old formats without issues. Sure, this creates some inconvenience for users of older versions of the software, but intervals between major versions are not that short in this case, you have plenty of time to update.

And for v6 formats absolutely had to change, as there were massive improvements. And for the schematic format had changed entirely, making it more like PCB format.

You can make an argument for more flexible fields that can be dropped by software versions that don't understand them, but in practice it will inevitably mean that older versions will not read full data. Sure, they may open it somewhat, but it will not be full data, which rarely makes sense.

[delfinom]

already getting 7.0 changes, the file format has already changed

Why do the file formats keep changing? One of the nice things about Altium is the backwards and forwards compatibility - I've used the '09 viewer to view '21 files and it works. Complains about new things it doesn't know about and outputs a list of them, but shows the essentials just fine. And, of course, the latest version opens the most ancient without quibbling.

I could understand it a format change happened once a decade or so, but every version? Suggests someone has not just dropped the ball but is dribbling with it.

Well file format change here is not a significant change. It's just there already have been numerous things added such as graphic line dash styles for PCB lines. This requires new flags being added to the PCB file format. The overall semantics however aren't changing anymore (i.e. we are sticking to sexpr).

We have no interest in forwards compatibility. It is a rabbit hole of "what ifs" and PCB design issues due to the software can be a big deal. Backwards compatibility we are committed to handling.

Just downloaded V6.00
Now I have V6 and V6.99 nightly so I can compare new features. I'll only be using V6 for any manufactured boards though.

Nothing super significant yet but there are already new DRC features, new graphic line styles for PCBs, etc
Changes are being posted here by Jeff who was working on most of these features while the rest of were tasked with preparing for a 6.0 release:
https://forum.kicad.info/t/post-v6-new-features-and-development-news/32633

There's definitely some merge requests that'll get merged in the next month or so that'll be of interest, namely
-Support for any font you want under the sun in the schematic and pcb editor
-3dconnexion mouse support (contributed by 3dconnexion) for Windows and macOS

[PlainName]

OK, thanks. I accept that forward compatibility can be a maze of twisty passages, although it's not impossible as Altium shows. However, there is the small matter of lots of money being involved to upgrade with Altium, whereas Kicad is zero cost (other than potential compatibility issues) so there is little reason not to upgrade anyway.

Season's greetings, BTW. And thanks for the Xmas prezzy

[ataradov]

-Support for any font you want under the sun in the schematic and pcb editor

How would that work? Does it mean TTF fonts? Do you embed the font somehow? What happens if the font is not present on the opening machine?

[delfinom]

-Support for any font you want under the sun in the schematic and pcb editor

How would that work? Does it mean TTF fonts? Do you embed the font somehow? What happens if the font is not present on the opening machine?

https://gitlab.com/kicad/code/kicad/-/merge_requests/613

TTF yes.

embedding/font presence issues TBD and honestly the footnote part of all the development

[Karel]

The KiCad project is proud to announce the release of version 6.0.0.

https://www.kicad.org/blog/2021/12/KiCad-6.0.0-Release/

[bateau020]

For those on MacOS: the readme has not yet been updated, and mentions 2 directories to copy. That is no longer true: /Library/ApplicationSupport/Kicad is no longer used (and you can delete it). The libs/models are now stored in /Application/Kicad/Kicad.app and fetched automatically once you set up the paths.
For all migrating from 5.1x to 6.0: If you use 3D View, you will probably want to add a path to the preferences: KISYS3DMOD should point to wherever KICAD6_3DMODEL_DIR points to. This way your old boards will be able to show 3D models.

[PlainName]

The KiCad project is proud to announce the release of version 6.0.0.

https://www.kicad.org/blog/2021/12/KiCad-6.0.0-Release/

Excellent, and thanks for the notification.

But... I just went to download and I see that Windows 7 is 'unsupported'. Nothing new there, except that Windows 8.1 is 'unsupported, will most likely still function'. The implication is that it won't work on Windows 7. Is that the case?

[delfinom]

The KiCad project is proud to announce the release of version 6.0.0.

https://www.kicad.org/blog/2021/12/KiCad-6.0.0-Release/

Excellent, and thanks for the notification.

But... I just went to download and I see that Windows 7 is 'unsupported'. Nothing new there, except that Windows 8.1 is 'unsupported, will most likely still function'. The implication is that it won't work on Windows 7. Is that the case?

We integrate python 3.9 which has switched to using windows API only present from windows 8 and onwards. Trying to run kicad 6 on windows 7 results in immediate crashes
 
Windows 8.1 will still work fine and it's not marked unsupported. Only windows 8 which Microsoft has dropped all support for a year or two ago. But it has the correct API so it'll still run.

[PlainName]

Oh, pity. I was looking forward to trying 6 but doesn't look like I will be after all

[langwadt]

Oh, pity. I was looking forward to trying 6 but doesn't look like I will be after all

some googling seems to indicated that it is possible to make python 3.9 work on win7 by copying some files from a newer windows

[PlainName]

Thanks. I'll give it a month or so until the definitive means to do it, or not, has been determined.

[holgerinput]

The kicad repo contains all patches to make their python 3.9 plugin compatible with windows 7:

https://gitlab.com/kicad/packaging/vcpkg/-/tree/kicad/ports/python3

But sadly this patch it is not activated by default.

Holger

[Quarlo Klobrigney]

I'd like to see that happen! I've been screwing around trying to get Python 3.9 working on Win 7. I even tried the Chocolatey package manager

https://chocolatey.org/ which says I can do it but then fails.....

Quote from: dunkemhigh on Today at 00:54:11

Oh, pity. Some googling seems to indicated that it is possible to make python 3.9 work on win7 by copying some files from a newer windows

[PlainName]

The kicad repo contains all patches to make their python 3.9 plugin compatible with windows 7:

https://gitlab.com/kicad/packaging/vcpkg/-/tree/kicad/ports/python3

But sadly this patch it is not activated by default.

Holger

Oooh! That could be good if it was active. Presumably Kicad needs to be built with that patch rather than having it post applied? (I am not set up to build it, don't have any knowledge of the process, and sadly the patch doesn't include instrutions.)

[holgerinput]

Hi!

Building Kicad is relatively easy

Its explained in detail here:
https://dev-docs.kicad.org/en/build/windows-msvc/

The "missing" patch can be added with a feature-flag.

In Step 1 the dependencies are installed/compiled with this line:

.\vcpkg install boost cairo curl glew gettext glm icu libxslt ngspice opencascade opengl openssl python3 wxpython wxwidgets zlib

you can enable the additional python3 patch with this line:
.\vcpkg install boost cairo curl glew gettext glm icu libxslt ngspice opencascade opengl openssl python3[deprecated-win7-support] wxpython wxwidgets zlib


The result is a complete Kicad version 6 with python 3.9 which runs fine under win7 without any restrictions.

Holger

edit:
There is one "restriction":

- Kicad "detects" this patch and shows a "nag-screen" on windows7 machines on startup.
So the developers spent A LOT OF WORK to prevent running the program on windows 7.
But you can disable the message by modifying this function:

https://gitlab.com/kicad/code/kicad/-/blob/master/libs/kiplatform/msw/app.cpp#L47

[PlainName]

Thanks, Holger. I don't have VS (and don't expect or want it, come to that) so it would have to be the msys route. Perhaps in the new year when other projects are less important...

[Quarlo Klobrigney]

Easy for you, difficult for me

I ran all the code paragraphs that was listed on the page, but all it downloaded was
"Kicad Ports Python3 vcpkg.zip" and vcpkg files in the vcpkg folder.
No other steps did anything after ".\vcpkg\vcpkg integrate install" except throw an error.

Applied user-wide integration for this vcpkg root.
All MSBuild C++ projects can now #include any installed libraries.
Linking will be handled automatically.
Installing new libraries will make them instantly available.

CMake projects should use: "-DCMAKE_TOOLCHAIN_FILE=/vcpkg/scripts/buildsystems/vcpkg.cmake"

So can you help the clueless (me) to get it  installed on my Win7?

Quote from: holgerinput on Today at 04:48:08

Hi! Building Kicad is relatively easy

Its explained in detail here:
>https://dev-docs.kicad.org/en/build/windows-msvc/

[Quarlo Klobrigney]

I checked for updates on a few bits of software that weren't installed with the Chocolatey package manager.
After installing them so they can be updated in the future, it insisted that my Kicad 5.1.12-1 was available for update to 6.00.
After clicking on update and waiting on the download, I was greeted with the error that "this is only available for win 8.1" or some such thing.
Oh well, I thought I had a way around it for Windows 7.

[delfinom]

- Kicad "detects" this patch and shows a "nag-screen" on windows7 machines on startup.
So the developers spent A LOT OF WORK to prevent running the program on windows 7.

No need to be dramatic. It took only 5 minutes to implement that.

[ataradov]

The comment in the code explains why it is necessary. They don't want to see bug reports from unsupported systems. This is not an empty or additional work, it serves the purpose bigger than just a nag screen.

This is the same reason why Linux detects "tainted" kernel with non-GPL drivers loaded.

[pierreraymondrondelle]

                   +++ RIP +++  Kicad 6,

I'm still under Win7 with NO MONEY for buying a new computer.
Thanks to all the developers for this brutal ending.

A happy new year to everybody who is still using Win7.

[JohanH]

                   +++ RIP +++  Kicad 6,

I'm still under Win7 with NO MONEY for buying a new computer.
Thanks to all the developers for this brutal ending.

A happy new year to everybody who is still using Win7.

One alternative is to install Linux on the same PC and use dual boot.

[PlainName]

Dual boot is a productivity killer. Unless there is a really really compelling reason to boot the other OS, and stay there for an appreciable time, it just won't happen after the first couple of goes.

VM would be much better but has it's own issues.

[PlainName]

The comment in the code explains why it is necessary. They don't want to see bug reports from unsupported systems. This is not an empty or additional work, it serves the purpose bigger than just a nag screen.

The program I use to do the company PAYE took this path. Instead of just saying "Well, it's Windows 7 and we don't support that" if they get such support calls (they never have from me) they actively stop the product working on Windows 7. My solution was to find a replacement, and as luck would have it it's a better product and I regret not looking around sooner!

Product inertia can count for a lot of income, which is why insurance companies think nothing of adding 50% to your premium every year - most people prefer to not change than check out the competition, so it's free money. Support must be be very expensive to be worth the loss of that income stream. (Of course, since Kicad is free the sums are very different there.)

[PlainName]

Thinking about this, I wonder if non-current Linux distros are prevented from running Kicad. I imagine there are many more versions of Linux around that are no longer supported, if they were actually supported in the first place, so is Kicad usage restricted to the known current 'big' distros? I suspect not.

And... Windows 7 shouldn't actually need support since it is a mature OS with isn't being constantly dicked around with by Microsoft. The functionality is there and fixed - no-one is going to turn it on one day and find either a completely different OS has been installed or the window furniture removed in the cause of flat-on-flat pure white display, so I would have though the support issues would be fewer than for weeping edge W11.

[JohanH]

Dual boot is a productivity killer. Unless there is a really really compelling reason to boot the other OS, and stay there for an appreciable time, it just won't happen after the first couple of goes.

VM would be much better but has it's own issues.

True, I have never bothered with dual boot myself. Switched over to Linux entirely on all my private desktops and laptops when Windows Vista came out.

[JohanH]

Thinking about this, I wonder if non-current Linux distros are prevented from running Kicad.

Yes, you will have issues in the same way. But on Linux, there is no reason to not always upgrade when there are updates available. If you somehow manage to steer into a corner needing a system that can't be upgraded, which would be a really special situation, you are on your own. I've been using desktop Linux for 15 years and my experience is that you always want to upgrade to get latest security features and support for new software etc.
Theoretically, you could try to backport on your own newer libraries to your older system so that you could run newer software. But that's a pain. Why bother doing the same that all developers out there already have done for you? By upgrading, you get all this for free.

[JohanH]

And... Windows 7 shouldn't actually need support since it is a mature OS with isn't being constantly dicked around with by Microsoft. The functionality is there and fixed

Well, that's quite an opinion. Windows 7 is completely insecure at this point. Don't ever connect a Windows 7 PC to a network or Internet or you will be eaten alive. Or keep it in a completely separate network from other devices, don't value any data you have on it, don't do anything important on it, don't put any personal data on it and don't ever connect or transfer any data from it to other PCs.

[Cerebus]

Support must be be very expensive to be worth the loss of that income stream. (Of course, since Kicad is free the sums are very different there.)

It is. For what every 1 ¹/₂ to 2 support people cost you could have a decent developer. One is an ongoing cost on existing products, the other is an investment in new products. It's easy to see how people could be tempted to scrimp.

Oddly enough people don't spot the economics when it works the other way around. For every good technical author you employ you can save on support people, and like developers once their work is done it can be 'sold' again and again for marginal cost. But when was the last time you saw a software product with good, extensive and complete documentation?

To keep it KiCad related, take a look at the 'help' for creating hierarchical documents in KiCad 6.0; it's not even grammatical, let alone good - some sentences don't even make sense like the first one in this completely unhelpful collection of paragraphs:

There are two types of hierarchy that can exist simultaneously: the first one has just been evoked and is of general use. The second consists in creating symbols in the library that appear like traditional symbols in the schematic, but which actually correspond to a schematic which describes their internal structure.

This second type is used to develop integrated circuits, because in this case you have to use function libraries in the schematic you are drawing.

Eeschema currently doesn’t treat this second case.

It appears to tell me that there are two types of hierarchy that can exist simultaneously, while also telling me that one can't exist and that the other has been evoked. Evoked by what? The smell of new mown hay?

People work on FOSS products for varied reasons. Some to work on types of projects/products that they don't get an opportunity to tackle at 'work'. Some for kudos. Some to get hold of tools that they can't get hold of economically any other way. When it comes to FOSS documentation I suspect some do it because at 'work' other people even lock up the crayons when they see them coming. "No, no, George it's OK. Don't bother yourself, we'll get someone else to write it up. [Soto voce to coworker] What does the Tea Lady do when she's not making Tea?" 

[ebastler]

                   +++ RIP +++  Kicad 6,

I'm still under Win7 with NO MONEY for buying a new computer.
Thanks to all the developers for this brutal ending.

A happy new year to everybody who is still using Win7.

Why would you need a new computer? Just install Windows 10 on the one you have; it does not need more hardware resources. Last I tried, Win 7 license keys were still accepted by Win 10, hence the upgrade is totally free.

If someone feels the need to still use Windows 7 (which has not received security patches for nearly two years now, I believe) for whatever reason, I don't mind. But please don't whine about the fact that new software released today does not support Win 7.

[Cerebus]

                   +++ RIP +++  Kicad 6,

I'm still under Win7 with NO MONEY for buying a new computer.
Thanks to all the developers for this brutal ending.

A happy new year to everybody who is still using Win7.

Why would you need a new computer? Just install Windows 10 on the one you have; it does not need more hardware resources. Last I tried, Win 7 license keys were still accepted by Win 10, hence the upgrade is totally free.

If someone feels the need to still use Windows 7 (which has not received security patches for nearly two years now, I believe) for whatever reason, I don't mind. But please don't whine about the fact that new software released today does not support Win 7.

Hey, he should demand his money back from the KiCad developers for being cruelly dumped like that! 

[langwadt]

                   +++ RIP +++  Kicad 6,

I'm still under Win7 with NO MONEY for buying a new computer.
Thanks to all the developers for this brutal ending.

A happy new year to everybody who is still using Win7.

Why would you need a new computer? Just install Windows 10 on the one you have; it does not need more hardware resources. Last I tried, Win 7 license keys were still accepted by Win 10, hence the upgrade is totally free.

If someone feels the need to still use Windows 7 (which has not received security patches for nearly two years now, I believe) for whatever reason, I don't mind. But please don't whine about the fact that new software released today does not support Win 7.

and once you have install win10 you can run any linux stuff in WSL which works very well

[PlainName]

Windows 7 is completely insecure at this point

So what? That has nothing to do with the functionality of running programs. It is not the business of an ECAD suite to worry about system security.

[PlainName]

But when was the last time you saw a software product with good, extensive and complete documentation?

Before a week ago, I wouldn't be able to tell you. But as of last week:

https://www.easydatatransform.com/

Runs out of steam in some descriptions in the help (real help, not a link to a web page short list) but tells you why you might want to use an option and not just repeat what the button label says.

[PlainName]

don't whine about the fact that new software released today does not support Win 7

The problem not that it doesn't support Win7, it is that it actively breaks on Win7. It's the difference between "If it works it works, otherwise tough shit" and "might have worked but we're making sure you can't try and find out".

[LazyJack]

Well,  Windows 7 is out of support. So why should we expect any new software support or even run on it? If someone is has no problem running an OS four major releases behind the current, then surely they would not worry about running an EDA suite one major release behind.
Supporting older versions are always a pain in the butt and takes valuable resources away from the current versions. I'd rather have KiCAD developing faster than complain about it not supporting old OSes.

[Cerebus]

don't whine about the fact that new software released today does not support Win 7

The problem not that it doesn't support Win7, it is that it actively breaks on Win7. It's the difference between "If it works it works, otherwise tough shit" and "might have worked but we're making sure you can't try and find out".

My local tame, and very competent, Windows developer says that this kind of thing is actually Microsoft's fault. APIs have appeared, disappeared and changed in such ways that supporting older and current versions of Windows means maintaining more than one branch of your source code. You can't just say "I'll use the Windows 7 API, ignore newer APIs and rely on backwards compatibility to keep it going". There are new features that you must support to have your code run and to support those you must use the new APIs, you can't just call the new APIs for the new features and the old APIs for the old features. (Received wisdom, I'm not competent to debate whether this is the case or not.)

Talking about software development in general, and having done it myself in the past, it is insanity to run parallel branches of source code if you have any way you can avoid doing so. If doing so is necessary to support say less than 5% of users, then I can fully appreciate why a sane bunch of developers would say "Tough luck, if you've a very much older OS, you're stuck with a much older version of the product".

[gnuarm]

don't whine about the fact that new software released today does not support Win 7

The problem not that it doesn't support Win7, it is that it actively breaks on Win7. It's the difference between "If it works it works, otherwise tough shit" and "might have worked but we're making sure you can't try and find out".

I seem to recall one of the reasons for having an OS was to provide a consistent programming interface across different hardware and time.  If the OS itself breaks software over time, that seems to be counter productive.  I suppose it does seem to match the thinking of the day where everything is disposable and it is expected that you throw away the old and just buy the new all the time. 

It seems to me it should be imperative that a new OS not break the old software.  But I think you are talking about a new release of the software not working on the older OS.  That's a different matter. 

Someone referred to Windows 7 as being four versions back from the current.  So Win 10, Win 8, Win 7...  I only count two.  Vista was before 7, right?  Did I miss another named release after Win 7?  Was it Windows Ralph maybe?

[gnuarm]

I have a friend who develops code for browsers to interpret.  They have to support all sorts of old platforms including IE.  Different browsers, different versions, different bugs.  It's a bloody mess. 

I have issues with some company web sites working under Firefox.  Seems they've abandoned FF and tell me to use something else.  So not everyone tries very hard to maintain compatibility.  I thought HTML 5 would work on all browsers, but it seems the coders still like to use "special features", so the leave some browsers behind. 

[langwadt]

don't whine about the fact that new software released today does not support Win 7

The problem not that it doesn't support Win7, it is that it actively breaks on Win7. It's the difference between "If it works it works, otherwise tough shit" and "might have worked but we're making sure you can't try and find out".

I seem to recall one of the reasons for having an OS was to provide a consistent programming interface across different hardware and time.  If the OS itself breaks software over time, that seems to be counter productive.  I suppose it does seem to match the thinking of the day where everything is disposable and it is expected that you throw away the old and just buy the new all the time. 

It seems to me it should be imperative that a new OS not break the old software.  But I think you are talking about a new release of the software not working on the older OS.  That's a different matter. 

yes, if you insist on an old OS, you might be stuck with old software

Someone referred to Windows 7 as being four versions back from the current.  So Win 10, Win 8, Win 7...  I only count two.  Vista was before 7, right?  Did I miss another named release after Win 7?  Was it Windows Ralph maybe?

https://en.wikipedia.org/wiki/List_of_Microsoft_Windows_versions

7,8,8.1,10,11

[JohanH]

Windows 7 is completely insecure at this point

So what? That has nothing to do with the functionality of running programs. It is not the business of an ECAD suite to worry about system security.

It's your business as computer admin. Security wise, not so big deal with an ECAD suite, but possibly other software. If you develop PC or embedded software or remotely control other devices they can become infected. And of course other devices on the same network. I would not trust a business that run computers with Win 7. Win 7 is dead. Deal with it. It's gone. Good riddance.

[gnuarm]

don't whine about the fact that new software released today does not support Win 7

The problem not that it doesn't support Win7, it is that it actively breaks on Win7. It's the difference between "If it works it works, otherwise tough shit" and "might have worked but we're making sure you can't try and find out".

I seem to recall one of the reasons for having an OS was to provide a consistent programming interface across different hardware and time.  If the OS itself breaks software over time, that seems to be counter productive.  I suppose it does seem to match the thinking of the day where everything is disposable and it is expected that you throw away the old and just buy the new all the time. 

It seems to me it should be imperative that a new OS not break the old software.  But I think you are talking about a new release of the software not working on the older OS.  That's a different matter. 

yes, if you insist on an old OS, you might be stuck with old software

Someone referred to Windows 7 as being four versions back from the current.  So Win 10, Win 8, Win 7...  I only count two.  Vista was before 7, right?  Did I miss another named release after Win 7?  Was it Windows Ralph maybe?

https://en.wikipedia.org/wiki/List_of_Microsoft_Windows_versions

7,8,8.1,10,11

I didn't know 8.1 was a "version".  Is 11 a thing?  I thought we were still on 10!  I'm obsolete and didn't know it.  I guess my software will stop running now?  Oh, wait, my software is from 2000, Codewright and Eudora.  I'm safe! 

[JohanH]

  Is 11 a thing?  I thought we were still on 10! 

Most are still on Win 10 (unless they are running Linux or Mac). But I'm already testing software on Win 11 at $day-job.

[PlainName]

My local tame, and very competent, Windows developer says that this kind of thing is actually Microsoft's fault.

Yes, it is one of the tricks Microsoft use to push users to Windows 10. No technical reason for it, just a way of fucking up people that want proper windows with actual borders and discoverable features.

[PlainName]

Windows 7 is completely insecure at this point

So what? That has nothing to do with the functionality of running programs. It is not the business of an ECAD suite to worry about system security.

It's your business as computer admin.

My computer admin says relevant precautions are in place and it is fine.

[ataradov]

I don't get the complaints here. Win has excellent backwards compatibility. Stuff from 1995 still works today.  But they have to introduce new APIs in new versions of windows, and if software uses those APIs, then it will obviously not work on older versions. Sure, you can write software using only APIs from 1995, but why?

[PlainName]

I don't get the complaints here.

Indeed. Perhaps because:

if software uses those APIs, then it will obviously not work on older versions

That's fine. But if it doesn't use those APIs? The complaint is that we are not being allowed to find out, or to provide compatible APIs. Instead, it is being actively broken.

Consider the usual open source licenses which dictate that the user must be able to replace libraries and the like with other compatible ones. That's the entire reason for having the LGPL alongside the GPL, allowing proprietary source to coexist with the principles of open source.

[ataradov]

That's fine. But if it doesn't use those APIs? The complaint is that we are not being allowed to find out, or to provide compatible APIs. Instead, it is being actively broken.

Who is stopping you from providing it? Python could check the version of the OS and provide its own implementation of the missing API if it wants to work on W7. Or not use that API at all.

You will have to rebuild from source, of course, but adding the ability to provide binary code sounds like a non-trivial thing to do. Can you do this with any OS?

[bson]

But if it doesn't use those APIs?

Win7 used CodePage 65001 and had completely fragmented UTF-8 support.  Win10 uses UTF-8 natively and has only partial legacy support for CP65001 using special interfaces intended for things like file conversion and migration to UTF-8.  The two are not compatible, and having to support both is grossly impractical since character set encoding is used everywhere, and this particular MS legacy stuff is not easily encapsulated or abstracted out.  Basically, MS has abandoned CP65001 in favor of UTF-8.  Which is a good technical reason to throw Win7 in the trashbin of history where it belongs.

[JohanH]

My computer admin says relevant precautions are in place and it is fine.

Then you obviously are competent at taking care of those things. But it should not be the default, general advice to most people!

[Quarlo Klobrigney]

And the less knowledgeable folks that just check their mail and look at cat videos. Then comes the call where they say "The computer screen was all black and I couldn't do anything so I powered it off and back on. Now the thing won't do anything for hours now!" How many times have I had that call.

Quote from: dunkemhigh on Today at 02:12:55

Yes, it is one of the tricks Microsoft use to push users to Windows 10. No technical reason for it, just a way of fucking up people that want proper windows with actual borders and discoverable features.

[delfinom]

But if it doesn't use those APIs?

Win7 used CodePage 65001 and had completely fragmented UTF-8 support.  Win10 uses UTF-8 natively and has only partial legacy support for CP65001 using special interfaces intended for things like file conversion and migration to UTF-8.  The two are not compatible, and having to support both is grossly impractical since character set encoding is used everywhere, and this particular MS legacy stuff is not easily encapsulated or abstracted out.  Basically, MS has abandoned CP65001 in favor of UTF-8.  Which is a good technical reason to throw Win7 in the trashbin of history where it belongs.

I'm confused, CP65001 _is_ literally the designation for UTF-8 on Windows 10. It must still be enabled on a per-app basis but it's more capable now. We do in Kicad v6 now tell Windows to always give us CP65001. It converts pretty much all Win32 api over to utf-8 mode and even Windows UI framework functions (which has fixed a few render bugs)

Windows 10 and 11 otherwise still default to UCS16

[gnuarm]

I don't get the complaints here. Win has excellent backwards compatibility. Stuff from 1995 still works today.  But they have to introduce new APIs in new versions of windows, and if software uses those APIs, then it will obviously not work on older versions. Sure, you can write software using only APIs from 1995, but why?

Old software only works by chance, not by design or mandate.  I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

Microsoft does not offer support for older program interfaces.  In fact, they intentionally end support for some things like help files.  I suppose there is a reason why, but I can't see where this doesn't hurt MS in the end.  I guess it's just a tiny paper cut.  I hope it doesn't get infected.   

[delfinom]

I don't get the complaints here. Win has excellent backwards compatibility. Stuff from 1995 still works today.  But they have to introduce new APIs in new versions of windows, and if software uses those APIs, then it will obviously not work on older versions. Sure, you can write software using only APIs from 1995, but why?

Old software only works by chance, not by design or mandate.  I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

Microsoft does not offer support for older program interfaces.  In fact, they intentionally end support for some things like help files.  I suppose there is a reason why, but I can't see where this doesn't hurt MS in the end.  I guess it's just a tiny paper cut.  I hope it doesn't get infected.   

The help files system aka CHM had weekly CVEs filed and they threw them in the dumpster. It's bad design dating back to the IE era of development.

I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

This is an interesting statement because Microsoft intentionally has kept the file open dialogs exactly the same across Windows versions depending on the API you use. So on Windows 10, programs using the Windows 98 era API will get the 98 era file dialog which is awful. Getting the Vista+ dialog requires using the Vista+ API.

[gnuarm]

I don't get the complaints here. Win has excellent backwards compatibility. Stuff from 1995 still works today.  But they have to introduce new APIs in new versions of windows, and if software uses those APIs, then it will obviously not work on older versions. Sure, you can write software using only APIs from 1995, but why?

Old software only works by chance, not by design or mandate.  I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

Microsoft does not offer support for older program interfaces.  In fact, they intentionally end support for some things like help files.  I suppose there is a reason why, but I can't see where this doesn't hurt MS in the end.  I guess it's just a tiny paper cut.  I hope it doesn't get infected.   

The help files system aka CHM had weekly CVEs filed and they threw them in the dumpster. It's bad design dating back to the IE era of development.

Ok, so you agree that MS does not support older program interfaces and trashes some features intentionally. 

I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

This is an interesting statement because Microsoft intentionally has kept the file open dialogs exactly the same across Windows versions depending on the API you use. So on Windows 10, programs using the Windows 98 era API will get the 98 era file dialog which is awful. Getting the Vista+ dialog requires using the Vista+ API.

I have no idea what you mean by "exactly the same across Windows versions".  The bottom line is the program has worked under every version of Windows from 95 up to at least Vista.  I think it started crapping out on Win 8, but I'm not certain (I don't have a copy of Win 7 running).  I know it works poorly under Win 10.  So where is the preservation???  When you talk about using an API, you are talking about the application using the API.  I don't think the program has changed anything.  The variable is the version of Windows. 

[delfinom]

I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

This is an interesting statement because Microsoft intentionally has kept the file open dialogs exactly the same across Windows versions depending on the API you use. So on Windows 10, programs using the Windows 98 era API will get the 98 era file dialog which is awful. Getting the Vista+ dialog requires using the Vista+ API.

I have no idea what you mean by "exactly the same across Windows versions".  The bottom line is the program has worked under every version of Windows from 95 up to at least Vista.  I think it started crapping out on Win 8, but I'm not certain (I don't have a copy of Win 7 running).  I know it works poorly under Win 10.  So where is the preservation???  When you talk about using an API, you are talking about the application using the API.  I don't think the program has changed anything.  The variable is the version of Windows.

Yes, Microsoft has elected to "box in" the file dialogs to APIs. Depending on which API you use, that's the generation of file dialog you get even on Windows 10 which is using the Vista

This page roughly shows you the three generation if dialogs https://www.codeproject.com/Articles/16678/Vista-Goodies-in-C-Using-the-New-Vista-File-Dialog

Your old programs should be using the first, Windows 2000 dialog. It works the same as it always have.

Ok, so you agree that MS does not support older program interfaces and trashes some features intentionally.

Yes, some features that are better off dead. Nobody used those horrible CHM interfaces.

[gnuarm]

I have some old software that works under Win 10, but just barely.  The File Open dialog is really messed up and many other features of the UI are crippled. 

This is an interesting statement because Microsoft intentionally has kept the file open dialogs exactly the same across Windows versions depending on the API you use. So on Windows 10, programs using the Windows 98 era API will get the 98 era file dialog which is awful. Getting the Vista+ dialog requires using the Vista+ API.

I have no idea what you mean by "exactly the same across Windows versions".  The bottom line is the program has worked under every version of Windows from 95 up to at least Vista.  I think it started crapping out on Win 8, but I'm not certain (I don't have a copy of Win 7 running).  I know it works poorly under Win 10.  So where is the preservation???  When you talk about using an API, you are talking about the application using the API.  I don't think the program has changed anything.  The variable is the version of Windows.

Yes, Microsoft has elected to "box in" the file dialogs to APIs. Depending on which API you use, that's the generation of file dialog you get even on Windows 10 which is using the Vista

This page roughly shows you the three generation if dialogs https://www.codeproject.com/Articles/16678/Vista-Goodies-in-C-Using-the-New-Vista-File-Dialog

Your old programs should be using the first, Windows 2000 dialog. It works the same as it always have.

If that is true, why does my program work correctly under WinMe, W2K, WinXP, WinVista... but not under Win10?

Ok, so you agree that MS does not support older program interfaces and trashes some features intentionally.

Yes, some features that are better off dead. Nobody used those horrible CHM interfaces.

Of course they did.  That's why so many old programs no longer have help.  It was used extensively and now unsupported.  I don't know why you call them "horrible".  I don't care about the programming if that's what you mean.  To the user it is much, much better than having no help at all.  There's no longer even a manual you can find on the web.

[ataradov]

Of course they did.  That's why so many old programs no longer have help.  It was used extensively and now unsupported.

There are a lot of independent CHM viewers.

Not supporting it menas that new applications would not use it as a help format. For old applications - just use a standalone viewer.

[PlainName]

Nobody used those horrible CHM interfaces.

What? You cannot be serious! The only local help I seem to have all come as either .hlp or .chm (mostly the latter). And a quick check shows 1955 of them on the filesystem.

They may be horrible to you, but as the only source of help info they are still jolly useful.

[PlainName]

Not supporting it menas that new applications would not use it as a help format.

Pah! 27 new ones to my system in 2021 used it.

[ataradov]

Pah! 27 new ones to my system in 2021 used it.

But do they come from 2021 software? Of course you will always have some in any year if you keep installing software from 1995.

[PlainName]

Latest one was started in 2019.

Edit: and I  would like it known that I really REALLY very seriously HATE bloody online webpages as help, which seems to be the trend you're referring to.

[alexanderbrevig]

I just had to make a Gruvbox theme!

You can use it too: https://github.com/AlexanderBrevig/kicad-gruvbox-theme 

[0xFFF0]

How to overcome the the installation lock on Windows 7? 

[ElektroQuark]

Opening CHM files natively on Windows 10 and 11:

CLICK!

[SiliconWizard]

How to overcome the the installation lock on Windows 7?

Yep, I just tried before reading the latest posts here, and noted Windows 7 was not supported anymore. How sweet. =)

Then, in case it was just an "installation lock" as you said, I tried extracting the files directly. You can open the installer exe as an archive, with 7-zip for instance, and extract the files in a directory. You get the same structure as a Kicad install.

But, as one could suspect, Kicad doesn't start. If they "locked" this in the installer, that was for a good reason: the executables were built for Windows 8.1 minimum. So the executables require some DLLs that don't exist on Windows 7. Digging a bit deeper, it seems to mainly (not quite sure it's the only cause though) come from the fact they are linked against Python 3.9, which itself doesn't support Windows 7 anymore.

I don't know yet if it's possible to build Kicad 6 ourselves (using MSYS2 for instance) for Windows 7. But meanwhile, we're screwed.

[PlainName]

Python 3.9, which itself doesn't support Windows 7 anymore.

I haven't looked into that yet, but my StreamDeck doesn't support Windows 7 in that it won't install, and if you extract the files it won't run because some DLLs are missing. The trick there is to shove fake DLLs in the install directory, and it runs. The only purpose of those DLLs is to answer the question "is this W8.1+?". The DLLs do nothing other than fake that. Maybe Python9 is similar.

[SiliconWizard]

Python 3.9, which itself doesn't support Windows 7 anymore.

I haven't looked into that yet, but my StreamDeck doesn't support Windows 7 in that it won't install, and if you extract the files it won't run because some DLLs are missing. The trick there is to shove fake DLLs in the install directory, and it runs. The only purpose of those DLLs is to answer the question "is this W8.1+?". The DLLs do nothing other than fake that. Maybe Python9 is similar.

Nah, unfortunately. This is a bunch of api-ms-win* DLLs here that were introduced in Win 8 or 8.1 and I don't think you can fake them to get any useful result. They contain a lot of system functions.
If someone manages to "fake" that, that would make a number of applications usable in Win 7, such as Python 3.9+, Vivado, KiCad, and many others...

Now I tried to build KiCad from source, that can be done on Windows with MSYS2, and would thus yield executables working on Win 7. This is documented there: https://dev-docs.kicad.org/en/build/windows-msys2/ , and I am used to building cross-platform apps using MSYS2, so no problem with that. So far, it won't build though, because KiCad 6 apparently requires the very latest version of wxWindows (3.1.5 IIRC), while MSYS still has wxWindows 3.0... so we'll have to wait till MSYS2 updates wxWindows. And then I'll keep you posted. If it does build, that will be a solution.

[nctnico]

Building wxwidgets from source is not that difficult on Windows. Though I find it a bit ignorant that Kicad binaries don't work on Windows 7 out of the box. Most of the professional CAD software does because a lot of places still use Windows 7.

[Karel]

Now I tried to build KiCad from source, that can be done on Windows with MSYS2, and would thus yield executables working on Win 7. This is documented there: https://dev-docs.kicad.org/en/build/windows-msys2/ , and I am used to building cross-platform apps using MSYS2, so no problem with that. So far, it won't build though, because KiCad 6 apparently requires the very latest version of wxWindows (3.1.5 IIRC), while MSYS still has wxWindows 3.0... so we'll have to wait till MSYS2 updates wxWindows. And then I'll keep you posted. If it does build, that will be a solution.

Wouldn't it be easier to use the Linux version of KiCad?
You'll have to dual-boot with windows 7 which is not ideal, but still saves you much time,
also because I expect monthly updates for KiCad V6.

[PlainName]

You'll have to dual-boot with windows 7 which is not ideal

Dead loss. I often fire up Altium briefly to check something, and if that means saving and closing everything, restarting with a differerent OS then 2 mins later do the reverse... Not going to happen.

[Karel]

You'll have to dual-boot with windows 7 which is not ideal

Dead loss. I often fire up Altium briefly to check something, and if that means saving and closing everything, restarting with a differerent OS then 2 mins later do the reverse... Not going to happen.

In that case, what about running altium in VMware on Linux? You'll have to pay $249 but that's one time only.

[nctnico]

You'll have to dual-boot with windows 7 which is not ideal

Dead loss. I often fire up Altium briefly to check something, and if that means saving and closing everything, restarting with a differerent OS then 2 mins later do the reverse... Not going to happen.

In that case, what about running altium in VMware on Linux? You'll have to pay $249 but that's one time only.

That is another option. The free VMWare allows for only one VM but that could be a limitation because I recon you'll want to run more VMs simultaneously.

[PlainName]

What's the $249 for?

Main problem for me is I have a significant investment in Windows apps, many of which are not functionally replicated on Linux. I am not about to through all that away just to try Kicad 6. And without trying the possibility that I might find it useful (and perhaps review my priorities) is pretty remote, so it's a bit of a  circle there.

[PlainName]

The free VMWare allows for only one VM

I have a not-too-recent Workstation license, but prefer VirtualBox, particularly for host integration.

[Cerebus]

You'll have to dual-boot with windows 7 which is not ideal

Dead loss. I often fire up Altium briefly to check something, and if that means saving and closing everything, restarting with a differerent OS then 2 mins later do the reverse... Not going to happen.

In that case, what about running altium in VMware on Linux? You'll have to pay $249 but that's one time only.

Why, if one was a Linux user, would one choose a proprietary hypervisor when there are perfectly good free ones? The only reason to be running VMware workstation under Linux that I can see is if one wanted to maintain integration/compatibility with a VMware ESXi estate, or if VMware is all that one understood how to 'drive'. (I've nothing against VMware itself, I use VMware fusion myself on macOS, but I do have ESXi machines to integrate with.)

[Karel]

What's the $249 for?

It's the price of a VMware Workstation Pro license.

[Karel]

Why, if one was a Linux user, would one choose a proprietary hypervisor when there are perfectly good free ones?

Because, from what I understood from some other people here who already tried it,
altium (the layout part) doesn't seem to work reliable/stable with other hypervisors on Linux.

[madires]

I haven't tested Altium. However, for software requiring 3D capabilties you have to make sure that 3D is set up properly on the linux host, that 3D support is enabled for the VM and that the VM specific driver package is installed. Alternatively - if supported - you can also try GPU pass-through.

[ebastler]

Wouldn't it be easier to use the Linux version of KiCad?
You'll have to dual-boot with windows 7 which is not ideal [...]

If one is mainly working in a Windows environment, why not upgrade from Win 7 to Win 10? Doesn't need more hardware resources, receives regular updates and security patches, license is free (using the Win 7 key), and should run whatever Windows software one uses besides KiCad.

Privacy controls have much improved in recent Win 10 versions, by the way. It actively asks during the installation process whether you want various "services" (location, ad personalization...) or not. The only nuisance I am aware of is that Microsoft really wants you to create an online Microsoft account and use it as the user account on your machine -- but you can still bypass that by not having an internet connection while walking through the initial Win 10 setup.

Am I missing anything else? Why would one insist on keeping Win 7?

[nctnico]

Why, if one was a Linux user, would one choose a proprietary hypervisor when there are perfectly good free ones? The only reason to be running VMware workstation under Linux that I can see is if one wanted to maintain integration/compatibility with a VMware ESXi estate, or if VMware is all that one understood how to 'drive'. (I've nothing against VMware itself, I use VMware fusion myself on macOS, but I do have ESXi machines to integrate with.)

This is a bit off-topic but what VMware can and Virtualbox does not is being able to run 3D accelleration in a virtual machine. Or better put: all my attempts to use 3D (like OpenGl) in Virtualbox has failed where with VMware it worked right out of the box.

[Cerebus]

Why, if one was a Linux user, would one choose a proprietary hypervisor when there are perfectly good free ones? The only reason to be running VMware workstation under Linux that I can see is if one wanted to maintain integration/compatibility with a VMware ESXi estate, or if VMware is all that one understood how to 'drive'. (I've nothing against VMware itself, I use VMware fusion myself on macOS, but I do have ESXi machines to integrate with.)

This is a bit off-topic but what VMware can and Virtualbox does not is being able to run 3D accelleration in a virtual machine. Or better put: all my attempts to use 3D (like OpenGl) in Virtualbox has failed where with VMware it worked right out of the box.

I've had Altium working "well enough" in a VirtualBox VM on Linux and macOS, but I'd agree one's chances of first time success are higher with VMWare.

[chickenHeadKnob]

Why, if one was a Linux user, would one choose a proprietary hypervisor when there are perfectly good free ones? The only reason to be running VMware workstation under Linux that I can see is if one wanted to maintain integration/compatibility with a VMware ESXi estate, or if VMware is all that one understood how to 'drive'. (I've nothing against VMware itself, I use VMware fusion myself on macOS, but I do have ESXi machines to integrate with.)

This is a bit off-topic but what VMware can and Virtualbox does not is being able to run 3D accelleration in a virtual machine. Or better put: all my attempts to use 3D (like OpenGl) in Virtualbox has failed where with VMware it worked right out of the box.

I've had Altium working "well enough" in a VirtualBox VM on Linux and macOS, but I'd agree one's chances of first time success are higher with VMWare.

A few pages back I complained that I couldn't get the 6.0 nightlies running in virtualbox mint linux 20.2. What was happening (I think) was that eeschema,PCBnew ect were seg-faulting on graphics. The only reason I was trying this on VirtualBox was that I was loath to upgrade my base system (Mint 18.x) as I had invested a lot of time tweaking everything. Instead of diagnosing further I bit the bullet and shit-canned the whole thing and did a clean install of Mint 20.2 and natively run 6.0 official which works fine so far. The whole upgrade of my system was a big dick-around I wanted to avoid but that has nothing to do with Kicad.

[holgerinput]

Python 3.9, which itself doesn't support Windows 7 anymore.

I haven't looked into that yet, but my StreamDeck doesn't support Windows 7 in that it won't install, and if you extract the files it won't run because some DLLs are missing. The trick there is to shove fake DLLs in the install directory, and it runs. The only purpose of those DLLs is to answer the question "is this W8.1+?". The DLLs do nothing other than fake that. Maybe Python9 is similar.

Nah, unfortunately. This is a bunch of api-ms-win* DLLs here that were introduced in Win 8 or 8.1 and I don't think you can fake them to get any useful result. They contain a lot of system functions.
If someone manages to "fake" that, that would make a number of applications usable in Win 7, such as Python 3.9+, Vivado, KiCad, and many others...

Now I tried to build KiCad from source, that can be done on Windows with MSYS2, and would thus yield executables working on Win 7. This is documented there: https://dev-docs.kicad.org/en/build/windows-msys2/ , and I am used to building cross-platform apps using MSYS2, so no problem with that. So far, it won't build though, because KiCad 6 apparently requires the very latest version of wxWindows (3.1.5 IIRC), while MSYS still has wxWindows 3.0... so we'll have to wait till MSYS2 updates wxWindows. And then I'll keep you posted. If it does build, that will be a solution.

HI!

I've explained a working solution two or three pages earlier:
https://www.eevblog.com/forum/kicad/kicad-6-is-coming!/msg3897506/#msg3897506
"vcpkg/vs2019" is the officially way to build kicad 6.0.
If you enable the win7 compatibility patch you'll get a set of binaries which run fine under win7

[SiliconWizard]

Now I tried to build KiCad from source, that can be done on Windows with MSYS2, and would thus yield executables working on Win 7. This is documented there: https://dev-docs.kicad.org/en/build/windows-msys2/ , and I am used to building cross-platform apps using MSYS2, so no problem with that. So far, it won't build though, because KiCad 6 apparently requires the very latest version of wxWindows (3.1.5 IIRC), while MSYS still has wxWindows 3.0... so we'll have to wait till MSYS2 updates wxWindows. And then I'll keep you posted. If it does build, that will be a solution.

Wouldn't it be easier to use the Linux version of KiCad?
You'll have to dual-boot with windows 7 which is not ideal, but still saves you much time,
also because I expect monthly updates for KiCad V6.

Yeah no. If I'm going to use Windows, I don't want this hassle. Otherwise, I'd switch completely to Linux. It's likely going to happen, but not quite yet. I have a Linux headless box connected to the network, that I can access through ssh or VNC for GUI apps. But routing PCBs using VNC, no =)

Note that the freshly released KiCad 6.0 on some Linux distros is still a bit shaky. On the aforementioned box, there are issues with the schematic and PCB editor. We'll probably have to wait for a few more days until the first dependency quirks are ironed out.

I don't hugely mind having to build KiCad from source. That's something I've not done so far, but that I was considering doing anyway.

[SiliconWizard]

Building wxwidgets from source is not that difficult on Windows. Though I find it a bit ignorant that Kicad binaries don't work on Windows 7 out of the box. Most of the professional CAD software does because a lot of places still use Windows 7.

You're right about wxWidgets, but MSYS2 already had it as a package, and I want to limit the number of extra dependencies I have to build manually, especially when they would "clash" with the official packages. But yes, that would be an option!

As to Windows 7, yes. I can understand why many commercial vendors have now stopped supporting Windows 7, and now also build their apps so that they CAN'T even run on it (like Vivado), but for open-source software, it's a lot more questionable.

Ironically, Firefox, for instance, still perfectly works on Windows 7. And you'd think if the security of the OS was the real reason for stopping supporting it, a web browser would be much higher on the list than CAD software...

That said, for open-source software, distributed binaries are a bonus and they come with some limitations in a number of projects, because maintainers probably don't want the hassle of having to build for tens of different platforms, and test, even minimally, the artefacts. So if we have to build it from source - which certainly would be a major drawback for many people - I don't hugely mind. As long as building it is actually possible without having to go through multiple hoops, or having to use very specific tools and versions for everything.

[Smokey]

... A few pages back I complained that I couldn't get the 6.0 nightlies running in virtualbox mint linux 20.2. ...

Not sure if it's related, but I had similar issues with 6rc1 on an ubuntu virtualbox.  Just crashed a lot when I tried to open designs.  I figured out that you need to turn up all the graphics setting on the virtual machine.  Give it more graphics memory specifically.  Then it mostly worked.

[Smokey]

I'll further add to this ongoing tangent...
I'm sort of surprised that so many people are seeing dropping windows7 support as a show stopper.  Windows7 was good.  Accepted.  But windows 10 is also good.  And they give it away for free to all the windows7 users.  Really no great excuse to still being running windows7.  And I'm running one box with some old ass 2011 era hardware and windows10 works fine. 

[The Doktor]

Python 3.9, which itself doesn't support Windows 7 anymore.

I haven't looked into that yet, but my StreamDeck doesn't support Windows 7 in that it won't install, and if you extract the files it won't run because some DLLs are missing. The trick there is to shove fake DLLs in the install directory, and it runs. The only purpose of those DLLs is to answer the question "is this W8.1+?". The DLLs do nothing other than fake that. Maybe Python9 is similar.

Nah, unfortunately. This is a bunch of api-ms-win* DLLs here that were introduced in Win 8 or 8.1 and I don't think you can fake them to get any useful result. They contain a lot of system functions.
If someone manages to "fake" that, that would make a number of applications usable in Win 7, such as Python 3.9+, Vivado, KiCad, and many others...

Now I tried to build KiCad from source, that can be done on Windows with MSYS2, and would thus yield executables working on Win 7. This is documented there: https://dev-docs.kicad.org/en/build/windows-msys2/ , and I am used to building cross-platform apps using MSYS2, so no problem with that. So far, it won't build though, because KiCad 6 apparently requires the very latest version of wxWindows (3.1.5 IIRC), while MSYS still has wxWindows 3.0... so we'll have to wait till MSYS2 updates wxWindows. And then I'll keep you posted. If it does build, that will be a solution.

HI!

I've explained a working solution two or three pages earlier:
https://www.eevblog.com/forum/kicad/kicad-6-is-coming!/msg3897506/#msg3897506
"vcpkg/vs2019" is the officially way to build kicad 6.0.
If you enable the win7 compatibility patch you'll get a set of binaries which run fine under win7

Do you know of any place where I can download Win7 compatible binaries?

[PlainName]

But windows 10 is also good

Windows 10 is pretty much the same as Windows 7 under the bonnet so technically it is fine. The problem with it is that the user interface is shit, confusing and ignores even Microsoft's own guidelines. But it has to do that so it can be different and 'new'. There's also the forced updates, lack of quality control, the push online income streams, etc.

[SiliconWizard]

But windows 10 is also good

Windows 10 is pretty much the same as Windows 7 under the bonnet so technically it is fine. The problem with it is that the user interface is shit, confusing and ignores even Microsoft's own guidelines. But it has to do that so it can be different and 'new'. There's also the forced updates, lack of quality control, the push online income streams, etc.

Yep, that's it in a nutshell. The UI is atrocious, and it's not even pretty. The updates are bound to break things on a regular basis. The amount of telemetry is insane, and as some have shown, even when disabling everything that can be disabled, there's still some phoning home. The only sure way of stopping that is to cut the internet access, which, uh, is not really practical for everyday use.

If it was just a better 7, I would have long updated.

[Karel]

And as long as most people continue to use windows, microsoft gets away with it and nothing will change.

Only drug dealers and software companies call their customers "users".

[John B]

I started shifting over projects from version 5 to 6. Firstly I needed to add my custom symbols and footprints, but it's also worth noting there's quite a few changes to the standard library of symbols. Quite a few of the standard schematic symbols need to have the library reference updated, otherwise when you go into the PCB editor it will delete the footprint. I was partway through a project, so the schematic was complete but the PCB was half done. Probably my own fault there...

The 3D model references all need to be reselected too, otherwise nothing will show up in the 3D view of the PCB, but that seems to be purely cosmetic. The symbols in the schematic can be updated in bulk with only a few clicks but I can't see how to update all 3D models in one go.

Also, does anyone find the path finding ability when laying tracks almost broken? The PCB editor starts to lag badly after 1 or 2 bend in the tracks and never wants to lay a track through obviously large enough gaps unless its only 1 bend. There has obviously been significant changes in that regard compared to version 5.

[SiliconWizard]

I started shifting over projects from version 5 to 6. Firstly I needed to add my custom symbols and footprints, but it's also worth noting there's quite a few changes to the standard library of symbols. Quite a few of the standard schematic symbols need to have the library reference updated, otherwise when you go into the PCB editor it will delete the footprint. I was partway through a project, so the schematic was complete but the PCB was half done. Probably my own fault there...

The 3D model references all need to be reselected too, otherwise nothing will show up in the 3D view of the PCB, but that seems to be purely cosmetic. The symbols in the schematic can be updated in bulk with only a few clicks but I can't see how to update all 3D models in one go.

Also, does anyone find the path finding ability when laying tracks almost broken? The PCB editor starts to lag badly after 1 or 2 bend in the tracks and never wants to lay a track through obviously large enough gaps unless its only 1 bend. There has obviously been significant changes in that regard compared to version 5.

Thanks for the heads up. I've tried it on Linux, and only very little so far, and indeed opening any existing project made with v5 was a 'weird' experience. I can confirm the no 3D model and haven't investigated it so far, so thanks for the info.

All in all, there is a LOT of new features, new formats, new everything, and if you have to work on existing projects, at least professionally, I'd suggest waiting for a while...

[PlainName]

And as long as most people continue to use windows, microsoft gets away with it and nothing will change.

Only drug dealers and software companies call their customers "users".

As long as most people switch to Windows 10, you mean. Nothing wrong with Windows pre-8 that couldn't have been done without the screwup of Metro. But instead of holding out, like happened to force a rethink of Win8 and ME, everyone succumbs to the W10 giveaway and Microsoft's dirty tricks.

How's systemd going for you?

[ataradov]

How's systemd going for you?

I personally like systemd and it improved my life quite a bit. I had no issues with it, and creating services never been easier.

As far as migration goes, I just use my own libraries for everything to avoid issues like that. It is more work at the time, but pays off in long run.

[Cerebus]

Note that the freshly released KiCad 6.0 on some Linux distros is still a bit shaky.

I suspect that is "all platforms". The huge holes in the help files suggests that someone doesn't quite know what "finished and releasable" means. That kind of thinking about something so user visible doesn't bode well for what's going on behind the scenes.

e.g. The entire "Footprints and footprint libraries" section is all TODO.

7. Footprints and footprint libraries
7.1. Managing footprint libraries
NOTE
TODO: Write this section
7.2. Creating and editing footprints
NOTE
TODO: Write this section
7.2.1. Custom pad shapes
7.2.2. Footprint attributes
NOTE
Mention net ties here
7.2.3. Footprint wizards
For more information about creating new footprint wizards, see the Scripting section of the Advanced Topics chapter.

8. Advanced topics

It's not as if this is the only place, there are 29 occurrences of "TODO" in the pcbnew help file alone.

Personally I find this very embarrassing. I've been doing quite a lot of proselytising recently telling people who have previously tried and discarded KiCad things like "No, you should try KiCad again, they're getting better and better". Now I'm going to have to eat my words, "No, you were right, it is still amateur hour.".

[Smokey]

And as long as most people continue to use windows, microsoft gets away with it and nothing will change.

Only drug dealers and software companies call their customers "users".

As long as most people switch to Windows 10, you mean. Nothing wrong with Windows pre-8 that couldn't have been done without the screwup of Metro. But instead of holding out, like happened to force a rethink of Win8 and ME, everyone succumbs to the W10 giveaway and Microsoft's dirty tricks.
...

... Just think of the quality of horses we could have had if everyone had just "held out" on that whole automobile fad in 1909.   

I tend to agree with you that the metro UI is stupid, and I personally think Ubuntu Unity interface is worse, but it's crazy to say we should all be holding out on Ubuntu 10.10 from 2010 as some sort of protest.

[PlainName]

it's crazy to say we should all be holding out

That's the rational response to "as long as most people continue to use windows, microsoft gets away with it". The argument would be just which version we are discussing, and whether we let them or don't let them 'get away with it'.

However, much of the bad stuff that happens does so precisely because we 'let them get away with it'. If we don't nip it in the bud then whatever the bad thing is, it ratchets up to become the new normal and harder to stop. One upon a time a company called Phorm tried to force adverts on us by - get this - inspecting our Internet traffic for context! But they had to back down precisely because there was a big fuss about it. Fast forward to today and most people are quite content with the likes of Google and Facebook doing far worse than that, because they managed to slip the stuff by us in a way that only a few could object. We have let them get away with it.

It is crazy to think we should just sit back and let whoever do whatever they want without any hassle from us. No, if we are averse to their plans for global domination we should make that known and NOT let them get away with it. The problem is merely what qualifies for such action - how does the average pleb know that this is the cause to stand behind, which will (probably only temporarily) head off the later bad shit.

Nevertheless, we did almost do for Windows 10. Microsoft had to stiff Metro, at least notionally, and they still only managed to get it past us by giving the thing away free and actively stopping stuff working on Windows 7. We did indeed try to hold out, but weakness got the better of many of us, and here we are. (And I would like to take this opportunity to say "it is what it is".) 

[JohnG]

It's not as if this is the only place, there are 29 occurrences of "TODO" in the pcbnew help file alone.

Personally I find this very embarrassing. I've been doing quite a lot of proselytising recently telling people who have previously tried and discarded KiCad things like "No, you should try KiCad again, they're getting better and better". Now I'm going to have to eat my words, "No, you were right, it is still amateur hour.".

In Altium, they don't bother listing the TODOs, so you don't know they are missing until you get bitten by something. You are free to discover what is missing or undocumented by yourself. Maybe if you are lucky, you will find the solution on some online forum somewhere. Or, you can put in a ticket, and be told that "your problem is not actually a problem, according to our marketing". And if it's really a problem, you are SOL until the next version. Jest sayin'...

Kicad is not yet a replacement for Altium, but it sure seems on the right track. For my company, the key missing feature is database library support, so we continue to use Altium (which does the job for us, despite my complaining).

John

[djsb]

Instead of complaining about the help files, do something about it. I spent 3 years (2004-2007) translating the original French help files into English with the help of several volunteer translators. I started off using machine translation as I don't speak French.

[Cerebus]

Instead of complaining about the help files, do something about it. I spent 3 years (2004-2007) translating the original French help files into English with the help of several volunteer translators. I started off using machine translation as I don't speak French.

No, not having this. Answering a criticism of "Fundamental, basic, documentation is missing in what is supposed to be a release" with "You should write it" is an asinine response. Complete sections of basic help missing in a release? No, this is basic amateur hour - "Let's release this with no documentation at all on footprints and footprint tools and don't forget to not document zones or pads too". Literally everything that I've needed to look up in the help in the past few days has led to a "TODO". Would it be OK to "release" it without the footprint editor in place? No. Without tools to manipulate pads or zones? No. So why is it OK to "release" it without the documentation necessary to use that those?

4.4. Working with footprints
NOTE
TODO: Write this section - covers footprint properties, updating from library, etc.
4.5. Working with pads
NOTE
TODO: Write this section - covers pad properties
4.6. Working with zones
NOTE
TODO: Write this section

The truly stupid thing is that this documentation existed for 5.x. At least the 5.99 nightlies had the old helpfiles with basic information albeit out of date. So did 6.0 RC1. In 6.0 it's missing entirely. It would have been better to release 6.0 with the 5.x helpfiles in place with a caveat of "these are out of date" but someone decided to replace them after the release candidate stage with "new for 6.0" documentation that wasn't even half finished.

The issue here isn't that the documentation needs work to bring it up to date, the issue is that someone or some people made the decision to push 6.0 to "release" status with significant chunks of fundamental, crucial documentation missing entirely. It's telling that the response of a typical KiCad fanboy isn't "Oh, that's bad!", but "It's your fault becuase you didn't volunteer to fix it.". Volunteering to fix the documentation post facto wouldn't fix an incredibly poor, amateurish project management decision. In the commercial world a decision like that would get a project manager fired the next working day after the "release" got into the hands of customers with the letter from HR citing "reputational damage", "damage to the standing of the product in the marketplace", "costs to rectify", "loss of goodwill" and so on.

[JohnG]

In the commercial world a decision like that would get a project manager fired the next working day after the "release" got into the hands of customers with the letter from HR citing "reputational damage", "damage to the standing of the product in the marketplace", "costs to rectify", "loss of goodwill" and so on.

Yes, I suppose that this occasionally happens with commercial software.

What normally happens is that they circle the wagons around the executive who made the decision (and who got a bonus for early or on-time release), and maybe find some scapegoat engineer to fire if the heat gets bad enough...

John

[SiliconWizard]

Anyway, I managed building it on MSYS2. Actually, MSYS2 has both wxWidgets 3.0 and 3.1 packages, so it was a matter of forcing using 3.1 - needed some tweaking in one CMake file. After that, the build process went smoothly.

But was it all it would take? You bet your ass not. Kicad executables built this way run fine on Windows 7. But. Once you have passed the initial dialog asking you if you want to import your old settings (oh, I was happy when I got this dialog!), KiCad then greets you with a nice message "This operating system is not supported by KiCad and its dependencies. Any issues with KiCad on this system cannot be reported to the official bugtracker."

At this point, I thought, OK, annoying, but it's probably just a one-time warning and you can move on. You think? Nope! Once you click OK, the app just exits.
One can probably patch the source code to remove this, I'll have to see. But good lord, the KiCad team has gone to greath lengths to make sure Windows 7 users would suffer forever and not be able to use KiCad ever.

Yeah, I know some of you will say that it's perfectly fine, that Win 7 is dead, that... etc. Let's not go over this again, this has been done tens of times in other threads already. Point is, whereas I can understand it from a commercial vendor, I find it unacceptable for open-source software. Open-source software should never add anything to explicitely prevent people from building and/or running the software on any platform they see fit, under their own responsibility. I find this pretty bad. As I said earlier, even Firefox, which would be much more sensitive to potential security issues, has not done this.

So, I'll keep you posted, but I can say I do not like what they did here.

[voltsandjolts]

Point is, whereas I can understand it from a commercial vendor, I find it unacceptable for open-source software. Open-source software should never add anything to explicitely prevent people from building and/or running the software on any platform they see fit, under their own responsibility.

I see your point but I imagine their reasoning was something like 'moving to newer api's makes our life easier and we simply don't have the resources to support deprecated api's or systems'

I find this pretty bad. As I said earlier, even Firefox, which would be much more sensitive to potential security issues, has not done this.

I think your comparison is pretty bad.

Combined income of Mozilla Foundation and Mozilla Corporation
2015    $420 million       
2016    $520 million       
2017    $562 million       
2018    $436 million

I really don't think they took delight in screwing win7, they just felt they had to move on and put the resources they have to best use.

[SiliconWizard]

So, I looked up in the source code to locate this check, and got the bottom line. It's only, and entirely due to Python. From comments in the source code:

// Python 3.8 switched to Windows 8+ API, we do not support Windows 7 and will not
    // attempt to hack around it. A normal user will never get here because the Python DLL
    // is missing dependencies - and because it is not dynamically loaded, KiCad will not even
    // start without patching Python or its WinAPI dependency. This is just to create a nag dialog
    // for those who run patched Python and prevent them from submitting bug reports.

So it's nothing related to KiCad source itself, or even wxWidgets, but Python only. There's actually another app I use that requires Python, and the same problem - latest versions don't run anymore.

Now, under MSYS2, Python is at 3.9.7, and it works - so the MSYS2 Python package is built accordingly. This is why KiCad builds without an issue and has no problem launching... But still exits when it initializes Python. The reason is that I try running the built KiCad outside of MSYS2. One easy workaround is to run KiCad from within MSYS2 - actually, I noticed MSYS2 has a package for KiCad 6.0.0, so it should work there.

But I'd like to make it independent of MSYS2. And, I don't much care about Python scripting anyway in KiCad, so I'm currently re-building without this. Of course, it should also be possible to build Python 3.9 so that it runs on Win 7, and use that. Might look into that later.

I'll keep you posted... I guess a few people at least may be interested. And if not, at least you have the bottom line.

[nctnico]

So, I looked up in the source code to locate this check, and got the bottom line. It's only, and entirely due to Python. From comments in the source code:

// Python 3.8 switched to Windows 8+ API, we do not support Windows 7 and will not
    // attempt to hack around it. A normal user will never get here because the Python DLL
    // is missing dependencies - and because it is not dynamically loaded, KiCad will not even
    // start without patching Python or its WinAPI dependency. This is just to create a nag dialog
    // for those who run patched Python and prevent them from submitting bug reports.

So it's nothing related to KiCad source itself, or even wxWidgets, but Python only. There's actually another app I use that requires Python, and the same problem - latest versions don't run anymore.

Now, under MSYS2, Python is at 3.9.7, and it works - so the MSYS2 Python package is built accordingly. This is why KiCad builds without an issue and has no problem launching... But still exits when it initializes Python. The reason is that I try running the built KiCad outside of MSYS2. One easy workaround is to run KiCad from within MSYS2 - actually, I noticed MSYS2 has a package for KiCad 6.0.0, so it should work there.

But I'd like to make it independent of MSYS2. And, I don't much care about Python scripting anyway in KiCad, so I'm currently re-building without this. Of course, it should also be possible to build Python 3.9 so that it runs on Win 7, and use that. Might look into that later.

I'll keep you posted... I guess a few people at least may be interested. And if not, at least you have the bottom line.

Maybe you can get Kicad to make your binaries -if you get that far- available on their website. I'm sure you'll be doing a lot of people a favour. Not supporting Windows7 is way too soon.

[madires]

Based on https://docs.microsoft.com/en-us/lifecycle/faq/extended-security-updates the ESU for WIndows 7 ends:
- January 10, 2023 for Professional, Enterprise, and Professional for Embedded Systems
- October 10, 2023 for Windows Embedded Standard 7
- October 8, 2024 for Windows Embedded POSReady 7

[LazyJack]

How about creating a new topic on whining about Windows 7 support by KiCAD and staying with actual V6.0 news in this topic.

[PlainName]

Isn't W7 support actual Kicad 6.0 news?

[Cerebus]

How about creating a new topic on whining about Windows 7 support by KiCAD and staying with actual V6.0 news in this topic.

In what way is Windows 7 support, or rather lack of it, in V6.0 not news related to V6.0? And characterising it as "whining" is not helpful, but rather smacks of uncritical fanboyism.

Windows support is of no interest to me, I'm a macOS/Linux/Unix boy, but that doesn't mean that I regard it as unimportant or unworthy of discussion. After all, users of lesser operating systems need all the help they can get. 

[SiliconWizard]

Just a bit of follow-up. (Regardless of those saying it is "whining"... meanwhile, we are trying to find solutions rather than just "whine", while people saying we "whine" are themselves, in fact, just whining about us discussing this.  )

* While up to KiCad 5, it was possible to configure builds without Python scripting support (it was just a matter of a couple build options), they have apparently stopped making it optional. The only option you have now is to disable wxPython (giving you wxWidgets from within your Python scripts), but not Python scripting itself. It's probably now considered a mandatory part of KiCad.

* I did manage to build a fully working KiCad for Windows 7, including Pcbnew which was the hardest part - this is the one that absolutely requires Python, the other KiCad executables can run without it. But so far, it's using Python 3.9 from MSYS2 and I need to set a couple environment variables, which is cumbersome. I am currently looking at the code in 'python_scripting.cpp' to better understand how KiCad deals with it, and also I'm looking at how to build Python 3.9+ so it can be used outside of MSYS2, and ideally put in KiCad's install directory, as it's normally done when you install it with the official installers, and which should not require playing with environment variables.

Still some work to do, but getting there.

[SiliconWizard]

The 3D model references all need to be reselected too, otherwise nothing will show up in the 3D view of the PCB, but that seems to be purely cosmetic. The symbols in the schematic can be updated in bulk with only a few clicks but I can't see how to update all 3D models in one go.

Yeah. After looking at this, it's because the name of the environment variables for the various paths, including 3D models, have changed. So any existing design made with an older version will suffer here.
I think they could have done something to ease the migration...

[thinkfat]

The 3D model references all need to be reselected too, otherwise nothing will show up in the 3D view of the PCB, but that seems to be purely cosmetic. The symbols in the schematic can be updated in bulk with only a few clicks but I can't see how to update all 3D models in one go.

Yeah. After looking at this, it's because the name of the environment variables for the various paths, including 3D models, have changed. So any existing design made with an older version will suffer here.
I think they could have done something to ease the migration...

You can fix that in bulk by just setting KISYS3DMOD to whatever KICAD6_3DMODEL_DIR points to.

[John B]

I added a new directory alias so all the KISYS3DMOD will work now. But am I right in thinking the old KiCad 5 .mod files wont work in 6? They are .wrl instead. The model doesn't work when I select the old models from KiCad 5. Some of them are missing in 6.

[tycz]

* I did manage to build a fully working KiCad for Windows 7, including Pcbnew which was the hardest part - this is the one that absolutely requires Python, the other KiCad executables can run without it.

Are parts of the program written in Python now, like what Eagle did with its scripting language? Or is it just the coder mentality of the devs shining through?

I never really came across a good use case for scripting in PCB CAD software. The only time I've heard of it being useful is for placing components (LED clocks are a good example). But that could be achieved by importing the component positions from a file, created in the user's favourite scripting language.

[eugene]

It sounds too easy, so there must be a reason it won't work, but how about going to www.python.org and downloading and installing 3.8 on your Win 7 machine.

[ataradov]

It sounds too easy, so there must be a reason it won't work, but how about going to www.python.org and downloading and installing 3.8 on your Win 7 machine.

That would not do anything to KiCad. They are using Python as a library for scripting. Specific version of Python is linked to the final binary, it does not use anything from the outside system.

[thinkfat]

* I did manage to build a fully working KiCad for Windows 7, including Pcbnew which was the hardest part - this is the one that absolutely requires Python, the other KiCad executables can run without it.

Are parts of the program written in Python now, like what Eagle did with its scripting language? Or is it just the coder mentality of the devs shining through?

I never really came across a good use case for scripting in PCB CAD software. The only time I've heard of it being useful is for placing components (LED clocks are a good example). But that could be achieved by importing the component positions from a file, created in the user's favourite scripting language.

Panelization, via stitching, track manipulation etc. There was a nice plugin for KiCAD 5 that did track rounding and solder mask removal for RF, I've used it frequently. If the scripting API is comprehensive, it is a great possibility to enhance the software with features that you need but are too specialized to be generally useful and are therefore unlikely to appear in any release.

[thinkfat]

I added a new directory alias so all the KISYS3DMOD will work now. But am I right in thinking the old KiCad 5 .mod files wont work in 6? They are .wrl instead. The model doesn't work when I select the old models from KiCad 5. Some of them are missing in 6.

.MOD files are old-style footprint files, not 3d models, no? Kicad still supports STEP and WRL for component models.

[woofy]

You can fix that in bulk by just setting KISYS3DMOD to whatever KICAD6_3DMODEL_DIR points to.

That has sorted it for me. Thanks.

[SiliconWizard]

So yes:

* For the 3D models, it's just a matter of adding the 'KISYS3DMOD' variable in the Path list. ("Configure Paths...") They changed a number of variable names for the paths - and prefixing them with 'KICAD6' tends to show they want to make it possible to "break" compatibility with future major versions, while forcing user to take action instead of possibly getting unintended results. It makes sense, although I prefer keeping backward compatibility as much as possible in software - but hey. Here, they are doing the same as most other EDA vendors do. Cadstar is notorious for breaking libraries at every new major version, requiring upgrading actions that are annoying - and sometimes require technical support (which they can sell... )

* Yes .step and .wrl are still supported. I don't think there is any new format used for 3D models anyway?

* For Python: yes scripting is a nice bonus. Eagle had that (although with their own language IIRC) a very long time ago already, and it was very useful. Now the choice of Python is questionable, but let's not get into that. Could be an endless debate. As to why Python decided not to support Windows 7 anymore here, just ask the team. (Just adding here, even if that's beyond the topic, that Windows 7 is STILL supported by MS. It has just become a paid support, but at least until 2023 or 2024. So in particular, some companies are bound to still use it and pay for the support. Thus, deciding to end being compatible with it is all the more questionable, and that's also for all commercial software that has ended supporting it...)

I managed to use Python 3.9.7 from MSYS2 (which works on Windows 7, but requires a lot of patches, that can be seen in the PKGBUILD file for Python - MSYS2 uses pacman, similar to Arch Linux). And, I found a way to "embed it" in KiCad's install directories to make it behave as intended while not requiring fiddling with environment variables. So now I have something fully usable. I also included the latest ngspice (36), which appears to work fine in the simulator too.

From the tests I've done so far, it seems to work perfectly fine. I'll do some more testing though, and will keep you posted. I'll probably make it accessible through a download at some point. If so, there won't be any installer (you'll have to extract it anywhere you like), and it won't come with libraries, which you'll have to download separately (the whole libraries take up 7 GB or so...)

[PlainName]

Excellent work, that man!

there won't be any installer (you'll have to extract it anywhere you like)

Might well be a bonus 

[SiliconWizard]

Oh, and I've seen that it now supports importing Cadstar and Altium? It's worth a shot, haven't tried it yet! Has anyone tried?

[Cerebus]

Oh, and I've seen that it now supports importing Cadstar and Altium? It's worth a shot, haven't tried it yet! Has anyone tried?

I haven't tried either of those, but I tried the Eagle import recently for a third party board that I had a very poor pinout guide for but could get the original Eagle files. It did a good job as far as it went (the Eagle schematic and Eagle PCB are still two separate, independent objects as far as KiCad is concerned but it served my, read-only, purposes). Last time I tried it on an older version it was a disaster.

[Bassman59]

The 3D model references all need to be reselected too, otherwise nothing will show up in the 3D view of the PCB, but that seems to be purely cosmetic. The symbols in the schematic can be updated in bulk with only a few clicks but I can't see how to update all 3D models in one go.

Yeah. After looking at this, it's because the name of the environment variables for the various paths, including 3D models, have changed. So any existing design made with an older version will suffer here.
I think they could have done something to ease the migration...

My libraries were all created before Kicad 6 (actually, some date from Kicad 4 and were moved forward to 5) so they still live (as a Subversion working copy) in the "old" standard directories. On the Mac, that was ~/Library/Application Support/kicad, though at some point they moved it to /Library/Application Support/kicad so the libraries were system-wide instead of per-user. I kept mine in the user Library.

Now with Kicad 6, the "standard" libraries provided by Kicad are in the application bundle, that is, /Applications/KiCad/KiCad.app/Contents/SharedSupport/ -- and the environment variables relevant to them are all prefixed with KICAD6_. But if you had previous environment variables, those were retained, so all of your libraries t hat refer to them do not break.

ANYWAY. The whole point of the library tables (fp-lib-table, sym-lib-table) is to map a nickname for a library to a .pretty (a footprint library) or .kicad_sym (a symbol library) somewhere in your file system. Since a library table is a per-system thing (on the Mac it's in ~/Library/Preferences/kicad) then as long as the library tables point to the actual libraries you really don't need the environment variables.

The oddball with not having EVs is that pointers to the 3D models are embedded in footprints. Now these pointers could be hard paths, which is fine if you actually standardize on file locations and you stick with just one OS. But a path on the Mac won't work in Windows, so you still need an EV to point to the base location of your 3D models. My footprints all call out the older KISYS3DMOD.  As long as KISYS3DMOD is listed in the "Configure Paths" setting, Kicad can find the models.

[SiliconWizard]

Oh, and I've seen that it now supports importing Cadstar and Altium? It's worth a shot, haven't tried it yet! Has anyone tried?

I haven't tried either of those, but I tried the Eagle import recently for a third party board that I had a very poor pinout guide for but could get the original Eagle files. It did a good job as far as it went (the Eagle schematic and Eagle PCB are still two separate, independent objects as far as KiCad is concerned but it served my, read-only, purposes). Last time I tried it on an older version it was a disaster.

Eagle import was a bit shaky in v5 - often required a lot of fixing. If it has improved, all good.

I do have a number of CadStar projects, as we were using that up to a few years ago, but I currently don't have a CadStar license... and unfortunately, all the project files I have are in binary format, whereas KiCad import only handles the ASCII format. You can export to that format from CadStar, but a license is needed... =) So, at this point, no luck here.

I tried importing a couple Altium schematics, and it appears to work reasonably well!

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

[Cerebus]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Somebody has done an independent add-on, but more oriented to presentation elements than replacing the Hershey fonts; native support for "any font, not just Hershey" is slated for V7.

[delfinom]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Somebody has done an independent add-on, but more oriented to presentation elements than replacing the Hershey fonts; native support for "any font, not just Hershey" is slated for V7.

Custom fonts already added to the latest 7.0 nightlies. Only 3d viewer support is lacking and some bug fixes over time needed.

[delfinom]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Go into the Preferences and udner Dispaly Options, turn on Anti-aliasing. They are off by default, I think it's ridiculous and I got some pushback on making it default at some point. The Anti-aliasing was fixed in v6 a bit with some mathematical errors in the shaders corrected that make it funky before.

Otherwise, even with the recent custom font support, those custom fonts look just as annoying without anti aliasing. Why? Because we need to translate fonts into lines and polygons which are the graphic primitives behind the rest of the system from ERC for schematics and DRC for PCBs. We can't just do SVG style rendering with these. Everything is meant to be accurately translatable between what you see and what gets printed or made into a PCB.


And about Windows 7:

For the record, I couldn't even start a new Windows 7 VM months ago to potentially be nice here with support. Microsoft has dropped all the Windows update servers for Windows 7 so I can't even obtain the Hyper-V drivers to install into the VM to get the correct behaviors.  They also nuked all downloads from their web pages for Windows 7.

[PlainName]

Microsoft has dropped all the Windows update servers for Windows 7 [...] They also nuked all downloads from their web pages for Windows 7

Yeah, all the tricks they used against the likes of Netscape, DrDOS, Go, Caldera, Stac, the multiple billion dollar fines, etc, are now turned towards its own customers in order to save face with The Interface Formerly Known As Metro. IMO you'd be doing the world a great favour if you made W7 work and dropped support (actively, of course) for W10+

[SiliconWizard]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Somebody has done an independent add-on, but more oriented to presentation elements than replacing the Hershey fonts; native support for "any font, not just Hershey" is slated for V7.

Custom fonts already added to the latest 7.0 nightlies.

Have you tried it? And if so, could you maybe post a screenshot of a schematic using a custom font, so we can see how it renders?

[ataradov]

I tried it. Here is an assortment of random fonts. Currently there seems to be a bug where only standard font gets selection highlight outline. So it it impossible to see selected items.

Also, no anti-aliasing, since those labels disappear with AA, at least in my VM, which does not have ideal 3D acceleration.

[SiliconWizard]

Thanks. So for now, it doesn't really help readability and kinda looks like shit. I'm not considering the bugs - of course there will be for a while. But I'm afraid the rendering aspect itself may never really improve unless they make a major modification... (I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

I understand that the work is tougher for PCB layout (because ultimately it needs to be rendered as just basic vector graphics), but for schematics? Couldn't they just have added text rendering in the schematic editor using *text rendering* and not reusing the underlying graphics engine, which kind of bites? That may not seem a big deal to some, but to me that's a real issue. (To be fair, Eagle schematics look even more horrible - but pretty much all other major EDA manage to do this properly...)

[ataradov]

I would not expect those fonts to be great even when they are done. All it will do is make random schematics look goofy because some people unironically use comic sans.  I personally would prefer improvements in the standard vector font. I feel like Eagle one is somewhat cleaner, although I really don't care. The one that is here now works for me.

One thing you can do is similar to Altium where there is a differentiation between purely graphics items and layout items. You can do Bezier curves in silk screen, but not in copper. And DRC for silk screen does not need to be all that accurate.

Also one thing I noticed is that size control is harder with custom fonts. The same size in points results in different visual heights. So "C6" and "22 uF" have the same size, I only changed the font name. Yet they look very different. So alignment with other schematic elements may be an issue. Probably less of an issue if you stick with one font. "TP6" and "OUT" were on the same level in the original font, but here they are shifted with two different fonts. Edit: Although this one is just the optical effect because of different sizes, baseline is the same.

[ebastler]

(I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

I did not realize schematics entry and PCB layout had a common legacy in KiCad. To me they alwas felt like they were originally developed by two entirely separate teams?

[ataradov]

I think they are separate to a large extent, but apart from enabling AA, I don't know if this will improve much no matter what. This is very early days, so all this has a chance for improvement, of course.

And judicial use of fonts may make things look ok.

[SiliconWizard]

(I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

I did not realize schematics entry and PCB layout had a common legacy in KiCad. To me they alwas felt like they were originally developed by two entirely separate teams?

What teams? The original KiCad was essentially developed by one guy.

And, I didn't take a deep look at the source code - especially recently - but this would be my guess that the graphics engine for drawing schematics and PCBs is essentially the same, or at least with a large chunk of common stuff.

[ebastler]

What teams? The original KiCad was essentially developed by one guy.

Thanks, I had not realized. I have been assuming all along that eeschema and pcbnew came from different origins since their user interfaces feel so different. (Different command letters for equivalent commands, different logic for commands with/without prior selection etc.).

Is a "history of KiCad" document available anywhere, or interviews with the original creator or early contributors? I'd be curious to learn more but did not find much during a quick search.

[ataradov]

And I think I need to clarify, Test Label Text and LBMxxxx are written in goofy fonts, font rendering is not that bad The jagged lines from no AA or subpixel rendering is all there though.

[SiliconWizard]

What teams? The original KiCad was essentially developed by one guy.

Thanks, I had not realized. I have been assuming all along that eeschema and pcbnew came from different origins since their user interfaces feel so different. (Different command letters for equivalent commands, different logic for commands with/without prior selection etc.).

The real divergence is relatively "new" - comes from the overhaul KiCad got when the CERN invested in it. Yes, it's been years now, but there was a KiCad before the CERN.
They have also focused more on the layout editor, probably because it was what would really make KiCad take off and stand out.

And you can imagine, even though they have put a lot of work in it, that there still is a log of legacy code and data structures.

A very short history can be found here: https://en.wikipedia.org/wiki/KiCad
Basically, that was the work of one university professor for years (with occasional contributions) before the CERN invested in it.

[SiliconWizard]

And I think I need to clarify, Test Label Text and LBMxxxx are written in goofy fonts, font rendering is not that bad The jagged lines from no AA or subpixel rendering is all there though.

I personally find this relatively ugly and hard to read - again especially at lower zoom levels. But sure I am pretty hard to please when it comes to GUI in general, and editors in particular.

Certainly, if you're one of those who can't tell that Eagle schematics look terrible (in particular for the lack of antialiasing, and questionable symbols), I can understand it's not a problem for you.

If you want to see what decent schematic rendering looks like - Altium is a good start.

Interestingly, one of the best looking schematic editor is the humble sPlan - unfortunately, it is way, way less featureful otherwise, and doesn't export to any usable format other than graphics, at least that I know of. But that gives an idea about what can be achieved with limited means. That's close to publishing-quality schematics, but I personally see no reason a decent schematic editor should be any less than this. Just MHO of course. But just so you know what I mean.

[ataradov]

Certainly, if you're one of those who can't tell that Eagle schematics look terrible (in particular for the lack of antialiasing, and questionable symbols), I can understand it's not a problem for you.

I like how Eagle schematics look like, so obviously it is all a matter of personal preference.

If you want to see what decent schematic rendering looks like - Altium is a good start.

I've seen plenty of shitty schematics done in Altium. That's  where my concern with allowing custom fonts comes from.

I do like to see improvements in standard fonts, of course. By no means they are ideal. But I would not call them horrible either.

But also, I hate AA in general. If there are pixels, I want to see them. All my system fonts set to no AA or blurring of any kind.

[delfinom]

I tried it. Here is an assortment of random fonts. Currently there seems to be a bug where only standard font gets selection highlight outline. So it it impossible to see selected items.

Also, no anti-aliasing, since those labels disappear with AA, at least in my VM, which does not have ideal 3D acceleration.

* Unfinished implementation

It is the nightly after all aka unreleased interim build. It'll take some time for it to get ironed out. The old concepts of "single fonts" needs to be dissected and replaced in various places with the new abstraction.

[delfinom]

Also one thing I noticed is that size control is harder with custom fonts. The same size in points results in different visual heights. So "C6" and "22 uF" have the same size, I only changed the font name. Yet they look very different. So alignment with other schematic elements may be an issue. Probably less of an issue if you stick with one font. "TP6" and "OUT" were on the same level in the original font, but here they are shifted with two different fonts. Edit: Although this one is just the optical effect because of different sizes, baseline is the same.

Editing properties of say, all the text on a PCB to the same font and size should get easier later in v7. We are moving to a properties pane ala Altium for everything

[delfinom]

Thanks. So for now, it doesn't really help readability and kinda looks like shit. I'm not considering the bugs - of course there will be for a while. But I'm afraid the rendering aspect itself may never really improve unless they make a major modification... (I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

It comes down to fonts.
Serif fonts will always look awful. Standard San Serif fonts your mileage will vary. What you really want is to use Monospace programmer fonts.

Attached are the "KiCad Font" vs. the "Cascadia Code" microsoft font for programming compared on the same schematic section

[3roomlab]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Somebody has done an independent add-on, but more oriented to presentation elements than replacing the Hershey fonts; native support for "any font, not just Hershey" is slated for V7.

Custom fonts already added to the latest 7.0 nightlies. Only 3d viewer support is lacking and some bug fixes over time needed.

any idea about inclusion of korean/japanese fonts ?

[delfinom]

Now just a note regarding the schematic editor: while it got a few new nice features, there's something that is, IMO, still severely lacking: the support of TrueType fonts. The default font is OK(-ish) for PCBs, although I would also like being able to select different fonts, but for schematics, it's really annoying. It doesn't look clean on schematics, especially on screen, and even makes it hard to read at lower zoom levels. I know this was one feature listed for v6 actually - and I'm disappointed it didn't end up in it.

Somebody has done an independent add-on, but more oriented to presentation elements than replacing the Hershey fonts; native support for "any font, not just Hershey" is slated for V7.

Custom fonts already added to the latest 7.0 nightlies. Only 3d viewer support is lacking and some bug fixes over time needed.

any idea about inclusion of korean/japanese fonts ?

Just use a unicode font?

[SiliconWizard]

Thanks delfinom, that definitely looks better than what I feared =)

[nctnico]

(I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

I did not realize schematics entry and PCB layout had a common legacy in KiCad. To me they alwas felt like they were originally developed by two entirely separate teams?

What teams? The original KiCad was essentially developed by one guy.

And, I didn't take a deep look at the source code - especially recently - but this would be my guess that the graphics engine for drawing schematics and PCBs is essentially the same, or at least with a large chunk of common stuff.

Likely none of the original code exists today  As part of a small team I've been working on an inherited project as well for a couple of years and by now I think we have touched about 95% of the original code.

[Cerebus]

Hmm, looks like the macOS builds of the nightlies have been failing for the last 19 days. I wanted to grab the latest 6.99 to look at the font stuff and the cupboard was empty - no nightlies at all. The font stuff hadn't made it into the nightlies for the last macOS build of 6.99 that was available before xmas (and which is now missing too).

[delfinom]

Hmm, looks like the macOS builds of the nightlies have been failing for the last 19 days. I wanted to grab the latest 6.99 to look at the font stuff and the cupboard was empty - no nightlies at all. The font stuff hadn't made it into the nightlies for the last macOS build of 6.99 that was available before xmas (and which is now missing too).

macOS unforunately is a one man operation and he is busy. And Apple keeps changing stuff as they slowly lock down macOS into iOS

[Bassman59]

(I get it that the graphics engine is probably the same for schematics and layout, which IMO is a mistake, but that's probably KiCad's long legacy.)

I did not realize schematics entry and PCB layout had a common legacy in KiCad. To me they alwas felt like they were originally developed by two entirely separate teams?

What teams? The original KiCad was essentially developed by one guy.

And, I didn't take a deep look at the source code - especially recently - but this would be my guess that the graphics engine for drawing schematics and PCBs is essentially the same, or at least with a large chunk of common stuff.

Actually -- originally schematic and PCB were two entirely separate programs. None of the features we take for granted (like clicking a footprint and having the schematic zoom to the corresponding symbol) were possible.

Work on PCB moved forward to take advantage of things like OpenGL and better rendering and speed and such. Schematic was not modernized as such. The two remained separate, with different rendering and all. This was the case with Kicad 4 and earlier "releases."

Kicad 5 was the first release that moved Schematic forward to OpenGL and put it on par with PCB. This was a development goal: make schematic and PCB work the same and work well together, like the Kiway messaging between the two so that selection feature I mentioned would work. (I think that existed prior to 5, I don't remember.)

Kicad 6 is much more evolved -- part symbols are finally embedded in the schematic instead of being referenced to the originating libraries, for example. Using 6 now I see that from the user's perspective it really is one program, with two views of the design (schematic, layout).

Thing is: this is all what the users have wanted since forever (other than a different library system, on which there is never going to be an agreement about "what is best."). The biggest complaints were that schematic and PCB were separate and obviously so!

Anyway.

[delfinom]

Depending on how things go. The standalone executables for schematic and pcb may be gone on v7 if you want to talk about complete merge of the two tools

[SiliconWizard]

Anyway, point was not about the tools being separate executables, but the fact they probably share a significant chunk of code, in particular for the graphics rendering. They may have diverged a bit for the OpenGL backend while eeschema didn't have it yet, but otherwise there's probably a lot in common. delfinom can confirm that.

And, if you want to merge the two in one executable, why not. But I liked being able to edit a layout without a schematic attached (but just its netlist), so as long as you keep that possibility, I'm fine with whatever.

[Bassman59]

Anyway, point was not about the tools being separate executables, but the fact they probably share a significant chunk of code, in particular for the graphics rendering. They may have diverged a bit for the OpenGL backend while eeschema didn't have it yet, but otherwise there's probably a lot in common. delfinom can confirm that.

Please re-read what I wrote.

In the beginning there was no code shared between PCB and schematic. Kicad 5 was the first version where there was a lot of sharing, because that was when the schematic side had OpenGL. Kicad 6 is "unified," if you will.

[rockola]

Couldn't they just have added text rendering in the schematic editor using *text rendering* and not reusing the underlying graphics engine, which kind of bites?

"They" did not do it because the code to use the underlying engine can be used for Gerbers, which was originally the whole point: I implemented outline fonts for KiCad because I wanted to have nicer looking silkscreens.



I agree that pixel-based text rendering would look nicer in the schematic editor, and am looking forward to someone else implementing it and submitting a merge request. KiCad 7 comes out in a year, so hopefully that someone will get a round tuit.

The examples above look jagged because antialiasing was off.

[ebastler]

As a side note, I recommend against using "Comic Sans". Ever. For anything. 

[RoGeorge]

For a newcomer, what would be the right version to start learning KiCad, v5 or v6?

[PlainName]

As a side note, I recommend against using "Comic Sans". Ever. For anything. 

Why?

[Cerebus]

As a side note, I recommend against using "Comic Sans". Ever. For anything. 

HOW PRAY WOULD ONE WRITE A STRONGLY WORDED LETTER ABOUT A SUBJECT IMPORTANT TO THE MORAL FIBRE OF THE NATION TO THE EDITOR OF A NATIONAL NEWSPAPER WITHOUT USING COMIC SANS

 

[ataradov]

For a newcomer, what would be the right version to start learning KiCad, v5 or v6?

V6, of course. Why learn on an old version?

[thinkfat]

For a newcomer, what would be the right version to start learning KiCad, v5 or v6?

That's a good question. On one hand, if you have no legacy with v5, v6 would seemingly be the appropriate choice. On the other hand, if you're an absolute beginner to electronics design and ECAD, all the tutorials out there are based on v5. While the two versions are not fundamentally different and share many usage paradigms, some people might lost in the differences. So, if you foresee having troubles if you cannot follow a tutorial to the letter, better start with v5.

[ebastler]

As a side note, I recommend against using "Comic Sans". Ever. For anything. 

Why?

[rant]
Comic Sans was a forced attempt at "being informal" already when it came out in the 90s, in my opinion. It has been overused ad nauseam in informal contexts, and has for some absurd reason even been (ab)used in professional contexts. Just don't!

OK, I'll grant you an exception for use in primary school handouts. But that's the only acceptable use!
[/rant]

And yes, that's totally subjective and a matter of personal taste, of course. 

[SiliconWizard]

Anyway, point was not about the tools being separate executables, but the fact they probably share a significant chunk of code, in particular for the graphics rendering. They may have diverged a bit for the OpenGL backend while eeschema didn't have it yet, but otherwise there's probably a lot in common. delfinom can confirm that.

Please re-read what I wrote.

In the beginning there was no code shared between PCB and schematic.

Oh really. Do you have proof of that? And are you saying that the original author of KiCad was so stupid that he never reused any part of his code in both tools?
Are you in particular (because this was my point) saying that no code at all was ever shared for the font rendering, while it was the same font and looked exactly the same in both? Uh?

[SiliconWizard]

Couldn't they just have added text rendering in the schematic editor using *text rendering* and not reusing the underlying graphics engine, which kind of bites?

"They" did not do it because the code to use the underlying engine can be used for Gerbers, which was originally the whole point

This has been exactly my point all along inded, despite some people claiming that there was absolutely no code shared between all those tools. Of course that was the reason. All I'm saying is that it looks bad on screen in general, because text rendering is difficult and is best done with appropriate approaches and not just naive rendering.

Since there is a relatively long-standing demand for truetype fonts in KiCad (and they are working on it), I suppose I'm not the only one.

I gave sPlan as an example of what I consider proper-looking schematics. As I said, I see no reason why we should settle for less in a schematic editor. But of course, it's all mostly a matter of legacy for existing EDA software, so it's definitely not an easy task for KiCad developers.

[Munyua44]

i have been using the kicad 5.1 and it has very poor simulation features. Am hoping that Kicad 6 has improved on the simulation features such tat we do not need to look for another external source of simulation.
Also hope the autorouting process is back so that it can be very helpful when training our beginner students.

[woofy]

There is still a lot to do in kicad and I'd like to see pin/gate swapping with back propagation to the schematic, so I hope they do not waste time on an auto-router. Although I always manually route my boards I have no objection to the presence of an auto-router. It's just that I have never seen a worthwhile one. My first was Bartels Auto Engineer back in 386/486 DOS days. It was one of the best auto-routers around at the time but the results were awful. Later Proteus gained an auto-router and again the results were (and still are) awful. So nice as it sounds, I think there are other features for the core kicad team to prioritise.

[PlainName]

I've never seen a decent autorouter either, and always manually route boards. However...

I worked at a client producing reasonably complex PCBs (essentially a PC motherboard with the usual parts memory, SATA ports, video, etc) and quite often the chap designing the PCBs would modify the schematic in the morning then let the autorouter loose over lunch and farm out the gerbers in the afternoon. The results were, to put it politely, a mess with tracks looping back on themselves and every kind of horror you might hear about autorouting being present somewhere on the boards. Nevertheless, the boards never failed to work (at least, when they did fail it was either a design or software issue).

It is very hard to argue against results (and my colleague at the time joined me in trying!)

[madires]

There's no magic autoroute-the-way-I-would-do-it button! To get acceptable results one has to play a lot with the autorouter settings and rules, and has to learn how the specific autorouter reacts/works. In most cases that investment makes only sense if you regularly route complex PCBs.

[thinkfat]

I've tried freerouter in the past with KiCAD and spend quite some time tinkering with costs for vias and traces, orientation, netclasses and stuff. The results were always so-so, but at least not catastrophic. And at least once the router showed me potential routing paths I hadn't thought about. I took the advice and routed the board manually.

Then I tried one of those "deep learning" autorouters on the web. It was a bit disappointing given all the fanfare by its creators.

[Cerebus]

There is still a lot to do in kicad and I'd like to see pin/gate swapping with back propagation to the schematic, so I hope they do not waste time on an auto-router.

That's one of the things very high on my wish list, the omission on the PCB while still having everything ready for gate swaps on the schematic side is one of those puzzling things - a bit like finding a shoe factory that only has equipment for making left shoes. "But where's the other half?".

Also high on the wish list is the analogous facility for back annotating bus pins. I'm in the middle of a board with an FPGA on it and deliberately left all the FPGA pins tied to a bus but individually unassigned because of course I can assign the pin functions inside the FPGA to suit the easiest routing, and manually back annotating all the functional pins on an FPGA one-by-one is/was a slow, painful and error prone process. Pin swaps between identical functional groups on a microprocessor is a similar, if slightly more involved wish list item too.

I'd rather see an auto-place algorithm that wasn't completely useless first before any effort was wasted expended on auto-route. The current parts placement, both on simple 'import changes' and auto-place is generally worse than if it just preserved the same relationship between components that exists in the schematic with just enough room to read the labels. In fact an 'import changes' that left room to read annotations would be a big leap forward.

[thinkfat]

An autoplace that minimized the length and crossing points of ratsnest lines would be a nice addition. If that could be combined with gate and pin swapping - true bliss. If would be really helpful if GPIO pins could be swapped automatically, or at least manually during routing.

[nfmax]

Placement is nine tenths of routing

[thinkfat]

i have been using the kicad 5.1 and it has very poor simulation features. Am hoping that Kicad 6 has improved on the simulation features such tat we do not need to look for another external source of simulation.
Also hope the autorouting process is back so that it can be very helpful when training our beginner students.

I don't think much has changed there. The simulation engine is still ngspice and you still have to hunt high and low for spice or xspice models for your components.

Autorouting is not coming back.

[Bassman59]

Anyway, point was not about the tools being separate executables, but the fact they probably share a significant chunk of code, in particular for the graphics rendering. They may have diverged a bit for the OpenGL backend while eeschema didn't have it yet, but otherwise there's probably a lot in common. delfinom can confirm that.

Please re-read what I wrote.

In the beginning there was no code shared between PCB and schematic.

Oh really. Do you have proof of that? And are you saying that the original author of KiCad was so stupid that he never reused any part of his code in both tools?
Are you in particular (because this was my point) saying that no code at all was ever shared for the font rendering, while it was the same font and looked exactly the same in both? Uh?

The only "proof" I have is that I've been using Kicad since before the v4 was released. The programs were different, with different code bases. I thought I had old source tarballs somewhere but I guess I finally got rid of them.

Really, Jean-Pierre Charras, whom I don't think is stupid, started it back in the early 90s as separate programs, I think the schematic was first. The PCB went through various changes -- it was called "PCBNEW" until recently because it wasn't even the first attempt at a layout tool. There were other developers involved, too. The path towards integrating the disparate parts into a single program was really started seriously when Wayne Stambaugh took over as project lead, and it really did take quite a while to unify everything.

Now you don't have to believe me. Maybe you can ping Jean-Pierre or Wayne on the Kicad developers' email list and ask them yourself.

Can I ask you a question: how long have you been using Kicad?

[delfinom]

i have been using the kicad 5.1 and it has very poor simulation features. Am hoping that Kicad 6 has improved on the simulation features such tat we do not need to look for another external source of simulation.
Also hope the autorouting process is back so that it can be very helpful when training our beginner students.

I don't think much has changed there. The simulation engine is still ngspice and you still have to hunt high and low for spice or xspice models for your components.

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

[JohnG]

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

Even Altium has switched to ngspice: https://resources.altium.com/p/a-first-look-at-altium-designer-20. I hope some of these companies adopting ngspice throw in some resources to the developers.

Very glad to hear that there are big plans for including simulation in KiCad. I hope they focus on smoothing the workflow and output graphs first. I will keep an eye out.

John

John

[nctnico]

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

Even Altium has switched to ngspice: https://resources.altium.com/p/a-first-look-at-altium-designer-20. I hope some of these companies adopting ngspice throw in some resources to the developers.

Very glad to hear that there are big plans for including simulation in KiCad. I hope they focus on smoothing the workflow and output graphs first. I will keep an eye out.

Well, being able to use a component database should come first. That is the most important feature (that is still missing) to be able to use KiCad professionally for real.

[Uky]

Had a problem with a Gerber file yesterday that would not print correctly with the Pentalogix ViewMate. Downloaded KiCad 6.0.1 and imported the file.
(Metric 4.3) While the display canvas showed the file correctly - the print engine with the preview lacked several traces.

 

[JohnG]

Well, being able to use a component database should come first. That is the most important feature (that is still missing) to be able to use KiCad professionally for real.

I agree 100%.

[delfinom]

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

Even Altium has switched to ngspice: https://resources.altium.com/p/a-first-look-at-altium-designer-20. I hope some of these companies adopting ngspice throw in some resources to the developers.

Very glad to hear that there are big plans for including simulation in KiCad. I hope they focus on smoothing the workflow and output graphs first. I will keep an eye out.

Well, being able to use a component database should come first. That is the most important feature (that is still missing) to be able to use KiCad professionally for real.

The few people planning to work on the simulator are also ones that have no interest in the component database.
There is someone else throwing around that feature for v7, whether it makes it in or not is a different story


We are a open source project, nobody is forced to work on things they don't have interest in. Unlike a job where you have to suck it up and do it 

[Uky]

IMHO: Having a decent interface that can import or export things are the most valuable feature of any CAE software.
Adding propertys to a schematic drawing and then being able to export a netlist to an external tool such as a Spice
simulator would take some efforts out of having to write such a software within a suite.

Importing propertys from external sources as well. It would perhaps be better if KiCAD made it possible to connect
to an external database software. There are several free versions of eg. SQL than to create a proprietary system.

Flexibility is worth a lot!

[nctnico]

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

Even Altium has switched to ngspice: https://resources.altium.com/p/a-first-look-at-altium-designer-20. I hope some of these companies adopting ngspice throw in some resources to the developers.

Very glad to hear that there are big plans for including simulation in KiCad. I hope they focus on smoothing the workflow and output graphs first. I will keep an eye out.

Well, being able to use a component database should come first. That is the most important feature (that is still missing) to be able to use KiCad professionally for real.

The few people planning to work on the simulator are also ones that have no interest in the component database.
There is someone else throwing around that feature for v7, whether it makes it in or not is a different story


We are a open source project, nobody is forced to work on things they don't have interest in. Unlike a job where you have to suck it up and do it 

I get that but having a database driven component management system would propel Kicad up to the level where Altium and Orcad are. Simulation is kinda meh. With free software like Ltspice and Microcap being around, simulation is pretty well served already.

You are not going to use one schematic for both a circuit board and simulation anyway because simulation often needs the use of dummy/extra components. Also having excess components slows simulation down so typically you only simulate parts of a circuit. IOW: don't fall into the trap thinking that having simulation integrated saves time for drawing a schematic twice; it doesn't.

[Bassman59]

ngspice is fine, even EAGLE/autodesk have integrated it...not that it's saying anything but hey, it works and there's not much alternative because of how extremely niche and high level mathematics the simulators are.
v6 fixes some bugs and tweak but it's not much of an change over v5 I think.

However, there is now a maintainer for the simulator and he has big plans for v7, and potentially alot more awesome goodies

Even Altium has switched to ngspice: https://resources.altium.com/p/a-first-look-at-altium-designer-20. I hope some of these companies adopting ngspice throw in some resources to the developers.

Very glad to hear that there are big plans for including simulation in KiCad. I hope they focus on smoothing the workflow and output graphs first. I will keep an eye out.

Well, being able to use a component database should come first. That is the most important feature (that is still missing) to be able to use KiCad professionally for real.

The few people planning to work on the simulator are also ones that have no interest in the component database.
There is someone else throwing around that feature for v7, whether it makes it in or not is a different story


We are a open source project, nobody is forced to work on things they don't have interest in. Unlike a job where you have to suck it up and do it 

I get that but having a database driven component management system would propel Kicad up to the level where Altium and Orcad are. Simulation is kinda meh. With free software like Ltspice and Microcap being around, simulation is pretty well served already.

You are not going to use one schematic for both a circuit board and simulation anyway because simulation often needs the use of dummy/extra components. Also having excess components slows simulation down so typically you only simulate parts of a circuit. IOW: don't fall into the trap thinking that having simulation integrated saves time for drawing a schematic twice; it doesn't.

I agree in principle that a database-driven library system would be a worthwhile addition to the Kicad feature set.

But out of everyone who says "Kicad needs database libraries!" zero of them have been willing to step up and write a description of what that actually means and offer a design proposal for discussion.

And since none of the database enthusiasts have done so, the developers have continued fixing and refining the program's main functionality.

Is anyone willing to start that discussion with the developers on the Gitlab site or the mailing list, or will it just remain on forums?

[nctnico]

As the reply of delfinom already indicates; the developers of Kicad seem to be unaware of the huge value of having a database driven component system. Love has to come from both sides...

A couple of years ago I have seriously considered of shelling out several k euro for someone to add a database driven component system to Kicad. However my (bad) experience with contributing to other open source projects has made me decide not to go this route. If there is no love from the core team for a feature then it won't be maintained and it is likely to be abandoned in future versions. So I spend the money at the local Orcad distributor.

AFAIK there is a specification for what a database driven component system for Kicad should look like and IIRC it looks pretty complete. And it is not like this is something new; I have been using Orcad with a database driven component system for 20+ years. There is tons of information from Altium and Orcad on how such a system should work. Copy & paste

[John B]

In KiCad 5, I used a plugin to duplicate PCB footprint layouts based on hierarchical sheets:

https://github.com/MitjaNemec/Kicad_action_plugins

Are there any other options with KiCad 6, or just hang tight till there's some plugins available?

[delfinom]

As the reply of delfinom already indicates; the developers of Kicad seem to be unaware of the huge value of having a database driven component system. Love has to come from both sides...

A couple of years ago I have seriously considered of shelling out several k euro for someone to add a database driven component system to Kicad. However my (bad) experience with contributing to other open source projects has made me decide not to go this route. If there is no love from the core team for a feature then it won't be maintained and it is likely to be abandoned in future versions. So I spend the money at the local Orcad distributor.

AFAIK there is a specification for what a database driven component system for Kicad should look like and IIRC it looks pretty complete. And it is not like this is something new; I have been using Orcad with a database driven component system for 20+ years. There is tons of information from Altium and Orcad on how such a system should work. Copy & paste

Again, if you actually read my post. There is already somebody planning to work on it.
There are others planning work on the simulator.

You don't have a carpenter do your plumbing. The same thing applies here. Otherwise, they both can use hand tools, so why can't they do either job   The only thing I can otherwise say is you are just being disrespectful of programmers if you are trying to say "why don't they all work on this to serve me", abstract organizational tooling is the same as linear algebra in spice right?

Is anyone willing to start that discussion with the developers on the Gitlab site or the mailing list, or will it just remain on forums?

I wouldn't worry about it. It's been brought up on the mailing list before and it usually dies off for the same reason, people who make no code contribution argue in endless circles. Design by committee almost never works or else you just get USB 3.2 2x2 Extreme Edition. The gitlab issue is open here: https://gitlab.com/kicad/code/kicad/-/issues/7436

Since the Jon assigned it to himself, it'll probably get done for v7

[PlainName]

you are just being disrespectful of programmers if you are trying to say "why don't they all work on this to serve me"

But that isn't what they are saying. They are actually saying that if you want the product to appeal to the end user, and be treated as a grown-up product, it needs to do such and such. Being open source and all that, the developers are free to do whatever their desire dictates, but then they shouldn't be surprised if the end result doesn't appeal too much.

What if one is not a programmer, or not a programmer of desktop programs, or not a programmer of the language du jour? Are suggestions/requirements automatically void because you can't implement them yourself? How about if you've contributed with funds - does that make you more or less worthy than someone who's contributed a lot of code that wasn't up to scratch and got rejected?

[nctnico]

you are just being disrespectful of programmers if you are trying to say "why don't they all work on this to serve me"

But that isn't what they are saying. They are actually saying that if you want the product to appeal to the end user, and be treated as a grown-up product, it needs to do such and such. Being open source and all that, the developers are free to do whatever their desire dictates, but then they shouldn't be surprised if the end result doesn't appeal too much.

Precisely! As an outsider looking in I'm simply stating what is missing. It is up to the developers to pick it up or not.

From https://www.kicad.org/about/kicad/
Mission Statement
The goal of the KiCad project is to provide the best possible cross platform electronics design application for professional electronics designers. Every effort is made to hide the complexity of advanced design features so that KiCad remains approachable by new and inexperienced users, but when determining the direction of the project and the priority of new features, the needs of professional users take precedence.

Professional users need a database driven component system so why hasn't this been implemented yet? And why get hostile towards people pointing that out?  The so typical  'do it yourself if you want this feature' response is getting a bit old.

[Cerebus]

Professional users need a database driven component system so why hasn't this been implemented yet? And why get hostile towards people pointing that out? 

This seems to be the KiCad ethos, at least as put forward by KiCad people I have seen commenting on the EEVBlog forum. There seems to be a general antipathy from them towards experienced electronics professionals remarking on what an electronics CAD tool needs. The usual final retort is "well if you want that why don't you contribute" where contribute means "write code" completely ignoring that contributing requirements and insights is also contributing.

The problem with these kind of projects is that people work on what they want to, what interests them, rather than what is needed. That's the nature of the beast and it would take much cannier project management than you ever see to steer the ship in the direction of doing "what's needed" rather than "what Johnny, who can do this, feels like playing with". Case in point, I see "linear algebra" mentioned above. That suggests to me that someone who wants to write their own circuit simulator from scratch is working on the simulation side of things, because you need to know next to nothing about linear algebra to improve the integration with existing simulation tools such as ngspice.

[sethhillbrand]

This seems to be the KiCad ethos, at least as put forward by KiCad people I have seen commenting on the EEVBlog forum. There seems to be a general antipathy from them towards experienced electronics professionals remarking on what an electronics CAD tool needs. The usual final retort is "well if you want that why don't you contribute" where contribute means "write code" completely ignoring that contributing requirements and insights is also contributing.

Just a quick note as a "KiCad person".  No lead developer of KiCad would ever take this approach.  We actively seek out experienced users and mine them for ideas about what would help their workflows.  We maintain a contact sheet with large design houses to identify where their pain points are when using KiCad and actively work to remove these.

Now, KiCad users posting on eevblog forums are another story.  They are welcome to broadcast any sort of provocative talk they like.  That doesn't represent the KiCad project.  If you are a professional CAD designer/engineer, we would welcome your input.  This doesn't happen on forum postings, though.  You can post issues on GitLab and we can arrange offline contact options as well to facilitate the sort of information exchange that we need to improve the system for everyone.

[RoGeorge]

This seems to be the KiCad ethos, at least as put forward by KiCad people I have seen commenting on the EEVBlog forum. There seems to be a general antipathy from them towards experienced electronics professionals remarking on what an electronics CAD tool needs. The usual final retort is "well if you want that why don't you contribute" where contribute means "write code" completely ignoring that contributing requirements and insights is also contributing.

Just a quick note as a "KiCad person".  No lead developer of KiCad would ever take this approach.  We actively seek out experienced users and mine them for ideas about what would help their workflows.  We maintain a contact sheet with large design houses to identify where their pain points are when using KiCad and actively work to remove these.

Now, KiCad users posting on eevblog forums are another story.  They are welcome to broadcast any sort of provocative talk they like.  That doesn't represent the KiCad project.  If you are a professional CAD designer/engineer, we would welcome your input.  This doesn't happen on forum postings, though.  You can post issues on GitLab and we can arrange offline contact options as well to facilitate the sort of information exchange that we need to improve the system for everyone.

That's a very complicated way to say "KiCad devs don't give a dead rat about what you all are talking here", because forums are usually lame, therefore you all must be noobs talking bollocks.

[fourfathom]

If you are a professional CAD designer/engineer, we would welcome your input.

OK.  Since this seems to be how KiCad wants to plan future enhancements, does anyone here (EEVblog) want to engage on the component database issue, via GitLab or elsewhere? 

Not me though.  I'm a retired EE and we had CAD people to do the boards and component people to manage the database, so I don't know any of the details.  But I do know that I would like to be able to integrate my personal component database into my KiCad design flow.  I would be happy to change out my database if that's what it took.

[Cerebus]

That's a very complicated way to say "KiCad devs don't give a dead rat about what you all are talking here", because forums are usually lame, therefore you all must be noobs talking bollocks.

That's a rather jaundiced view, but given the tenor of one of the other 9 contributions in total that sethhillbrand has made here I'm not sure that it's necessarily unjustified.

You know I would pay for Kicad if the paid option was for a useable interace.

OK.

https://giving.web.cern.ch/civicrm/contribute/transact?reset=1&id=6

That glib quip isn't what I'll call a highly valuable contribution from someone who presents themselves as in a position to influence KiCad's direction, especially when it makes up 11% of someone's posting to date.

The EEVBlog forum probably presents the highest concentration of EEs online. Unlike some other electronics fora it's not loaded with incentives to gain points or popularity which can heavily influence who makes contributions and why. Here you only speak up if you think you've something to contribute, and don't get a little dopamine rush of contributor points for doing so. I think that makes this a very valuable place to gather opinions, ideas, and use as a sounding board. If it was me, and I was genuinely looking for input I'd try and find a way to leverage that, rather than direct people towards vaguely defined routes to contribute and hint at privileged access to "large design houses".

I find it all rather unconvincing. People all over the place have been bitching, justifiably, about the KiCad UI for ages. In 6.0 they've cleaned a lot up, but tab ordering is still random and crazy (especially for a hotkey centric app), keystrokes still do different things in different places, and so on. If anyone was actually interested in listening and acting on what they hear those niggles would have been fixed ages ago.

I wish I could find more positive things to say, but every time I start discussing KiCad (like pointing out the embarrassing release of 6.0 with big chunks of basic documentation missing) I get faced with someone whose only answer is to ignore the criticisms and fall back on "Why don't you go off and fix it?". Not everybody who has a criticism, insight, suggestion or whatever has the skillset to "Go and fix it.". Similarly, sethhillbrand's remarks don't sound like "OK. Go ahead, I'm listening" but a handwaving "If you are a professional CAD designer/engineer, we would welcome your input." followed by "You can post issues on GitLab and we can arrange offline contact options as well to facilitate the sort of information exchange that we need to improve the system for everyone.". As I've read through quite a lot of other people's KiCad tickets on exactly the issues that I find pain points I've noticed that they tend to just get closed, not acted on, no matter how much effort someone has put into authoring the ticket. Thus I'm inclined to be highly reluctant to go and waste a few hours on creating tickets just to find them quietly and unaccountably closed a while later.

People don't want to write tickets that they suspect might quietly close without action, they don't want to engage in an "offline contact option"*  to participate in a "facilitated information exchange"*. They want to talk, and be listened to.


*Do I need to point out that those phrases sound like bullshit? Maybe they are, maybe they aren't, but they sound like it.

[Ed.Kloonk]

Keep in mind as a dev of open source, you start a project because of what else is around that bugs you. Sometimes, it's just one thing that annoys you so much you make the start. The problem is these things grow legs and turn into a monster.

[delfinom]

The problem with these kind of projects is that people work on what they want to, what interests them, rather than what is needed. That's the nature of the beast and it would take much cannier project management than you ever see to steer the ship in the direction of doing "what's needed"

That's the problem my dude. This isn't a commercial entity, the majority aren't paid to do this work and it would be unreasonable to demand that kind of compliance. People do this as a hobby, not a second job that pays nothing. There are certain reasonable expectations as a result.

I get faced with someone whose only answer is to ignore the criticisms and fall back on "Why don't you go off and fix it?". Not everybody who has a criticism, insight, suggestion or whatever has the skillset to "Go and fix it."

I'm really confused why you are going on this tangent. I have not seen any of this in this thread.

I've noticed that they tend to just get closed, not acted on, no matter how much effort someone has put into authoring the ticket

* citation required

If you haven't noticed there's a large pile of issues, and we aren't here to provide customer support. Crashes get fixed first, bugs get fixed second, then there is staring at wishlight items over time which can be months or years unforunately as there is not enough manpower or time to do everything.

Also v6 had rocky development, it was in development for too long, it was in feature freeze for too long and the result is that wishlist pile was ignored even longer.

[Cerebus]

I get faced with someone whose only answer is to ignore the criticisms and fall back on "Why don't you go off and fix it?". Not everybody who has a criticism, insight, suggestion or whatever has the skillset to "Go and fix it."

I'm really confused why you are going on this tangent. I have not seen any of this in this thread.

It's a recurring theme. Just because it hasn't happened in this thread doesn't mean it hasn't happened, and moreover happened often enough in front of me to form the opinion that it is stereotypical behaviour.

I've noticed that they tend to just get closed, not acted on, no matter how much effort someone has put into authoring the ticket

* citation required

If you haven't noticed there's a large pile of issues, and we aren't here to provide customer support. Crashes get fixed first, bugs get fixed second, then there is staring at wishlight items over time which can be months or years unforunately as there is not enough manpower or time to do everything.

I'm not going to spend 2 hours finding an example in the 8,842 closed issues, or weeks monitoring the  1,404 open issues purely to find an example just to prove a point - my time is worth more to me than that. And while I wasn't saying that sethhillbrand's "open an issue" was a deliverate brush-off, merely less than helpful, you do seem to be saying so if there are too many issues for anyone to take notice of a new one that isn't panic inducing. Which one is it? Opening an issue would be a valuable way of providing input, or it would just be ignored because there are too many? I believe that this ticking petard is yours.

[thinkfat]

So, summing up the arguments: "You need to do what I want and if you don't you're stupid"!
Did I get that right?

[MitjaN]

In KiCad 5, I used a plugin to duplicate PCB footprint layouts based on hierarchical sheets:

https://github.com/MitjaNemec/Kicad_action_plugins

Are there any other options with KiCad 6, or just hang tight till there's some plugins available?

Well I’ve started on porting the plugin, but as I’ll also refactor it, I don’t expect to finish before middle of February.

If you are in a hurry though, I believe that existing ReplicateLayout plugin from “5.99_test” branch should work with KiCad 6.0.x

[John B]

In KiCad 5, I used a plugin to duplicate PCB footprint layouts based on hierarchical sheets:

https://github.com/MitjaNemec/Kicad_action_plugins

Are there any other options with KiCad 6, or just hang tight till there's some plugins available?

Well I’ve started on porting the plugin, but as I’ll also refactor it, I don’t expect to finish before middle of February.

If you are in a hurry though, I believe that existing ReplicateLayout plugin from “5.99_test” branch should work with KiCad 6.0.x

Thanks so much for your effort (also getting the thread back on track  )

I'll give the old plugin a shot.

[JohnG]

I really wish there was a "whining" or "whinging" filter for this forum...

I demand someone else do something about it!!!

[Cerebus]

So, summing up the arguments: "You need to do what I want and if you don't you're stupid"!
Did I get that right?

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

[Cerebus]

I really wish there was a "whining" or "whinging" filter for this forum...

I demand someone else do something about it!!!

There's an "ignore user" feature. If you don't like what someone says you can use it and live in a little bubble of people that all agree with you.

[ebastler]

I really wish there was a "whining" or "whinging" filter for this forum...
I demand someone else do something about it!!!

There's an "ignore user" feature. If you don't like what someone says you can use it and live in a little bubble of people that all agree with you.

You might have your "ignore tongue-in-cheek subtext" filter still turned on? 

That aside, you are putting JohnG down for a feature he did not ask for. He was not suggesting to ignore all posts from user X, but ignore all whining posts.

[Cerebus]

I really wish there was a "whining" or "whinging" filter for this forum...
I demand someone else do something about it!!!

There's an "ignore user" feature. If you don't like what someone says you can use it and live in a little bubble of people that all agree with you.

You might have your "ignore tongue-in-cheek subtext" filter still turned on? 

That aside, you are putting JohnG down for a feature he did not ask for. He was not suggesting to ignore all posts from user X, but ignore all whining posts.

You'll get that when you get a "moon on a stick, and a pony" tickbox as well.

[delfinom]

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Hahah. O god you don't know the magnitude of the whining we get with the documentation.

1. OMG IT MUST BE TRANSLATED TO 50 LANGUAGES!
2. OMG IT MUST COMPLY WITH LINUX DISTROS XYZ POLICY
3. OMG IT MUST USE ABC TOOLCHAIN AND GENERATE AN EPUB FILE COMPATIBLE WITH MY 10 YEAR OLD KINDLE

The result is, all the documentation contributors have quit and we have been left with outdated docs.
We just had a 50+ angry mailing list chain kickoff because somehow, people think we are idiots and don't know how HTML works. Also making some claims about things without even having explored the topic (Just use asciidoctor! it does PDFs just fine! yea....except it has an outstanding defect of CJK font support being broken).

We are listening but the moment anything is attmepted there's another group that goes NOT LIKE THAT.
We are trying to redo the entire documentation tooling to actually make contributors not need to have a degree in Linux System Administration to write a sentence.

[Cerebus]

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Hahah. O god you don't know the magnitude of the whining we get with the documentation.

1. OMG IT MUST BE TRANSLATED TO 50 LANGUAGES!
2. OMG IT MUST COMPLY WITH LINUX DISTROS XYZ POLICY
3. OMG IT MUST USE ABC TOOLCHAIN AND GENERATE AN EPUB FILE COMPATIBLE WITH MY 10 YEAR OLD KINDLE

The result is, all the documentation contributors have quit and we have been left with outdated docs.
We just had a 50+ angry mailing list chain kickoff because somehow, people think we are idiots and don't know how HTML works. Also making some claims about things without even having explored the topic (Just use asciidoctor! it does PDFs just fine! yea....except it has an outstanding defect of CJK font support being broken).

We are listening but the moment anything is attmepted there's another group that goes NOT LIKE THAT.
We are trying to redo the entire documentation tooling to actually make contributors not need to have a degree in Linux System Administration to write a sentence.

My complaint is just "present, please" the current release has significant holes in it that weren't there in previous releases.

4.3. Editing object properties
All objects have properties that are editable in a dialog. Use the hotkey E or select Properties from the right-click context menu to edit the properties of selected item(s). You can only open the properties dialog if all the items you have selected are of the same type. To edit the properties of different types of items at one time, see the section below on bulk editing tools.

In properties dialogs, any field that contains a numeric value can also accept a basic math expression that results in a numeric value. For example, a dimension may be entered as 2 * 2mm, resulting in a value of 4mm. Basic arithmetic operators as well as parentheses for defining order of operations are supported.

4.4. Working with footprints
NOTE
TODO: Write this section - covers footprint properties, updating from library, etc.
4.5. Working with pads
NOTE
TODO: Write this section - covers pad properties
4.6. Working with zones
NOTE
TODO: Write this section
4.7. Graphical objects
Graphical objects (lines, arcs, rectangles, circles, polygons, and text) can exist on any layer but cannot be assigned to a net. Rectangles, circles, and polygons can be set to be filled or outlines in their properties dialogs. The line width property will control the width of the outline even for filled shapes. Line width can be set to 0 for filled shapes to disable the outline.

4.7.1. Creating graphical shapes
NOTE
TODO: Write this section

[PlainName]

The result is, all the documentation contributors have quit and we have been left with outdated docs.

What's needed to be able to write the docs? Other than a grasp of English

ISTM that even a current user might not be an appropriate person unless they have inside knowledge of how things are supposed to work. Documenting what you managed to find out through trial and error would be better than nothing, I guess, but the ideal would be to document what it's supposed to do. How would one go about getting involved in that?

[thinkfat]

So, summing up the arguments: "You need to do what I want and if you don't you're stupid"!
Did I get that right?

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Well, I didn't respond to you in particular, but "whom the shoe fits"...

What I got is "This release was not ready, because X", with X being of such importance it dwarfs every other achievement that makes the release a proper milestone. This is not much different from "Feature X MUST be worked on or you're stupid".

[phs]

Warning:  Long post ahead.  There is a “bottom line” towards the bottom....

Just wanted to give a tremendous thanks to the KiCad team!!  A number of years ago, we had some major projects lined up and had purchased some new seats of Altium.  The incompetent crooks at Altium (which we had been using and recommending for many years, including introducing their products to numerous Universities as well as to our own customers) screwed things up so badly that we were unable to use the new seats for many many months, and no amount of emailing or calling them helped.  Their sales director for our region at the time even told us through one of his underlings that they didn't have to compensate us in any way because he knew we'd HAVE to renew again the next year. 

HE WAS WRONG!

At that point we ended our relationship with Altium, forever.  We immediately moved to KiCad, as well as one of Altium's competitors, who was more than willing to give us a significant discount and actually had a competent customer service department.  I personally had had experience with GEDA at the time, which I still love, but at that time it seemed that significant resources had chosen to back the KiCad project.  Also, KiCad had gotten just far enough along that, especially combined with FreeCAD, it proved to be capable enough for us to save the day and complete our projects successfully -- all with KiCad and FreeCAD.  Didn't even need to use the seat of Altium's commercial competitor that we had purchased.

Of course back then there were numerous issues and inconveniences we had to learn and adjust to, and FreeCAD was also changing rapidly with a huge learning curve, but after a couple of months of very hard work, it was possible to establish an extremely capable, efficient and reliable workflow.  Being able to have full control to customize nearly every aspect of the development tool chain via scripting, plugins, etc. has proven to be a huge advantage over the constrictions Altium had forced on us for all of those previous years.  And not having to deal with the tremendous cost of Altium – often wasting many hours dealing with their ridiculous bugs and crashes, with no power whatsoever to influence them to add new features, was a surprisingly excellent stress-reliever.  Sure KiCAD had plenty of quirks, but there was so much more we could do about those issues than we ever could have with Altium. 

Were/are there things Altium is capable of that KiCad wasn’t/isn't -- yet --?  Of course, but KiCad’s development pace has been so much faster than I expected, that it has enabled the completion of some quite complicated and extensive projects.  And it’s great that our customers can use the same software that we use – without having to pay exorbitant fees and restrictions for the privilege of doing so.

Being cross-platform is a huge plus, too.  Not having to deal with Microsoft licensing – in fact, at this point not having to deal with any proprietary licensing at all, relieves stress levels as well.  We even have bootable USB sticks you can plug into just about any computer to boot Linux and run our complete dev toolchain.  Try that with Altium.

The other pleasant and heart-warming surprise was the discovery of the power of the KiCad forums and user networks.  Nearly any issue that came up was readily resolved by just reading, or on occasion, asking, on the forums.  There are extremely bright and helpful folks there who spend countless hours helping to improve KiCad and help its users in any way possible.  The same is true with FreeCAD.

A quick aside about FreeCAD:  Yes, it can be really intimidating at first – especially if you’re not comfortable with true parametric design, and/or don’t have a lot of experience using CAD tools.  It is also undergoing rather rapid development in a wide variety of areas, and there are many experiments with different branches of development, which all can be confusing and concerning to newcomers.

Make no mistake, however – FreeCAD is extremely capable, and can compete in many many areas with the big boys out there.  For anyone needing a seriously powerful, reliable mechanical CAD tool that also doesn’t lock in all of your design files to hostile proprietary file formats, has extensive scripting capabilities, and pretty seamless integration with KiCad, FreeCAD is worth the time and effort.  If you find it crashing on you frequently, don’t give up – once you’ve learned its ways (almost always “ways” that are simply best-practice CAD design principles), it is actually a remarkably reliable design tool.  And yes, it will take a substantial amount of time to get really comfortable with.  Robust parametric modeling isn’t something you just learn in a day or two – even with the most powerful CAD tools out there.  But there are many “pros” who complete commercial projects with FreeCAD every day, and prove it’s absolutely capable of a huge array of design tasks.

Another consideration regarding the time one invests in complex tools like these is that with the proprietary tools you have no control whatsoever about the ultimate direction their management might take in the future.  As many folks using the “free” proprietary packages have discovered recently, your tool might be bought, phased out, dramatically increased in price, severe changes in licensing, etc.  Who would ever want to go through what Eagle users have?  Many of them even got to PAY for that “service”!  And what do you then do with all of the data you’ve generated with those tools?

With KiCad, FreeCAD, and so many other powerful open source design tools available, there may be (or not so much) an additional time investment in the beginning to learn the use of the tools, as well as get them customized to maximize workflow efficiency, but the bonus is that they’ll be there for you for as long as you need them – even if their development project were to die, you’ll still be able to build from the source code if necessary, and your data won’t be locked up in a file format that you don’t have access to.  Not to mention there are no licensing restrictions that prevent you from packaging up your entire dev environment in virtual machines/containers that can serve as backup snapshots that can be usable indefinitely.

Of course, just because a tool is open source DOES NOT MEAN that it’s a great tool, or that it will do everything you need.  But there are more and more open source tools available these days that compete very closely with their commercial competitors.  And, many of the open source tools are leveling the playing field at what I believe is an astonishing pace.

KiCad, along with InvenTree, K-nTree, Interactive HTML BOM, RF plugin, StepUp plugin combined with FreeCAD, plus a few others allow me to now laugh at how sick I felt when Altium jerked us around, with us paying many thousands of dollars for the privilege of being completely screwed and possibly ruined.  NEVER AGAIN!  Must admit, though – in the end they did us a great favor!  Scary to consider how much time, money and terror that painful lesson cost, though...

The fact that Digi-key, LCSC, and other distributors are putting resources into supporting KiCad, and more and more manufacturing houses are also seamlessly supporting it, means the picture is getting better every day for a highly efficient, reliable development/production process.

Another great thing about KiCad is that if anyone has a feature/bug they’d like to see addressed, it’s incredibly easy to file the bug or request on the list the developers refer to.  Try that with Altium – they simply don’t care about the little company or individual, because as they made abundantly clear to us:

Either you have enough resources to sue us, or we don’t have to listen to you – at all. 

This is something I think all small businesses would do well to be acutely aware of.  If we hadn’t been able to rapidly come up to speed with affordable/capable alternatives to Altium, we surely would have lost important customers, and might have even had to deal with lawsuits.  This can be the utter destruction of a business, and you know what – ALTIUM DOESN’T CARE!  They fully admit that they think they have you locked in, and that you’re very unlikely to leave them because they pretty much own all of your data, locked up in their proprietary format.

When I see folks complaining about KiCad, FreeCAD, or other similarly capable tools I can certainly understand at least some of their frustration.  But there are easily accessed, numerous ways to get help, and if you are desperate to get a feature implemented, you can always work with the developers and offer to pay for it.  It might actually be far cheaper to pay KiCad devs for needed features than to pay for proprietary products – especially in the long run.

Another piece of advice I’d toss out to new adopters of these tools would be to keep up with the vast wealth of information on the user and developer forums.  Make sure to peruse them daily, if at all possible. 

OF COURSE DOCUMENTATION WILL LAG!  Those of us who depend on these tools want new features/bug fixes released as quickly as possible – we don’t care nearly as much about the documentation always being kept precisely up to date, because it is so simple to just check the forums, tutorials, and GASP – actually ask politely for some help.  I think you’ll find the KiCad forums far more accommodating and far less likely to delete any critical posts when compared to the freak show I witnessed on the Altium forums back in the day.

After having been forced, for a variety of reasons, to rely on so many proprietary tools over the years, I am so excited and relieved to have access to such power – not just the power of the tools themselves, but the power of being free from asinine, uncaring corporations, and the power of being free to customize the development toolchain and workflow to best suit our needs.  I’d much rather give KiCad, FreeCAD, and other open source projects my money, than to repeat the terrifying, expensive, and potentially ruinous experiences with the greedy corps.

Bottom line:  It takes arguably more work up front, but tools like KiCad, FreeCAD, ARM dev tools, Blender, Darktable Luxrender, JupyterLab, Linux itself, and so many more, are easily capable of being used professionally to produce quite complex commercial products.  People are doing it every day.  And the hundreds or thousands, of people-hours you might invest in these tools won’t be wasted because they’ll likely be here for you well into the future.  And, even if they’re not, there will always be ways to fully access and utilize the data you’ve created with them.  Plus you can help in so many different ways to make these tools even better.

So, when you’re frustrated, and feel the urge to moan about issues you’re facing, it might be worth asking a few questions, like:  Are these issues stopping other folks from getting their work done?  Are there any forum posts, tutorials, code repository tickets, or other resources out there that may be helpful?  Is it worth considering you may have to take a slightly different approach to how you use the tool, and that they new approach may even have some advantages that you aren’t able to recognize yet?  Have you politely asked on the forums for help?  And possibly most important:  Have you thought about the most constructive, polite, respectful ways that you might be able to help the situation?? 
And, yes, it’s always good to check the documentation before reaching out to avoid both wasting your own time, but also the very precious time of the developers and other users who try to help others.  Even if the docs might not be as up to date as you think they should be...

If instead, people decide to complain without regard to the fact that others are successfully just getting the job done, or ignoring the wealth of resources available, or aren’t willing to do the least bit to help out and make the situation better, then it’s probably best that they are ignored, because this just wastes everyone’s time and leaves a stench about that no reasonable person wants to deal with.

These are great times for powerful open tools that are rapidly leveling the playing field.  It is truly a miraculous time to be alive, so get away from the stench, and breath the crisp fresh air of exciting new possibilities and – enjoy the fact that even you can make beneficial contributions – no matter who you are!

Freedom to all and to all a good day!

[SiliconWizard]

We are listening but the moment anything is attmepted there's another group that goes NOT LIKE THAT.

That is an inherent problem with open projects. Sure in closed projects, you can also have some disagreement, but it's usually only a few people, and managers just make decisions and everyone follows.
In open projects, *anyone* (whether being active or not in the project) feels entitled to weigh in (and merely having an opinion makes some people think they are active contributors), and contrary to a "closed" environment, none of them feels like anyone in the core team can have the authority to act as a manager. Of course, in large open-source projects, there ARE actual "managers". Linus for Linux is a particularly telling example. But that's also why he is constantly being called a prick by many, who would dream of Linus being kicked out of Linux. (And then, the project would likely become a huge mess, but that's another story...)

Point is: projects need some form of management. And the distributed nature of open-source projects makes it highly probable that many people will question the management of the project, whatever it is.

Social networks in general (and forums in particular) are a good example of what happens when discussions go on "unmanaged".

[nctnico]

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Hahah. O god you don't know the magnitude of the whining we get with the documentation.

1. OMG IT MUST BE TRANSLATED TO 50 LANGUAGES!
2. OMG IT MUST COMPLY WITH LINUX DISTROS XYZ POLICY
3. OMG IT MUST USE ABC TOOLCHAIN AND GENERATE AN EPUB FILE COMPATIBLE WITH MY 10 YEAR OLD KINDLE

The result is, all the documentation contributors have quit and we have been left with outdated docs.
We just had a 50+ angry mailing list chain kickoff because somehow, people think we are idiots and don't know how HTML works. Also making some claims about things without even having explored the topic (Just use asciidoctor! it does PDFs just fine! yea....except it has an outstanding defect of CJK font support being broken).

We are listening but the moment anything is attmepted there's another group that goes NOT LIKE THAT.
We are trying to redo the entire documentation tooling to actually make contributors not need to have a degree in Linux System Administration to write a sentence.

Show me a PCB CAD package that has coherent documentation nowadays...  When I need to know how to get something done in Allegro (Orcad) or Atlium I use Google to find a forum post or instruction video. Usually that is far more informative than using the help function. 15+ years ago Altium came with (thick) paper manuals which described each step of each process (like making symbols) but that only covered the basics.

[delfinom]

The result is, all the documentation contributors have quit and we have been left with outdated docs.

What's needed to be able to write the docs? Other than a grasp of English

ISTM that even a current user might not be an appropriate person unless they have inside knowledge of how things are supposed to work. Documenting what you managed to find out through trial and error would be better than nothing, I guess, but the ideal would be to document what it's supposed to do. How would one go about getting involved in that?

1. Learn asciidoc/asciidoctor (it's not that bad, it's a sane alternative to markdown)
2. Know basic git and gitlab to open a merge request
3. Figure out how to get asciidoctor installed, if on Windows you need ruby + the gem
4. Install VSCode + Asciidoc Extension to get live previews

https://gitlab.com/kicad/services/kicad-doc

[Cerebus]

So, summing up the arguments: "You need to do what I want and if you don't you're stupid"!
Did I get that right?

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Well, I didn't respond to you in particular, but "whom the shoe fits"...

What I got is "This release was not ready, because X", with X being of such importance it dwarfs every other achievement that makes the release a proper milestone. This is not much different from "Feature X MUST be worked on or you're stupid".

In the case of basic, functioning documentation, as in help file "this is what control X does", yes, if it's missing then I would wholeheartedly endorse a point of view that it MUST be there. That's the equivalent of the garage telling your car is ready, and when you get there it's only got three wheels on.

You are the one throwing emotive words like 'stupid' around, the most insulting term I've used in relation to the matter in hand is 'amateurish' and that wasn't even in this thread. In fact I don't recall anyone who has said something to the effect of "I'd like to see X" use the kind of insulting language you're putting into their mouths. That borders on, perhaps, is, making ad-hominem arguments and there's no point in continuing to try and discuss this if you're determined to do that rather than offer reasoned counter arguments to the points made.

Chivvying people on to be better is not the same as calling them stupid. You may chose to insult people you wish to persuade if you like, but personally I find it rather counter productive.

Don't bother replying to me personally because I'm hitting "ignore thread" after this, circular arguments in place of constructive discussion are not what I want to spend my time on. It's ironic that earlier in the thread I'm the one urging people who had previously given up on it to give KiCad another go in light of recent improvements, and now I'm giving up in the face of unthinking zealotry. Go figure.

[Ed.Kloonk]

Warning:  Long post ahead.  There is a “bottom line” towards the bottom....

..../

Holy Moly. Nominal Animal has a new apprentice.

 

[phs]

Holy Moly. Nominal Animal has a new apprentice.

Haha, that one's been building up for a few years.  If you ventured in, I hope you made it out unscathed!

I do love the posts from the Animal, though!

[phs]

But I do know that I would like to be able to integrate my personal component database into my KiCad design flow.  I would be happy to change out my database if that's what it took.

fourfathom:  You may want to check out InvenTree/K-nTree...

[PlainName]

The result is, all the documentation contributors have quit and we have been left with outdated docs.

What's needed to be able to write the docs? Other than a grasp of English

ISTM that even a current user might not be an appropriate person unless they have inside knowledge of how things are supposed to work. Documenting what you managed to find out through trial and error would be better than nothing, I guess, but the ideal would be to document what it's supposed to do. How would one go about getting involved in that?

1. Learn asciidoc/asciidoctor (it's not that bad, it's a sane alternative to markdown)
2. Know basic git and gitlab to open a merge request
3. Figure out how to get asciidoctor installed, if on Windows you need ruby + the gem
4. Install VSCode + Asciidoc Extension to get live previews

https://gitlab.com/kicad/services/kicad-doc

Thanks, but what I meant was more: where does one get the knowledge to put in the documentation?

[PlainName]

You may want to check out InvenTree/K-nTree...

The InvenTree mention looks good - thanks for mentioning that (which hadn't crossed my radar yet and is pretty much what I am looking out for at the moment).

but Google/DuckDuck fail miserably to give any clue as to what K-nTree is. Do you have a link?

[fourfathom]

Here's K-nTree: https://github.com/sparkmicro/Ki-nTree/
(I found it while skimming the InvenTree site).  I have not actually looked at either of these yet, just following up on the phs post (thanks, phs!)

[delfinom]

The result is, all the documentation contributors have quit and we have been left with outdated docs.

What's needed to be able to write the docs? Other than a grasp of English

ISTM that even a current user might not be an appropriate person unless they have inside knowledge of how things are supposed to work. Documenting what you managed to find out through trial and error would be better than nothing, I guess, but the ideal would be to document what it's supposed to do. How would one go about getting involved in that?

1. Learn asciidoc/asciidoctor (it's not that bad, it's a sane alternative to markdown)
2. Know basic git and gitlab to open a merge request
3. Figure out how to get asciidoctor installed, if on Windows you need ruby + the gem
4. Install VSCode + Asciidoc Extension to get live previews

https://gitlab.com/kicad/services/kicad-doc

Thanks, but what I meant was more: where does one get the knowledge to put in the documentation?

Alot of it can be done just by being a normal user that figures it out/understands things. Some of it might require knowing developer intention, but many sections are like, "Want to make a curved track? Press this button and do this".
Or, "This button here does this"

[thinkfat]

So, summing up the arguments: "You need to do what I want and if you don't you're stupid"!
Did I get that right?

No, the argument is "Please listen to your users". I'm not even the one offering "do what I want suggestions" beyond "Consistent UI please" and "Please don't push it out the door with big chunks of the help file missing" which is a reasonably low bar to set for any software project.

Well, I didn't respond to you in particular, but "whom the shoe fits"...

What I got is "This release was not ready, because X", with X being of such importance it dwarfs every other achievement that makes the release a proper milestone. This is not much different from "Feature X MUST be worked on or you're stupid".

In the case of basic, functioning documentation, as in help file "this is what control X does", yes, if it's missing then I would wholeheartedly endorse a point of view that it MUST be there. That's the equivalent of the garage telling your car is ready, and when you get there it's only got three wheels on.

You are the one throwing emotive words like 'stupid' around, the most insulting term I've used in relation to the matter in hand is 'amateurish' and that wasn't even in this thread. In fact I don't recall anyone who has said something to the effect of "I'd like to see X" use the kind of insulting language you're putting into their mouths. That borders on, perhaps, is, making ad-hominem arguments and there's no point in continuing to try and discuss this if you're determined to do that rather than offer reasoned counter arguments to the points made.

Chivvying people on to be better is not the same as calling them stupid. You may chose to insult people you wish to persuade if you like, but personally I find it rather counter productive.

Don't bother replying to me personally because I'm hitting "ignore thread" after this, circular arguments in place of constructive discussion are not what I want to spend my time on. It's ironic that earlier in the thread I'm the one urging people who had previously given up on it to give KiCad another go in light of recent improvements, and now I'm giving up in the face of unthinking zealotry. Go figure.

Okay. Zealot. I've been called many things. But "Zealot" is new. Anyway I don't care if you read this or not, others will, and I'm writing for their sake.

Let me say a few things from my perspective both as a professional software developer as well as contributor to a number of open source software projects. Free Software lives by and large off of spare time dedicated by individuals. There are a few "big money" projects with a majority of contributors on some companies payroll, KiCAD is not one of them.

For every such project comes a point where you _have_ to make a release. Distributions, which are your primary customers, demand it. They want something they can package and ship. Your end users demand a release, because of all the shiny bling you have been showing on conferences for the past two years. So you take inventory of the outstanding issues. And since you know you cannot fix all of them, ever, you hold a triage. And then you sacrifice stuff, and if you don't have people to contribute documentation or translations, this is what gets chopped. Because you know it's not the end of the world for your users. But a bug in pcbnew that crashes it when you draw a trace, that will hurt. So you'll find someone to fix it and you will delay the release to have it fixed. Because that's essential for your users, where a "TODO" in a help file is not, because there's always a community of other users out there that they can ask.

In a professional setup, the story is a little different. Customer gets what is in the contract. If the contract says, you need to deliver documentation, you better have a work package for it and manpower reserved. And if push comes to shove, you can still play "Escalate!" and coax management into getting you an additional technical writer on the team to help you out. Because nobody wants to argue with the customer and talk him out of what he can rightfully expect.

Call me a zealot for knowing the difference. But don't call people "amateurish" who know what is _actually_ important for their users and what isn't.

[PlainName]

Here's K-nTree: https://github.com/sparkmicro/Ki-nTree/

Thanks! Using the correct spelling works wonders 

[MadScientist]

I moved from a professional license version of DipTrace to Kicad 6. ( I had used Kicad 5 for some personal designs ). I don’t find the lack of documentation that bad the  vast majority of Kicad features are intuitive and the rest get answers by the Kicad community . I just completed 6 designs last week and was very happy with Kicad 6 . With very little effort I was productive with Kicad.

Give me features over documentation  any day.

Great software , keep going guys

[thinkfat]

KiCAD 6.0.1 has been released.

[Ed.Kloonk]

KiCAD 6.0.1 has been released.

Now with full documentation of every feature. 

[langwadt]

KiCAD 6.0.1 has been released.

Now with full documentation of every feature. 

just look at the code

[IanJ]

With very little effort I was productive with Kicad.

This is a key feature of KiCad.
A big year coming for KiCad i think!

Ian