Author Topic: Just gave KiCAD another go after 5 years and wow!  (Read 1595 times)

0 Members and 1 Guest are viewing this topic.

Offline electrode

  • Regular Contributor
  • *
  • Posts: 127
  • Country: au
Just gave KiCAD another go after 5 years and wow!
« on: November 25, 2017, 08:34:54 AM »
I remember installing KiCAD in 2012, fumbling with the half-arsed Linux-ported Mac release, fighting to place a couple of components and then deleting all traces of the application. Now, I've tried it again, and I think it's good enough (for my needs) to drop Altium for good (which means no more Windows in my life :D ).

Just thought I'd post some of the things I love and issues I've had (comparing KiCAD 4.0.7 Mac to Altium Summer '09 – old, I know), to see if others had potential solutions/tips/tricks.

Things I love:
• It's free!
• Scrolling with the Mac trackpad works in all directions (no horizontal scrolling + choppy vertical scrolling in Altium in Virtualbox greatly hindered productivity).
• You can place components in the schematic and get on with it, without having to pause to search for a footprint that doesn't exist, then make it yourself and try to remember 30 mins later where you were up to with your schematic.
• Footprints seem to exist, ready to go, for just about everything. No more loading .Intlib files one folder at a time, hoping to find stuff.
• Assigning footprints is done in a really nice batch interface, where you can preview the footprint on a grid to verify it matches your model. Furthermore, they are consistently named and ordered really nicely (e.g. electrolytic capacitor by pitch and body diameters).

Things that are a pain:
• It assumes you have a backspace and a (forward) delete key, which is awkward on a Mac laptop keyboard. Wish you could just use delete for all delete functions.
• While the "select tool, apply tool" paradigm is a lot smoother than it used to be, it still seems to have shortcomings. There doesn't seem to be a way to batch edit in PCBnew, e.g. to select 3 labels at once and set them to "invisible" or to select 3 pads at once and edit their hole size. Instead, I've had to do it one at a time. Anyone found a solution to this?
• On a similar note, selecting multiple components seems to only give you the option to move them together (they start moving straight away). Is there a way to align by centres, distribute equally, etc.? One site said you could shift click the components, then right-click, but there's no visual indication that multiple are selected, and no option to align in the right-click menu for me...
• Hiding the F.fab layer seems to hide the component outlines on this layer, but not their value labels (that are also on this layer). :/
• After routing connections, the airwires don't refresh. I have to toggle ratsnest on and off to refresh them. Surely there's a better way?
• When editing values ('e' shortcut), the return key doesn't work for the OK button, even though it's blue. Pain to have to click. May have to report this as a bug.

Random tip: pressing '1' while routing a track highlights everything connected to that net, which I use all the time now. Is anyone aware of a way to make routing always highlight the pertinent pads like that?

Thanks, KiCADders!
 

Offline rhb

  • Regular Contributor
  • *
  • Posts: 236
  • Country: us
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #1 on: November 25, 2017, 11:41:23 AM »
It's rather funny that I *just* set up a Windows 7 machine to be able to run CAD software after refusing to use any Microsoft software for almost 30 years and only *very* reluctantly using Linux.  But Solaris is not the primary workstation environment now.  I've been amazed by how much software for electronics I have downloaded for free.

However, Windows *is* on a system with removable hard drives, so it runs whatever I want, when I want without fuss.
We all get what we deserve whether we want it or not, either as individuals or members of a group.  Sometimes this is as punishment and sometimes it's a blessing.  Which is always ambiguous and depends entirely upon what we do next.
 

Offline hermit

  • Frequent Contributor
  • **
  • Posts: 275
  • Country: us
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #2 on: November 25, 2017, 12:19:52 PM »
• While the "select tool, apply tool" paradigm is a lot smoother than it used to be, it still seems to have shortcomings. There doesn't seem to be a way to batch edit in PCBnew, e.g. to select 3 labels at once and set them to "invisible" or to select 3 pads at once and edit their hole size. Instead, I've had to do it one at a time. Anyone found a solution to this?

The files are text files so you can use scripts or a text editor to do changes.  At 3 pads/lables/etc  it might not be easier to use that though.

• Hiding the F.fab layer seems to hide the component outlines on this layer, but not their value labels (that are also on this layer). :/
Is the value only on the fab layer?  Sometimes it is also on the silkscreen.  You also have layer and render.  There is an option for values there too.

• After routing connections, the airwires don't refresh. I have to toggle ratsnest on and off to refresh them. Surely there's a better way?

I don't think that is normal behavior.

My knowledge is limited but I follow the Kicad user forum and it is quite active.  You stand a better chance of getting useful help there.

Version 5 is probably just a few months away.  I think mostly trying to get the library structure in order.
 
The following users thanked this post: electrode

Offline electrode

  • Regular Contributor
  • *
  • Posts: 127
  • Country: au
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #3 on: November 25, 2017, 02:19:54 PM »
Thanks guys.

It's rather funny that I *just* set up a Windows 7 machine to be able to run CAD software after refusing to use any Microsoft software for almost 30 years and only *very* reluctantly using Linux.  But Solaris is not the primary workstation environment now.  I've been amazed by how much software for electronics I have downloaded for free.

Haha, been there. I've done dual booting, dedicated Windows PC, Parallels, VMWare and Virtualbox. Before that, Virtual PC in the PPC days. The most seamless experience I've had is either a shared folder or dropbox, keeping Windows offline and still organising my files on the Mac side. That way, there's no wasted resources in running antivirus or coming back to your work all closed after Windows decides to auto-update and restart.  :-[

The files are text files so you can use scripts or a text editor to do changes.  At 3 pads/lables/etc  it might not be easier to use that though.

Is the value only on the fab layer?  Sometimes it is also on the silkscreen.  You also have layer and render.  There is an option for values there too.

Ooh, good point with the text files. Would probably be easy enough to hack up some scripts to do exactly what I want. Thanks!

I've attached a screenshot of the F.Fab thing. As far as I can tell, the labels aren't on the silkscreen layer too. Even if so, shouldn't they show in blue after hiding the F.Fab layer? Is this how it behaves on your install?
 

Offline hermit

  • Frequent Contributor
  • **
  • Posts: 275
  • Country: us
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #4 on: November 25, 2017, 02:35:45 PM »
First, are you using OpenGL canvas?  Hit F11 to make sure.  I'm running the dev version so some things might be different.  I think I kinda remember this problem when running stable.  That might be while there is a separate setting on the render tab.
 
The following users thanked this post: Janus

Offline electrode

  • Regular Contributor
  • *
  • Posts: 127
  • Country: au
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #5 on: November 25, 2017, 02:57:29 PM »
Wow, the OpenGL canvas is much different! Toggling the layer visibility now works as expected, thanks. Are there other advantages to OpenGL mode? Everyone seems to recommend it.
 

Offline hermit

  • Frequent Contributor
  • **
  • Posts: 275
  • Country: us
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #6 on: November 25, 2017, 03:06:50 PM »
All new features are made for it.  If you liked it before, wait till you see the interactive routing that is now available.  The shove and walk around router options.  Really, consider reading the Kicad forum for a while to get up to speed.
 
The following users thanked this post: electrode

Offline electrode

  • Regular Contributor
  • *
  • Posts: 127
  • Country: au
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #7 on: November 25, 2017, 04:42:57 PM »
OpenGL mode is amazing!
• Scrolling and zooming are incredibly smooth.
• Shift-clicking to select multiple things works now (with align, etc.; though not value editing)
• The airwires-not-refreshing thing seems to be fixed (I think).
• The pad highlighting thing I mentioned happens by default now.
• Tracks actually route around according to design rules (in normal canvas, they would place anywhere, but delete themselves if they violated a rule).

You've just made my experience twice as good as it was this morning. Thanks again.

I'll have a read around the official forums.
 

Offline b_force

  • Frequent Contributor
  • **
  • Posts: 757
  • Country: 00
    • One World Concepts
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #8 on: November 25, 2017, 11:37:04 PM »
KiCad has potential, but before they even start to implement any more fancy stuff, they really need to work out some bugs and put the interface on every position in the top 5 list.
It's very inconsistent, very confusing and absolutely not compatible with any other GUI from other programs.
For that reason companies or people with years of experience find it very hard to change to KiCad (close to impossible)
This gives these kind of professionals less confidence in their work, so less trust in the end product.
In my experience even to many failed prototypes or even (worse) production.

I lost several days of time and work trying to fix issues that simply weren't fixable in KiCad and so we had to make a (far) less ideal workaround.
The biggest complain I get from professionals is that they can't change the GUI to their liking.
Which I think is a must have anno 2017/2018 (I mean ALL big soft programs have these features nowadays).

From the cheaper alternatives, DipTrace is doing this a whole lot better (still not there though)

Not just only my personal two cents, but from quite some people who I talked to who (also) have been professional board designer for over 20-30 years.
Personally I don't believe KiCad is going to be anywhere in the next 5 years if they don't change their attitude about this and their priorities.
Which is a real shame to be very honest. Like I said it has tons of potential.
« Last Edit: November 25, 2017, 11:39:00 PM by b_force »
"If you can't explain it simply (or at all), you don't understand it well enough." A. Einstein

http://www.oneworldconcepts.com/
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1931
  • Country: fr
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #9 on: November 26, 2017, 12:01:06 AM »
Please, not this "I don't like the UI ergo the tool is unusable for anything serious" :horse:  again.

I do wonder which CAD package allows you to re-configure the UI in a way that some competitor uses (which is what you are pretty much asking here for).

Yes, if you are 20 years invested in some software, then changing to anything else is going to be hard - the muscle memory needs to be unlearned and new workflow/shortcuts/etc. learned. Takes time and can be infuriating. Heck, even changing between two versions of the same software can be hell (like when Office changed from normal menus to those dynamically changing ribbons).

That certainly doesn't mean that there is nothing that can be improved in KICAD, though.

So sorry, complaining about the UI isn't really an argument. If you don't have the resources (time, training availability, can afford the loss of productivity until you get up to speed again) to switch, then switching would be foolish, regardless of which software are you switching to.

 

Offline imidis

  • Frequent Contributor
  • **
  • Posts: 323
  • Country: ca
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #10 on: November 26, 2017, 12:36:36 AM »
Some programs are more and less intuitive than others. Some have licensing/cost benefits over others. The program you have been using is always going to seem easier or better in some ways. For vector based page layout I didn't have any experience with CorelDRAW but I did find it pretty intuitive to use. I found the way Adobe InDesign worked (UI wise) to be quite annoying, while I used it for some things it was not by any means my goto.

I'm pretty sure InDesign users would think InDesign does it better.  :scared:

But it's about preference, along with the fact that adobe moved to subscription only.... While CorelDRAW has both outright and sub licenses.

I still find myself thinking in some things even though they are a bit different programs - CorelDRAW does do some things better than KiCad UI wise.  :-//

But I would say if people are happy with what they are using it's easier and less frustrating to stay with what you are used to and what works for you.
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1931
  • Country: fr
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #11 on: November 26, 2017, 12:59:27 AM »
Yes, no question about that.

But my point was about the eternal argument that the software is unusable because its UI doesn't fit my preferences. Or rather, in the case of Kicad, "it has potential, but they need to fix the UI before companies will switch to it".

Which is nonsense - quite a few companies are happily using it with no problems. If someone is able to deal with e.g. Eagle and its idiosyncrasies, then Kicad is no problem once you learn it. And many even prefer it because Eagle lacks things such as the push & shove routing Kicad now has.

If someone wants to make an argument about Kicad being unsuitable for professional work, then things like library management, lack of higher end features in the PCB program, lack of mechanical CAD integration and similar would be valid points. But not the UI - companies have no problems using even software with much worse user interfaces if it does the job. It is only a matter of retraining the staff.

 

Offline b_force

  • Frequent Contributor
  • **
  • Posts: 757
  • Country: 00
    • One World Concepts
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #12 on: November 27, 2017, 07:22:59 AM »
Whatever you say.
I am only give opinions I hear over and over again from people working professionally in the field.
People can do whatever they want with that piece of information, I am not gonna sleep less about it.
I think it's only extremely naive and close to stupid to ignore the complains from your costumers or call it even nonsense.

Than you are simply only focusing on the technical side of a product.
That maybe worked 20-30 years back, but won't get you far nowadays anymore.
If people don't like the product because it's a horrendous beast to work with, they don't pay for it.

Once again, you may find it nonsense, others people think very different about that, for VERY good reasons.
"If you can't explain it simply (or at all), you don't understand it well enough." A. Einstein

http://www.oneworldconcepts.com/
 

Offline fourtytwo42

  • Frequent Contributor
  • **
  • Posts: 276
  • Country: gb
  • Interested in all things green/ECO
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #13 on: November 27, 2017, 08:14:31 AM »
In over 40 years of electronic design I have used just about every CAD system there is or has been, every one has its own quirks, lovers and haters! For a free system KiCad is amazing, it has many tools such as follow me, push and shove on layout that used to be associated with systems costing thousands a seat. Library management is different and painful like all CAD systems unless you use your own system independent methods. I have done several pcb's with it both multi-layer fast digital and heavy power and I commend it to anybody. Of course like all tools there is a learn curve but this one is definitely worth the investment IMOP. Actually just uploaded the latest version today and the install over an older version was seamless, well done guys. Just be sure to keep your own libraries and work entirely separate from the installation folders as any seasoned tool user should :) Ohh and always use an independent gerber viewer just for sanity before spending all that money on pcb fab :)
 

Offline saike

  • Contributor
  • Posts: 13
  • Country: gb
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #14 on: November 27, 2017, 08:35:21 AM »
Over the (many) years I have managed to end up with 6   'professional'  cad packages costing $$$ all of which have ceased to work because the dongles have failed or the online unlock code is not recognised, with the same response from all the vendors of "Upgrade to the latest and greatest version at only xxxxx dollars and all your problems will be solved".

I can live with any amount of clunkiness in Kicad at the price, it just takes a bit of effort to learn to work with something new and not be a quitter if you don't understand at the first few tries how the software works.

Edit: For some reason the formatter didn't like all the dollar signs above
« Last Edit: November 27, 2017, 08:36:54 AM by saike »
 

Offline miceuz

  • Frequent Contributor
  • **
  • Posts: 364
  • Country: lt
    • chirp - a soil moisture meter / plant watering alarm
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #15 on: November 27, 2017, 08:46:39 AM »
I am following KiCad development closely - I am running a nightly build and report some bugs from time to time.

For the upcoming release they are putting the schematics editor into order, copy paste is not far away. Also they have put some effort in integrating spice into it, but I think that was a waste of time, but whatever.

One of the major improvements is the work that is being done to libraries - both symbols and footprints. They have defined a standard for schematic symbols and footprints and are reworking the full library according to it. On Linux you can install library nightly and get new updates every day. A lot of basic components have footprints assigned having 3d models. Oh, BTW, 3d export to STEP works also, you need to have all the models in STEP, not legacy wrl though.

One nice feature they have released yesterday is expression evaluation in all the text fields in pcb tool. You just enter 1.27/2+0.6/2 into a field for pad placement when creating a footprint, or when placing something on the board.

Also there is the Python integration - all your pcb model is accessible and the sky is the limit.

All in all, I am using an unstable nightly build for professional work for more than two years now. No major hickups.

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1931
  • Country: fr
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #16 on: November 27, 2017, 08:47:56 AM »
I don't think it is nonsense but people don't buy or invest into a CAD system because of a GUI. That has been maybe true those 20 years ago when you had a choice between a DOS application (with no/primitive GUI) and a Windows UI. Or moving from old and expensive UNIX workstations to commodity PCs. Not today, unless the person with the strong opinions about the UI is also the one paying for it - which is rarely the case except for small businesses. The UIs are simply good enough, give or take some quirks and required training (which is normal, if someone expects to switch from one CAD to another from a different vendor without retraining, they are being delusional).

CAD is a long term investment, with a ton of time spent on training, support, subscriptions and what not. So if a company decides to change such package, they are going to have very good business or financial reasons for it - and the UI preferences of the individual users are not likely to be high on the list. What matters more is whether the new tool does things better than the old one (otherwise there would be no reason to switch). Anyhow, companies like that won't switch to Kicad anyway because it likely doesn't do what their high end package does.

Another such example of "a tool with an impossible UI" is Blender (https://www.blender.org/ - a 3D mesh modelling tool). It is a free program very much comparable with such industrial heavyweights as 3DS Max or Maya, both for videogame asset production and film animation/CGI stuff. And 99% of these UI complaints come from 3DS Max users who are unwilling to learn the shortcuts and are lost without the menus and toolboxes Max has. Even though once you learn it, the UI is extremely logical and well thought out - probably one of the best I have seen in similar software, both mesh modelling and CAD. You are able to perform common tasks much faster because you are working with both hands simultaneously, unlike e.g. in Max where you are spending a lot of time opening and collapsing toolboxes with the mouse while your other hand has nothing to do.

Quote
I think it's only extremely naive and close to stupid to ignore the complains from your costumers or call it even nonsense.

Nobody calls "complaints from customers" nonsense*** However, there are plenty of people who are keen to talk about "horrid UI" and the tool "having potential but not being usable for serious work until X,Y and Z"  whenever Kicad is mentioned - most often because it doesn't fit their own personal preferences and way of working. You have said you had unsurmountable issues with it - fair enough, maybe it is not a product suitable for your needs. However, that doesn't mean it won't work for anyone else neither.


*** Customer is someone who has actually paid for the product, btw - a small but an important distinction in this case. The Kicad devs don't owe anything to anyone here. It may be an arrogant thing to say but it is like that - unless you are paying the developers' bills (or contributing code), you don't really get to have a say in the product. It is no different than with anything else. I am not saying it would be smart to ignore it completely but given that everyone has some opinion about the UI (and you will never satisfy everyone) the developers cannot deal with cosmetic UI issues while there are large tasks still outstanding (e.g. finishing the OpenGL porting or the library system). Nothing would ever get done.
« Last Edit: November 27, 2017, 09:03:52 AM by janoc »
 

Offline b_force

  • Frequent Contributor
  • **
  • Posts: 757
  • Country: 00
    • One World Concepts
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #17 on: November 28, 2017, 10:05:00 AM »

Nobody calls "complaints from customers" nonsense*** However, there are plenty of people who are keen to talk about "horrid UI" and the tool "having potential but not being usable for serious work until X,Y and Z"  whenever Kicad is mentioned - most often because it doesn't fit their own personal preferences and way of working. You have said you had unsurmountable issues with it - fair enough, maybe it is not a product suitable for your needs. However, that doesn't mean it won't work for anyone else neither.


*** Customer is someone who has actually paid for the product, btw - a small but an important distinction in this case. The Kicad devs don't owe anything to anyone here.
I don't understand your 'definition' of  a customer. A customer is just someone using a product.
Paid or not, that is totally irrelevant. If a company brings out a product on the market and want it to grow/more people will be using it you simply need to invest a lot of time and effort in finding out what customers want.
I have worked on similar projects myself and a decent GUI is not complicated to make at all.

EVERY professional piece of software like Altium, Solid Works, PTC Creo, Maya and even thinks like Adobe Premiere Pro and similar suits, users can change a lot on the GUI, move windows, change a ton of short-cuts and so on.

In this case it's not only frustrating for users, but I think also even missed changes for the team that's working on KiCad.
The point is, I don't get why people get ALL worked up by it.
People give well meant feedback to make a very potential products awesome, and than your response is 'maybe it is not a product suitable for your needs'.
I am sorry, but that's not how you run a business anymore nowadays, if it's not only for just the attitude and the approach.

So ones again, to summarize. I am very much convinced that if the team behind KiCad totally nail down the user interface, this WILL be the best product on the professional market.
This means; make full usage not only of the keyboard, but window tools, mouse buttons en shortcut menus, as well as making the 'flow' to certain steps more smooth.
If you wanna make people even more happy, make a few build in interface presets, so people from Altium, PADS or Eagle can just use the keys and menus they were used to.
Guaranteed that people will be extremely happy to use KiCad. 


"If you can't explain it simply (or at all), you don't understand it well enough." A. Einstein

http://www.oneworldconcepts.com/
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1931
  • Country: fr
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #18 on: November 28, 2017, 12:23:27 PM »
I don't understand your 'definition' of  a customer. A customer is just someone using a product.
Paid or not, that is totally irrelevant. If a company brings out a product on the market and want it to grow/more people will be using it you simply need to invest a lot of time and effort in finding out what customers want.

That is exactly the root of the problem. You are treating this as another piece of software that you buy at a store and when something doesn't work to your liking, you go and complain to its author(s)/support. Which is perfectly ok, you have paid for a product and if it doesn't work as it should, you should have all the right to complain.

However, this concept flies totally out of the window the moment the "product" we are talking about is software written by a bunch of guys doing it in their free time (for the most part). They are giving you their work for free so you are literally complaining about a gift horse. That's what I have meant about the meaning of "customer".

Your (and many others') suggestions are certainly well intentioned. However, when there are 10 or more such "customers" constantly complaining about this and that or suggesting improvements to suit their needs on the development mailing list, it is a major distraction. This is not a company that has a dedicated team for doing customer support. So these requests won't get anything done any faster because those developers are still working on it in their free time - but now have to deal with these issues on top of it. At some point people simply have to say "no" (or ignore the requests) otherwise nothing would ever get done. And then you see them as arrogant and feel your feedback isn't welcome. From the developer's point of view it is a no-win situation.

Also keep in mind that these people work on the project because it is fun for them and they are enjoying it, not because you (or someone else) is paying them. Which also means they will prioritize work they find interesting and not the work that you may find important. That's the way it works and all this talk about what "customers want" and "growing a market" is simply meaningless in this context. It just doesn't apply, period. It is not a commercial project trying to gain market share.

I have worked on similar projects myself and a decent GUI is not complicated to make at all.

Then please do join the project and help out! Competent interaction designers are hard to come by. However, I am afraid that a decent GUI (I am not talking about slapping a few controls together in a UI designer!) is far from a trivial thing to design and implement, especially when it has to please many users with fairly strong opinions.

EVERY professional piece of software like Altium, Solid Works, PTC Creo, Maya and even thinks like Adobe Premiere Pro and similar suits, users can change a lot on the GUI, move windows, change a ton of short-cuts and so on.

Shortcuts you can redefine in Kicad too. The rest - I have personally never felt the need to reorganize the toolbars or menus (there aren't really any "windows" to change in Kicad), because I rarely if ever use them, with most of the work being done by the keyboard shortcuts. That doesn't mean that such feature wouldn't be useful but it is not really something I have been missing so far. There are much more annoying issues than this in the program - e.g. like having half of the PCB functionality available in the legacy canvas only and all the new stuff in the OpenGL one only, forcing constant switching between the two until everything gets ported to OpenGL.

However, everything on your list is big $$$ software developed by huge companies with dedicated teams working on this for decades. Not a fair comparison at all, IMO - the amount of resources available is several orders of magnitude different. If you want to compare, compare with something like Eagle - which has an even worse interface, IMO, and lacks plenty of features that Kicad now has, all that despite it has always been a (fairly expensive) commercial product with a company behind it.

If the Kicad team had e.g. the AutoDesk resources behind them like the Maya (or now the Eagle) team has, I am sure the development would go a lot faster (people could actually work full time on Kicad!) and things would be better. Alas, they don't. Based on the Github stats, there are about 2 active developers and 2 more semi-regular contributors. The closest they got to full time staff were about 2 developers from CERN working on some features for a while, one full time and one part time.

In this case it's not only frustrating for users, but I think also even missed changes for the team that's working on KiCad.
The point is, I don't get why people get ALL worked up by it.
People give well meant feedback to make a very potential products awesome, and than your response is 'maybe it is not a product suitable for your needs'.
I am sorry, but that's not how you run a business anymore nowadays, if it's not only for just the attitude and the approach.

People do get worked up because a ton of very good work done by a small team of people in their free time is constantly being dismissed and denigrated only because it doesn't meet some arbitrary criteria for "professionalism".

Everyone wants to offer feedback but few people if any contribute to the project in any manner. Be it by code, by monetary contributions, documentation, anything. Yes, you certainly don't run business in this way - but the Kicad team is not running a business! It is a community run project, with people trying to do something they enjoy (and to scratch their own itch) in their free time. So unless you understand this and realize that the project isn't running by the same rules as companies and commercial projects do, you will be hitting your head against a wall, complaining about arrogant and lazy developers and be frustrated be the lack of appreciation for your well intentioned effort.

The response about the product not being suitable for your needs was meant exactly as written - not as an offence but a statement of fact. If I test a tool and it doesn't work for what I am doing, I will look for another one, notabene if it is a free one and I won't loose any money by ditching it (apart from maybe time spent on it). I am using Kicad for my PCBs because it works for me, not because I am some sort of a fanatic - but I am keeping an eye out for other options and evaluating them from time to time (e.g. tried EasyEDA recently, used Eagle before, etc).

I am sorry if my message came across as "worked up", I didn't mean to offend you nor anyone else. I admit that the tone is a bit blunt because this topic is a never ending story, being brought up again and again. Always the same issues about the development team being unresponsive or not willing to accept suggestions from the professionals, "unusable UI", "having potential but will be usable only if ...", etc. always by people who expect the same things they get when they buy $5k/seat software package, be it features or "customer service" - from a 2 person team of volunteers working on their pet project. It is literally the :horse:, if there ever was one.


« Last Edit: November 28, 2017, 12:48:26 PM by janoc »
 
The following users thanked this post: MK14, niman2005, ramime

Offline b_force

  • Frequent Contributor
  • **
  • Posts: 757
  • Country: 00
    • One World Concepts
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #19 on: November 28, 2017, 01:28:05 PM »
Maybe it's just a different way of working on voluntary projects.
I have done many. So yes it's pretty OBVIOUS they do it in their FREE time.
You almost suggest that people don't know that, but it's pretty clear with a free piece of software.

The fact if it's a commercial business or money related or not is totally irrelevant in this whole discussion.
Than you are completely missing the point here.
If you think my intentions and background is only from a professional than you're clearly misreading something.

The point is getting an awesome piece of software out there that people enjoy using.
The fact is, even for free voluntary projects, that you need to stay open for feedback, criticism and need to find a way to 'sell' your program (= marketing).
The fact that it's free doesn't matter, this is just how it simply works to get proper attention.
Unless you don't have any ambition to make it any greater.

For the record, feedback is also a form of contribution.
Actually a very important one. Since most people who are going to use this program are electronic engineers and PCB designers, NOT software programmers.
That means that they have a lot of experience in similar software, have some very good thoughts about it, but just don't know how to code.
A lot of people always know more than just a few programmers. Not only on the hard skills, but also the soft skills.

The most important reason the have a look at your paid competitors, is because they know how extremely important a GUI actually is.
Much more important than most engineers even think it is.
You can even see that on Daves oscilloscope reviews. If the interface sucks people don't want the product, if it feels intuitive, people are willing to spends a few extra bucks for it.
An interface can totally make a break a product.
In my opinion, if it breaks a product you just wasted a significant amount of your time (free or paid), because ones again it's not so difficult to make (or copy) one that works.

I am more than willing to join the project if I feel their attitude and priorities will change.
People have been saying these things almost from the start from KiCad.
The only things we see on the 'todo' list, are things that people don't really care about, or are very minor in developing PCBs.
First thing that people want is something that is a joy to work with, a fancy 3D viewer or spice simulations is nice, but not really something on the priority list.

Ones again, I don't understand the whole debate.
People give well meant constructive feedback to improve a product.
If you can't handle that, no big deal, still friends, but just don't even start something like this.
This topic started were KiCad will be in 5 years time, so I just gave my few cents to that.
With the 'todo list' they showed a few months back in their presentation, I am extremely sorry to say that I don't really see it going anywhere. 
Hopefully I am wrong.
« Last Edit: November 28, 2017, 01:30:36 PM by b_force »
"If you can't explain it simply (or at all), you don't understand it well enough." A. Einstein

http://www.oneworldconcepts.com/
 

Offline HackedFridgeMagnet

  • Super Contributor
  • ***
  • Posts: 1853
  • Country: au
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #20 on: November 28, 2017, 02:45:55 PM »
With the 'todo list' they showed a few months back in their presentation, I am extremely sorry to say that I don't really see it going anywhere. 
If the last 5 years is anything to go by (refer to thread title), you are dead wrong.

Personally I find Kicad's UI far easier and more intuitive than Altium's.

I wouldn't even use Altium if it wasn't our workplace standard. I would be using Kicad.

The following happened in Altium straight after I read this thread. I suppose it is an easy fix I just have to google it an find out what 'obvious' thing I am missing.
I want to create an new schematic library, not and integrated one. I would already have it done in kicad.
No message just everything greyed out.

« Last Edit: November 28, 2017, 03:46:49 PM by HackedFridgeMagnet »
 

Offline HackedFridgeMagnet

  • Super Contributor
  • ***
  • Posts: 1853
  • Country: au
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #21 on: November 28, 2017, 02:51:31 PM »
I figured out what it is.
I had 2 instances of Altium open and it only lets one of them work at a time, the other just goes grey.
You have to change which one is connected to the licence server to do (some)stuff like save/create.

A good ui would have mentioned that. The program knows i have a licence.


 

Online lundmar

  • Regular Contributor
  • *
  • Posts: 201
  • Country: dk
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #22 on: November 28, 2017, 03:39:02 PM »
KiCad is excellent and I'm looking forward to the release of KiCad 5 hopefully in the beginning of next year.

Sure, KiCad is not perfect (yet) and there are still some quirks to get rid of but they are getting fewer and fewer with every new release. It is certainly fully capable of doing professional designs. Considering it is free open source software then that fact becomes pretty amazing.

A big thank you goes to the KiCad maintainers/contributors and CERN for putting a lot of resources into the project so that everyone, and especially the OSHW community, can benefit :-+
« Last Edit: November 28, 2017, 03:42:06 PM by lundmar »
https://lxi-tools.github.io - Open source LXI tools
https://tio.github.io - A simple TTY terminal I/O application
http://dc-power-supply.github.io - OSHW DC power supply project
 
The following users thanked this post: bitwelder, fourtytwo42, ramime

Offline donotdespisethesnake

  • Frequent Contributor
  • **
  • Posts: 492
  • Country: gb
  • Embedded stuff
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #23 on: December 07, 2017, 09:34:09 PM »
This topic started were KiCad will be in 5 years time, so I just gave my few cents to that.
With the 'todo list' they showed a few months back in their presentation, I am extremely sorry to say that I don't really see it going anywhere. 
Hopefully I am wrong.

Yes, you are wrong about everything. You don't even know what a customer is! People have been whining about the KiCad GUI for years, but they never contribute to KiCad. KiCad devs and users ignore the whining and just carry on.

Sorry, but saying "the GUI sucks" is not constructive feedback. There is a class of "professional users" who always complain about KiCad, but they are willing to pay $$ for the commercial offering, but don't donate a cent to Kicad. I think what is happening is that these users have already decided that they don't want to switch away from their comfort zone and learn a new tool, and complaining about a "bad GUI" is an easy target because no one ever agrees what a good interface is. It's a weak complaint though, because every tool has a weird and quirky GUI to someone, or even most people.

The idea that KiCad will fail in 5 years because a few non-users "don't like the GUI" is pure and utter nonsense. The KiCad project has no dependency on a commercial user base. KiCad will continue getting better because it has a pool of devs and users who like it and are willing to contribute time to it. Everything else is FUD.
Bob
 

Online hammy

  • Supporter
  • ****
  • Posts: 387
  • Country: 00
Re: Just gave KiCAD another go after 5 years and wow!
« Reply #24 on: December 07, 2017, 10:42:06 PM »
There is a class of "professional users" who always complain about KiCad, but they are willing to pay $$ for the commercial offering, but don't donate a cent to Kicad.

It is not that easy and not just black and white. A professional user is always watching the price-perfomance.
If I donate money to KiCAD I cannot pretend something. I just can wait and hope the software-development is going in a specific direction. On the other hand, if I buy some software I can evaluate before, I'm quite sure what I get.

At the end just one thing is important - idependent from the software: Can I get the job done?
If KiCAD is sufficient I can use it. If not, I can donate and hope for the best, but it does not solve a today requirement.

Participate and code myself is not an option. I'm not a programmer and I don't want to be one without talent in this field. So there is just one way for me. Donate and wait. If it is someday on a par with the software I use today I can consider a switch. If not I'm not going to run it in parallel, because I want to create my parts and footprints just once. Maintaining one part library is enough work. No need to to this twice and spent unpaid time for this.

Don't get it wrong, KiCAD is nice, it is an amazing software, opensource and free.  Every GUI is different and this is not the important factor for software-evaluation. For professional usage it is the time you spent. Time during layout, time during drc-setup, time during parts-creation and library management. How many advanced feature (delay and phase tuning, T-branch, fly-by) are included? How good is "the flow"? This "flow" is the most important thing. A good "flow" keeps you happy during layout and design.  ;D

I'm just wondering if KiCAD is really eager to evolve into such an advanced software. If it does, it is not for the hobby users any more. Such complex software needs a lot of time to get used to it. I'm sure this is not the focus and goal the development team after all. If it stays on the current level, it is ok. There is a need for it! All these Makers and Hobby users have the need for such a good software. But the other side of the coin is: It will not be on par with professional software. This is just not possible. If you win professionals you lose hobbyists.  :-//

Cheers
hammy
« Last Edit: December 07, 2017, 10:45:06 PM by hammy »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf