Author Topic: CERN's contribution to KiCAD  (Read 75544 times)

0 Members and 1 Guest are viewing this topic.

Offline elgonzo

  • Supporter
  • ****
  • Posts: 690
  • Country: 00
Re: CERN's contribution to KiCAD
« Reply #150 on: February 18, 2015, 08:53:45 pm »
I've been using the Ubuntu Software Centre's offering for Kicad (16-Jul-2014 BZR) but am very interested in the CERN advancements.

So two questions:-

a) Can someone point me to a CERN version compiled for Linux preferably with detailed installation instructions?

b) Does anyone with Ubuntu connections know if they intend to offer the new CERN Kicad upgrades any time?

As of now KiCad does not offer stable build (well, they do, but they are rather old).
A stable version with the current feature set is at least several months away.  :=\

More recent builds (based on snapshots from KiCad's source repository) can be found here: http://kicad.nosoftware.cz/
However, the builds offered there are for Kubuntu (i have no idea if it will run on a plain Ubuntu installation).

Build instructions for Linux are found on KiCad's homepage: http://www.kicad-pcb.org/display/DEV/Building+KiCad+on+Linux
« Last Edit: February 18, 2015, 08:55:30 pm by elgonzo »
 

Offline Dave Turner

  • Frequent Contributor
  • **
  • Posts: 439
  • Country: gb
Re: CERN's contribution to KiCAD
« Reply #151 on: February 19, 2015, 12:15:34 am »
Thanks Elgonzo. I'll wait a month or so until the CERN offering is further ahead.
 

Offline Icchan

  • Regular Contributor
  • *
  • Posts: 68
  • Country: fi
Re: CERN's contribution to KiCAD
« Reply #152 on: February 25, 2015, 11:20:20 pm »

Just a point - the Kicad developers don't believe in "official" or "stable" releases.

There is a "stable" development branch in their Launchpad repository but it has not been updated since January.

If they wanted to get more users (not more hackers/developers) they would believe in those. And as such they can't get it as an accepted tool for industry since it doesn't want to be. And since it doesn't want to be a tool for industry it doesn't get more support from industry in form of code / donations. (Industry needs functioning tools and stable releases that they can do work on)

I read developers mailing list the other day, and found out that KiCad doesn't have any "usability team" and it explains why the usability of the program is so clunky and many things just feel little or a lot off. I was however glad that someone had the skills and interest to try to fix many basic problems with KiCads usability, however it can take many months or years for that one particular individual to get those improvements implemented unless he gets help.

KiCad needs more developers but more importantly the development should be better organized (managed) than it is right now IMO... too much freedom creates programs that have a lot of stuff, but in which nothing's polished up.

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7805
  • Country: us
  • adieu
Re: CERN's contribution to KiCAD
« Reply #153 on: February 25, 2015, 11:23:06 pm »

Just a point - the Kicad developers don't believe in "official" or "stable" releases.

There is a "stable" development branch in their Launchpad repository but it has not been updated since January.

If they wanted to get more users (not more hackers/developers) they would believe in those.

Chatter on their IRC channel indicates they're targeting a new stable release this summer with the CERN stuff and a lot more polish.
No longer active here - try the IRC channel if you just can't be without me :)
 

Offline Warhawk

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: 00
    • Personal resume
Re: CERN's contribution to KiCAD
« Reply #154 on: February 26, 2015, 08:35:56 am »
And what is the deal with CERN releases ? I thought that CERN will create a separate branch aside of KiCad developers. Or not ? I also read somewhere that CERN should release something these days, right ?

There are so many information about CERN and KiCad around the web that I am lost, completely  :-//

Offline Icchan

  • Regular Contributor
  • *
  • Posts: 68
  • Country: fi
Re: CERN's contribution to KiCAD
« Reply #155 on: February 26, 2015, 04:19:49 pm »
Chatter on their IRC channel indicates they're targeting a new stable release this summer with the CERN stuff and a lot more polish.

That's good to hear. Hopefully in few years the KiCad really is a tool that someone less technically oriented can use. Or someone who wants a tool that works more than a tool that's free.

If that happens they will get the large industry interested and invested to the development and KiCad will wipe the floor with all others sooner or later.

Offline aveekbh

  • Regular Contributor
  • *
  • Posts: 62
  • Country: in
Re: CERN's contribution to KiCAD
« Reply #156 on: February 26, 2015, 04:38:39 pm »
I've been using the Ubuntu Software Centre's offering for Kicad (16-Jul-2014 BZR) but am very interested in the CERN advancements.
...
a) Can someone point me to a CERN version compiled for Linux preferably with detailed installation instructions?
I use the daily build PPA from https://code.launchpad.net/~js-reynaud/+archive/ubuntu/ppa-kicad, although I don't know if CERN's additions are included.

(Not sure if that helps  :-\ )
 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7805
  • Country: us
  • adieu
Re: CERN's contribution to KiCAD
« Reply #157 on: February 26, 2015, 04:46:58 pm »
And what is the deal with CERN releases ? I thought that CERN will create a separate branch aside of KiCad developers. Or not ? I also read somewhere that CERN should release something these days, right ?

There are so many information about CERN and KiCad around the web that I am lost, completely  :-//

There isn't a CERN-kicad branch per se. The CERN team is doing their own work on the side, and then as work units become complete, merging them into the main code. The team is also working directly on the main code to fix up what's already there. AFAIK, while the work is being done it is not publicly available, though I could be wrong. They are "releasing" it in complete - as in, working, not as in fully polished - units.
No longer active here - try the IRC channel if you just can't be without me :)
 

Online janoc

  • Super Contributor
  • ***
  • Posts: 2912
  • Country: fr
Re: CERN's contribution to KiCAD
« Reply #158 on: February 26, 2015, 04:47:13 pm »
Give the Kicad guys a break, please.

I am using a recently built-from-source version of Kicad. Why there isn't a "stable" release out yet? Well, because it isn't finished yet. If they released something half-assed that eats your files for lunch, the users would whine even more. Would you prefer that?

At this moment they really don't need more users to bombard them with complaints about stuff that is this or that. They need more developers to both lead the project and to polish the things that are there already.

Consider the fact that there were I think only 1 or 2 people from CERN working on the project (and probably not full time neither) and that there are only a few outside volunteer developers working on it. It certainly isn't a sexy project like writing a yet another re-implementation of a window manager or a theme engine or some crazy framework, that always tend to attract a lot of contributors, so no surprise that the pace is quite glacial.

And what is the deal with CERN releases ? I thought that CERN will create a separate branch aside of KiCad developers. Or not ? I also read somewhere that CERN should release something these days, right ?

The CERN guys are working as a part of the Kicad team, there isn't a separate "CERN" release, AFAIK. The CERN stuff is merged in the main code already, but not yet fully integrated. For example, the CERN push & shove routing works, but you must switch to the OpenGL rendering. It doesn't work in the "normal" one. But then you cannot use other tools, like copper pours and have to switch back. It is annoying. The library management is still very rough, I have noticed some bugs with pin placement in the schematic symbol editor, etc. It certainly isn't ready for prime time in the sense you download it and expect everything to work, like with commercial software. It is usable, though.






 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7805
  • Country: us
  • adieu
Re: CERN's contribution to KiCAD
« Reply #159 on: February 26, 2015, 04:50:36 pm »
Give the Kicad guys a break, please.

I am using a recently built-from-source version of Kicad. Why there isn't a "stable" release out yet? Well, because it isn't finished yet. If they released something half-assed that eats your files for lunch, the users would whine even more. Would you prefer that?

At this moment they really don't need more users to bombard them with complaints about stuff that is this or that. They need more developers to both lead the project and to polish the things that are there already.

This so much. They're making huge changes for the next version. This isn't released yet, though it's available because it's an open source project. You're free to use the code as they work on it, but it's incomplete, just as though an artist allowed you to see his unfinished painting. You don't get to whine that it's unfinished, that was the whole deal. Enjoy the unfinished painting if it's good enough for you, and if it's not, go look at the last painting for a while, it wasn't bad. The new, better one will be finished soon enough.
No longer active here - try the IRC channel if you just can't be without me :)
 

Offline ElektroQuark

  • Supporter
  • ****
  • Posts: 1216
  • Country: es
    • ElektroQuark
Re: CERN's contribution to KiCAD
« Reply #160 on: March 03, 2015, 01:58:51 pm »
KiCad Revision 5468:

"Merged the differential pair router & track length tuning tool."

And there are a bunch of other "less" important things added too.

It looks like the Rasberry Pi Fundations has donate a significant amount to the project.


Thanks to all of the pepople that are making this posible.

Offline djsb

  • Frequent Contributor
  • **
  • Posts: 578
  • Country: gb
Re: CERN's contribution to KiCAD
« Reply #161 on: March 03, 2015, 07:37:34 pm »
Video demo here

David
Hertfordshire,UK
 University Electronics Technician, London PIC,CCS C,Arduino,Kicad, Altium Designer,LPKF S103,S62 Operator, Electronics instructor.  http://debuggingrules.com/ Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime.
 

Offline Icchan

  • Regular Contributor
  • *
  • Posts: 68
  • Country: fi
Re: CERN's contribution to KiCAD
« Reply #162 on: March 04, 2015, 04:56:07 pm »
For example, the CERN push & shove routing works, but you must switch to the OpenGL rendering. It doesn't work in the "normal" one. But then you cannot use other tools, like copper pours and have to switch back. It is annoying. The library management is still very rough, I have noticed some bugs with pin placement in the schematic symbol editor, etc. It certainly isn't ready for prime time in the sense you download it and expect everything to work, like with commercial software. It is usable, though.

That's why KiCad developement needs to get little bit more organized IHMO. Goalposts, feature freezes, bug squash releases (to concentrate the efforts of people developing the software) and separate teams that would concentrate on specific areas of the development, usability, routing, graphics, schematic capture (and you could contribute to any of them of course). And someone to oversee the code that's brought to the project in general so it's high quality.

Of course this is a project run by volunteers without monetary gain to be had from the work they put in, but if industry gets a tool they can really use, it will feed back to the project in form of contributions from industry and more money for dedicated development of the tool that they rely on.

Of course that's quite long way away still. but it can happen if people who can, contribute to the project with code and bug-fixes and feature improvements.

Sadly, most electronic engineers are not coders and already own the tools they use to get stuff done. So it's a chicken or the egg problem of getting things going. But it's happening and that's something we can be thankful for.

Offline nuno

  • Frequent Contributor
  • **
  • Posts: 556
  • Country: pt
Re: CERN's contribution to KiCAD
« Reply #163 on: March 04, 2015, 05:29:26 pm »
The link on your signature doesn't work.

I think KiCAD is actually usable, although it lacks some organizing as you say, and has a few annoying interface bugs. It feels like it could easily become a great tool, and I hope it gets there soon.
 

Offline krivx

  • Frequent Contributor
  • **
  • Posts: 763
  • Country: ie
Re: CERN's contribution to KiCAD
« Reply #164 on: March 04, 2015, 05:59:30 pm »
I think KiCAD is actually usable, although it lacks some organizing as you say, and has a few annoying interface bugs. It feels like it could easily become a great tool, and I hope it gets there soon.

I agree it's usable, it's just a little frustrating that most interface bugs seem to be very minor issues. When using different rendering modes the hotkeys change, panning in OpenGL mode seems to be voodoo, some features are not implemented in different modes so I am constantly switching back and forth...

Does anyone know if there is any documentation on jumping into the Kicad code? I am happy to let the main developers work on differential routing if I can hack in consistent keyboard shortcuts.
 

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: CERN's contribution to KiCAD
« Reply #165 on: March 05, 2015, 06:00:04 pm »
; j
Quote from: c4757p link=topic=15721.ms5//  ;m g618487#msg618487 date=1424969436

Give the Kicad guys a break, please.

I am using a recently built-from-source version of Kicad. Why there isn't a "stable" release out yet? Well, because it isn't finished yet. If they released something half-assed that eats your files for lunch, the users would whine even more. Would you prefer that?

At this moment they really don't need more users to bombard them with complaints about stuff that is this or that. They need more developers to both lead the project and to polish the things that are there already.

This so much. They're making huge changes for the next version. This isn't released yet, though it's available because it's an open source project. You're free to use the code as they work on it, but it's incomplete, just as though an artist allowed you to see his unfinished painting. You don't get to whine that it's unfinished, that was the whole deal. Enjoy the unfinished painting if it's good enough for you, and if it's not, go look at the last painting for a while, it wasn't bad. The new, better one will be finished soon enough.

I found this:

http://www.kicad-pcb.org/display/DEV/KiCad+Development

Their site sucks, it's part of a giant TODO and it seems they lack of a web designer/developer for it.

http://bazaar.launchpad.net/~kicad-product-committers/kicad/product/view/head:/Documentation/development/road-map.md
« Last Edit: March 05, 2015, 06:14:33 pm by Circuiteromalaguito »
 

Offline elgonzo

  • Supporter
  • ****
  • Posts: 690
  • Country: 00
Re: CERN's contribution to KiCAD
« Reply #166 on: March 05, 2015, 08:55:51 pm »
I found this:

http://www.kicad-pcb.org/display/DEV/KiCad+Development

Their site sucks, it's part of a giant TODO and it seems they lack of a web designer/developer for it.

http://bazaar.launchpad.net/~kicad-product-committers/kicad/product/view/head:/Documentation/development/road-map.md
Beyond allowing easy downloads of releases and providing proper documentation -- plus some links to up-to-date tutorials and other 3rd-party content perhaps -- there is no pressing need to create a fancy web site. Good and accurate documentation is more important than offering eye candy. Still, the KiCad developers would certainly much appreciate web designers volunteering for this task...
« Last Edit: March 06, 2015, 01:32:37 am by elgonzo »
 

Offline mrflibble

  • Super Contributor
  • ***
  • Posts: 1946
  • Country: nl
Re: CERN's contribution to KiCAD
« Reply #167 on: March 06, 2015, 01:04:03 am »
 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7805
  • Country: us
  • adieu
Re: CERN's contribution to KiCAD
« Reply #168 on: March 06, 2015, 01:21:19 am »
Circuiteromalaguito, you appear not to have read the post you quoted. TL;DR?
No longer active here - try the IRC channel if you just can't be without me :)
 

Offline Icchan

  • Regular Contributor
  • *
  • Posts: 68
  • Country: fi
Re: CERN's contribution to KiCAD
« Reply #169 on: March 06, 2015, 01:42:03 am »
The link on your signature doesn't work.

Thanks, they've changed the address. And this forum software doesn't update your past messages if you change your signature... which is bad :(

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 9095
  • Country: my
  • reassessing directives...
Re: CERN's contribution to KiCAD
« Reply #170 on: March 06, 2015, 02:47:22 am »
Circuiteromalaguito, you appear not to have read the post you quoted. TL;DR?
another new guy on the block that mourns at free stuff...
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: CERN's contribution to KiCAD
« Reply #171 on: March 11, 2015, 11:21:45 pm »
Hello.

I tried to do a constructive opinion, it seems I failed.

I don't care about a fancy website, but a functional one. Take a look at projects such as OpenWRT, they provide a news section on main page so they say to the world their project is alive!  ;)

I just would like some volunteer would improve the site a bit, I'm very interested in KiCad. There's FreeEDA too, a way to integrate it with ngspice and others.

Regards.

 
Circuiteromalaguito, you appear not to have read the post you quoted. TL;DR?
another new guy on the block that mourns at free stuff...
 

Offline elgonzo

  • Supporter
  • ****
  • Posts: 690
  • Country: 00
Re: CERN's contribution to KiCAD
« Reply #172 on: March 12, 2015, 12:40:39 am »
I tried to do a constructive opinion, it seems I failed.
Yupp, you failed ;) Saying something sucks and not saying anything more is definitely an opinion (perhaps even a valid one); but it is not a constructive opinion...

Quote
I don't care about a fancy website, but a functional one. Take a look at projects such as OpenWRT, they provide a news section on main page so they say to the world their project is alive!  ;)
Now, here's where you start making some constructive suggestion... ;)
« Last Edit: March 12, 2015, 12:46:09 am by elgonzo »
 

Offline nickoe

  • Contributor
  • Posts: 24
Re: CERN's contribution to KiCAD
« Reply #173 on: April 03, 2015, 10:01:09 pm »
I believe the CERN developer working on KiCAD is actually Polish, not French. His name's Tomasz Wlostowski I think.

Anyway - Kicad is still light years behind DipTrace, not to mention Altium.

There is not only one. There is two, Maciej Sum?inski and Tomasz Wlostowski. And yes, they are both Polish. Also there is the Javier Serrano, which I think don't write code, but is the more administrative guy in relation between CERN and KiCad.

But for anyone interrested, Javier is requesting success stories see: https://lists.launchpad.net/kicad-developers/msg17537.html

This is what helps getting funding for KiCad, I am sure if you can show some success stories using KiCad in an educational environment, he would definitely also love to hear it.

I don't see your complaint. Is it that the official releases lag the development releases in features? Because duh. This stuff is brand spanking new and not tested in the wild. It's not suitable for general release yet. If you don't want to build the development version yourself, either stop complaining and wait for an official release to come or find some kind soul who has spared you the task of setting up a working build environment and download a binary.

Just a point - the Kicad developers don't believe in "official" or "stable" releases.

There is a "stable" development branch in their Launchpad repository but it has not been updated since January.

It is not quite true that the Developers don't believe in "official" or "stable" releases. Please read this: http://ci.kicad-pcb.org/job/kicad-doxygen/ws/Documentation/doxygen/html/md_Documentation_development_stable-release-policy.html

FYI: CERN gives KiCAD some exposure in the updates section of their web site, also mentioning the soon to be coming differential pair routing and trace length matching features...

It has been committed now! See: https://lists.launchpad.net/kicad-developers/msg17163.html


Just a point - the Kicad developers don't believe in "official" or "stable" releases.

There is a "stable" development branch in their Launchpad repository but it has not been updated since January.

If they wanted to get more users (not more hackers/developers) they would believe in those.

Chatter on their IRC channel indicates they're targeting a new stable release this summer with the CERN stuff and a lot more polish.

Feature freeze has been initiated, https://lists.launchpad.net/kicad-developers/msg17606.html

And what is the deal with CERN releases ? I thought that CERN will create a separate branch aside of KiCad developers. Or not ? I also read somewhere that CERN should release something these days, right ?

There are so many information about CERN and KiCad around the web that I am lost, completely  :-//

They will push to the product branch (the development branch intended to always be in working state watched by the continious integration build server). But when new features are to be released, like for example the differential paris and length tuning, they would only "release" it when it is more or less ready for the public testing.

And what is the deal with CERN releases ? I thought that CERN will create a separate branch aside of KiCad developers. Or not ? I also read somewhere that CERN should release something these days, right ?

There are so many information about CERN and KiCad around the web that I am lost, completely  :-//

There isn't a CERN-kicad branch per se. The CERN team is doing their own work on the side, and then as work units become complete, merging them into the main code. The team is also working directly on the main code to fix up what's already there. AFAIK, while the work is being done it is not publicly available, though I could be wrong. They are "releasing" it in complete - as in, working, not as in fully polished - units.

Ohh well, you also explained it pretty well.

Page where that markup (.md) is rendered:

https://github.com/KiCad/kicad-source-mirror/blob/master/Documentation/development/road-map.md

Or just read it in the compiled dev docs.

I think KiCAD is actually usable, although it lacks some organizing as you say, and has a few annoying interface bugs. It feels like it could easily become a great tool, and I hope it gets there soon.

I agree it's usable, it's just a little frustrating that most interface bugs seem to be very minor issues. When using different rendering modes the hotkeys change, panning in OpenGL mode seems to be voodoo, some features are not implemented in different modes so I am constantly switching back and forth...

Does anyone know if there is any documentation on jumping into the Kicad code? I am happy to let the main developers work on differential routing if I can hack in consistent keyboard shortcuts.

There is not one single document  as a quick start on digging into the code. I could find some references, but they might not help you at all. What you need to do is to get it building and figure out what you want to contribute with. I would suggest you to jump on the IRC channel #kicad@freenode. The main developers are not hanging around in there, but some other valuable contributers are. This should help you get started (if you haven't already).

---

I apoligise if some of theese questions has already been answerd, but just trawled the thread and tried to answer some questions while reading.
 

Offline Bassman59

  • Super Contributor
  • ***
  • Posts: 1257
  • Country: us
  • Yes, I do this for a living
Re: CERN's contribution to KiCAD
« Reply #174 on: April 03, 2015, 10:13:58 pm »
I don't see your complaint. Is it that the official releases lag the development releases in features? Because duh. This stuff is brand spanking new and not tested in the wild. It's not suitable for general release yet. If you don't want to build the development version yourself, either stop complaining and wait for an official release to come or find some kind soul who has spared you the task of setting up a working build environment and download a binary.

Just a point - the Kicad developers don't believe in "official" or "stable" releases.

There is a "stable" development branch in their Launchpad repository but it has not been updated since January.

It is not quite true that the Developers don't believe in "official" or "stable" releases.

You are correct -- remember that my post was from May of last year. I've been following the developer mailing list (and testing things when I can), and they've declared a feature freeze with the goal of a stable release on all supported platforms ASAP.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf