[Kicad-developers] Length matching/tuning V2.

2014-11-26 Thread Pierre Parent
Hi,

I worked a few month ago on a feature concerning trace length tunning.See This 
branch[1] , this bug 
repport[2], this youtube video[3].

Several people have been contacting me to tell me that it is very useful and I 
should keep working on it, so 
that it can be integrated in mainline.

So I'm going to spend  a few days on it. I will correct bugs, make code 
cleaner. But one of the thing I think 
should be done is port this to openGL mode. The thing is I don't know at all 
how it works. Here is what I 
need to do:

*Do something when the mouse is clicked and my toolkit is selected.
*Do something when the mouse moves and my toolkit is selected.
*Draw temporary lines of given width on the screen.
*Redraw everything, but only the real things (traces, pads ect), not temporary 
lines.
*Prompt a window in order to have the user input a value.

Do you have advices/hints on how to do that? (so that I don't loose too much 
time searching).

Do you see any other thing I should do so that it can be integrated in product 
branch?

Also I think CERN said they would work on length matching. Could you tell me 
more so that we don't work 
on 2 concurrent solutions?

Thank's in advance.

Pierre.


[1] https://code.launchpad.net/%7Epierre-parent-k/kicad/length-tunning
[2] https://bugs.launchpad.net/kicad/+bug/594089
[3] https://www.youtube.com/watch?v=XaBwm5anfYk
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] New documentation format.

2014-11-26 Thread Wayne Stambaugh
On 11/25/2014 8:59 PM, Fat-Zer wrote:
 2014-11-25 18:17 GMT+03:00 Wayne Stambaugh stambau...@verizon.net
 mailto:stambau...@verizon.net:
 
  Are there any plans on reorganization of doc/translation repository from
  scratch?
  e.g. IMO it would be a good decision to drop the history of generated
  files like pdf's and *.mo and exclude them from being indexed.
 
 Yes, all of the binary files will be removed from the source tree and
 generated using makefiles created by CMake.  It may take awhile to get
 all of the CMake stuff in place but that is the plan.
 
 I wanna help up with cmake scripting, I can prepare a template
 repository with ready to use cmake scripts and phony docs if it will be
 appreciated.

It would be greatly appreciated.  You should wait until we finalize the
required documentation tools so you can use CMake to check for the
availability of the tools.  Otherwise, you may have to redo it.  Brian
Sidebotham also volunteered to help with the CMake scripts.  Brian has a
lot of experience with CMake (he wrote the kicad-winbuilder) so he would
be a good resource.

I'm hoping we can nail down the document format and tools in the next
few weeks.  Please coordinate your efforts with Marco since he will most
likely be doing a lot of the document conversion from the current
format.  We have talked about using GitHub to host the documentation so
that decision will also need to be made.

In the future, please reply to the mailing list so other folks know your
intentions so they can help out and/or prevent duplicate efforts.  Thank
you for volunteering to help out with this important part of the project.

Cheers,

Wayne

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Nick Østergaard
This is indeed an annoyting bug, it has existed since 2009 at least.
There is already a bug report about it. See:

https://bugs.launchpad.net/kicad/+bug/1035621

Nick

2014-11-26 18:03 GMT+01:00 jp charras jp.char...@wanadoo.fr:
 Le 26/11/2014 17:18, Wayne Stambaugh a écrit :
 I just discovered something I never noticed before.  Trapezoidal SMD
 pads are not connected to zone files (see attached screen shot).  The
 zone fill also does not follow the contour of the pad outline.  Is this
 by design or should I file a bug report?


 Currently, this is by design.

 The primary goal of these trapezoidal pads is microwave applications.

 For these microwave applications, they are used to connect a large track
 to a narrow pin ( transistor or IC) ( or a narrow track to a large pin)
 *without discontinuity* of the copper width.
 (A discontinuity between a track and a copper pad area can create signal
 integrity issues. In fact any discontinuity on the signal path creates
 issues, at very high frequencies )
 A trapezoidal shape with an edge having the same size as the track, and
 the opposite edge having the size of the transistor or IC pin does not
 create discontinuity.
 One could use also rectangular pads, and trapezoidal track segments to
 avoid discontinuity, but Pcbnew does not know trapezoidal track
 segments, mainly because they are not easy to handle in DRC.

 When you are using trapezoidal shapes for pads, you are expected connect
 a track to these pads using the right edge, and the right track width
 (or the right pad size)
 If this is not the case, a rectangular pad or an oval pad is better
 (more easy to use).

 Adding thermal relief to a trapezoidal shape creates discontinuity, and
 the shape is no more a trapezoid.
 In fact, using a trapezoidal shape for a pad in a copper zone has no
 interest.

 For these reasons, I did not spent time to code thermal reliefs for
 trapezoidal shapes.

 Of course, trapezoidal shapes can have now applications outside the
 microwave applications, but I don't know these applications.

 --
 Jean-Pierre CHARRAS

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread jp charras
Le 26/11/2014 18:43, Nick Østergaard a écrit :
 This is indeed an annoyting bug, it has existed since 2009 at least.
 There is already a bug report about it. See:
 
 https://bugs.launchpad.net/kicad/+bug/1035621

Sorry, but I already said why a trapezoidal pad is designed.

The fact an user is thinking something is not expected is not a proof of
a bug.

Trapezoidal pads have a very specific purpose.
I am pretty sure a copper zone around tracks in a microwave design
breaks the design.

So :
First, define the purpose of trapezoidal pads.
After, and depending on this purpose, decide if thermal reliefs are good
or not.

 
 Nick
 
 2014-11-26 18:03 GMT+01:00 jp charras jp.char...@wanadoo.fr:
 Le 26/11/2014 17:18, Wayne Stambaugh a écrit :
 I just discovered something I never noticed before.  Trapezoidal SMD
 pads are not connected to zone files (see attached screen shot).  The
 zone fill also does not follow the contour of the pad outline.  Is this
 by design or should I file a bug report?


 Currently, this is by design.

 The primary goal of these trapezoidal pads is microwave applications.

 For these microwave applications, they are used to connect a large track
 to a narrow pin ( transistor or IC) ( or a narrow track to a large pin)
 *without discontinuity* of the copper width.
 (A discontinuity between a track and a copper pad area can create signal
 integrity issues. In fact any discontinuity on the signal path creates
 issues, at very high frequencies )
 A trapezoidal shape with an edge having the same size as the track, and
 the opposite edge having the size of the transistor or IC pin does not
 create discontinuity.
 One could use also rectangular pads, and trapezoidal track segments to
 avoid discontinuity, but Pcbnew does not know trapezoidal track
 segments, mainly because they are not easy to handle in DRC.

 When you are using trapezoidal shapes for pads, you are expected connect
 a track to these pads using the right edge, and the right track width
 (or the right pad size)
 If this is not the case, a rectangular pad or an oval pad is better
 (more easy to use).

 Adding thermal relief to a trapezoidal shape creates discontinuity, and
 the shape is no more a trapezoid.
 In fact, using a trapezoidal shape for a pad in a copper zone has no
 interest.

 For these reasons, I did not spent time to code thermal reliefs for
 trapezoidal shapes.

 Of course, trapezoidal shapes can have now applications outside the
 microwave applications, but I don't know these applications.

 --
 Jean-Pierre CHARRAS

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp
 


-- 
Jean-Pierre CHARRAS

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Nick Østergaard
2014-11-26 20:00 GMT+01:00 jp charras jp.char...@wanadoo.fr:
 Le 26/11/2014 18:43, Nick Østergaard a écrit :
 This is indeed an annoyting bug, it has existed since 2009 at least.
 There is already a bug report about it. See:

 https://bugs.launchpad.net/kicad/+bug/1035621

 Sorry, but I already said why a trapezoidal pad is designed.

 The fact an user is thinking something is not expected is not a proof of
 a bug.

I do follow you, but I still thought it was worth linking this thread
with the bug report.

 Trapezoidal pads have a very specific purpose.
 I am pretty sure a copper zone around tracks in a microwave design
 breaks the design.

Indeed, IIRC this special behaivour of the trapez pad is not described
in the manual, so one could only assume it was a bug.

On top of my head I can not come up with a design that requires a
trapez pad instead of an oval, rectangular or circular pad. Maybe
wayse could fill in, since he has them in his design.

 So :
 First, define the purpose of trapezoidal pads.
 After, and depending on this purpose, decide if thermal reliefs are good
 or not.


 Nick

 2014-11-26 18:03 GMT+01:00 jp charras jp.char...@wanadoo.fr:
 Le 26/11/2014 17:18, Wayne Stambaugh a écrit :
 I just discovered something I never noticed before.  Trapezoidal SMD
 pads are not connected to zone files (see attached screen shot).  The
 zone fill also does not follow the contour of the pad outline.  Is this
 by design or should I file a bug report?


 Currently, this is by design.

 The primary goal of these trapezoidal pads is microwave applications.

 For these microwave applications, they are used to connect a large track
 to a narrow pin ( transistor or IC) ( or a narrow track to a large pin)
 *without discontinuity* of the copper width.
 (A discontinuity between a track and a copper pad area can create signal
 integrity issues. In fact any discontinuity on the signal path creates
 issues, at very high frequencies )
 A trapezoidal shape with an edge having the same size as the track, and
 the opposite edge having the size of the transistor or IC pin does not
 create discontinuity.
 One could use also rectangular pads, and trapezoidal track segments to
 avoid discontinuity, but Pcbnew does not know trapezoidal track
 segments, mainly because they are not easy to handle in DRC.

 When you are using trapezoidal shapes for pads, you are expected connect
 a track to these pads using the right edge, and the right track width
 (or the right pad size)
 If this is not the case, a rectangular pad or an oval pad is better
 (more easy to use).

 Adding thermal relief to a trapezoidal shape creates discontinuity, and
 the shape is no more a trapezoid.
 In fact, using a trapezoidal shape for a pad in a copper zone has no
 interest.

 For these reasons, I did not spent time to code thermal reliefs for
 trapezoidal shapes.

 Of course, trapezoidal shapes can have now applications outside the
 microwave applications, but I don't know these applications.

 --
 Jean-Pierre CHARRAS

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp



 --
 Jean-Pierre CHARRAS

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] New documentation format.

2014-11-26 Thread Nick Østergaard
2014-11-26 15:13 GMT+01:00 Wayne Stambaugh stambau...@verizon.net:
 On 11/25/2014 8:59 PM, Fat-Zer wrote:
 2014-11-25 18:17 GMT+03:00 Wayne Stambaugh stambau...@verizon.net
 mailto:stambau...@verizon.net:

  Are there any plans on reorganization of doc/translation repository 
 from
  scratch?
  e.g. IMO it would be a good decision to drop the history of generated
  files like pdf's and *.mo and exclude them from being indexed.

 Yes, all of the binary files will be removed from the source tree and
 generated using makefiles created by CMake.  It may take awhile to get
 all of the CMake stuff in place but that is the plan.

 I wanna help up with cmake scripting, I can prepare a template
 repository with ready to use cmake scripts and phony docs if it will be
 appreciated.

 It would be greatly appreciated.  You should wait until we finalize the
 required documentation tools so you can use CMake to check for the
 availability of the tools.  Otherwise, you may have to redo it.  Brian
 Sidebotham also volunteered to help with the CMake scripts.  Brian has a
 lot of experience with CMake (he wrote the kicad-winbuilder) so he would
 be a good resource.

 I'm hoping we can nail down the document format and tools in the next
 few weeks.  Please coordinate your efforts with Marco since he will most
 likely be doing a lot of the document conversion from the current
 format.  We have talked about using GitHub to host the documentation so
 that decision will also need to be made.

For the moment the conversion/decision process is hosted on Marco's
git repo on github, so if the cmake'ers can't wait, they could work
from that already. I know that Marco has spent some time to write
normal Makefiles, so I guess they could help a little on what commands
actually is needed to generate the docs. I am sure he will be happy
with any help or involvement he can get to work with cmake. But not to
forget, since the format has not been decided yet, work will be lost
if one decide to support all the formats available for testing.

 In the future, please reply to the mailing list so other folks know your
 intentions so they can help out and/or prevent duplicate efforts.  Thank
 you for volunteering to help out with this important part of the project.

 Cheers,

 Wayne

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Adam Wolf
Nick,

I have used trapezoidal pads to connect between PCB antennas and traces.
There are precise rules on trace thickness for antennas, and this let me
have one thing I could modify in the footprint to handle multiple different
feedline widths.

Adam Wolf
Cofounder and Engineer
WL

On Wed, Nov 26, 2014 at 1:16 PM, Nick Østergaard oe.n...@gmail.com wrote:

 2014-11-26 20:00 GMT+01:00 jp charras jp.char...@wanadoo.fr:
  Le 26/11/2014 18:43, Nick Østergaard a écrit :
  This is indeed an annoyting bug, it has existed since 2009 at least.
  There is already a bug report about it. See:
 
  https://bugs.launchpad.net/kicad/+bug/1035621
 
  Sorry, but I already said why a trapezoidal pad is designed.
 
  The fact an user is thinking something is not expected is not a proof of
  a bug.

 I do follow you, but I still thought it was worth linking this thread
 with the bug report.

  Trapezoidal pads have a very specific purpose.
  I am pretty sure a copper zone around tracks in a microwave design
  breaks the design.

 Indeed, IIRC this special behaivour of the trapez pad is not described
 in the manual, so one could only assume it was a bug.

 On top of my head I can not come up with a design that requires a
 trapez pad instead of an oval, rectangular or circular pad. Maybe
 wayse could fill in, since he has them in his design.

  So :
  First, define the purpose of trapezoidal pads.
  After, and depending on this purpose, decide if thermal reliefs are good
  or not.
 
 
  Nick
 
  2014-11-26 18:03 GMT+01:00 jp charras jp.char...@wanadoo.fr:
  Le 26/11/2014 17:18, Wayne Stambaugh a écrit :
  I just discovered something I never noticed before.  Trapezoidal SMD
  pads are not connected to zone files (see attached screen shot).  The
  zone fill also does not follow the contour of the pad outline.  Is
 this
  by design or should I file a bug report?
 
 
  Currently, this is by design.
 
  The primary goal of these trapezoidal pads is microwave applications.
 
  For these microwave applications, they are used to connect a large
 track
  to a narrow pin ( transistor or IC) ( or a narrow track to a large pin)
  *without discontinuity* of the copper width.
  (A discontinuity between a track and a copper pad area can create
 signal
  integrity issues. In fact any discontinuity on the signal path creates
  issues, at very high frequencies )
  A trapezoidal shape with an edge having the same size as the track, and
  the opposite edge having the size of the transistor or IC pin does not
  create discontinuity.
  One could use also rectangular pads, and trapezoidal track segments to
  avoid discontinuity, but Pcbnew does not know trapezoidal track
  segments, mainly because they are not easy to handle in DRC.
 
  When you are using trapezoidal shapes for pads, you are expected
 connect
  a track to these pads using the right edge, and the right track width
  (or the right pad size)
  If this is not the case, a rectangular pad or an oval pad is better
  (more easy to use).
 
  Adding thermal relief to a trapezoidal shape creates discontinuity, and
  the shape is no more a trapezoid.
  In fact, using a trapezoidal shape for a pad in a copper zone has no
  interest.
 
  For these reasons, I did not spent time to code thermal reliefs for
  trapezoidal shapes.
 
  Of course, trapezoidal shapes can have now applications outside the
  microwave applications, but I don't know these applications.
 
  --
  Jean-Pierre CHARRAS
 
  ___
  Mailing list: https://launchpad.net/~kicad-developers
  Post to : kicad-developers@lists.launchpad.net
  Unsubscribe : https://launchpad.net/~kicad-developers
  More help   : https://help.launchpad.net/ListHelp
 
 
 
  --
  Jean-Pierre CHARRAS
 
  ___
  Mailing list: https://launchpad.net/~kicad-developers
  Post to : kicad-developers@lists.launchpad.net
  Unsubscribe : https://launchpad.net/~kicad-developers
  More help   : https://help.launchpad.net/ListHelp

 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Vesa Solonen
26/11/14, 21:16, Nick Østergaard kirjoitti:
 2014-11-26 20:00 GMT+01:00 jp charras jp.char...@wanadoo.fr:
 Le 26/11/2014 18:43, Nick Østergaard a écrit :
 This is indeed an annoyting bug, it has existed since 2009 at least.
 There is already a bug report about it. See:

 https://bugs.launchpad.net/kicad/+bug/1035621

 Sorry, but I already said why a trapezoidal pad is designed.

 The fact an user is thinking something is not expected is not a proof of
 a bug.
 
 I do follow you, but I still thought it was worth linking this thread
 with the bug report.
 
 Trapezoidal pads have a very specific purpose.
 I am pretty sure a copper zone around tracks in a microwave design
 breaks the design.


From an user perspective if xyz pad shape is supported, it should be
supported without artificial limitations. No tool designer may know how
an innovative user uses pad shapes.

The fact that a trapezoidal pad is usable to decrease reflections near
package should not limit it's usefulness for optimisation of, say wave
soldering.

If such shape is used for gradual impedance scaling, the fill keepout
must follow anyway or be far enough not to matter regarding the
transmission line properties. Just the fact that current state leaves
pads unconnected is enough to mark behavior as a bug. Pad left
unconnected from the assigned net is a serious no-no.

I've also used trapezoidal pads for the purpose stated by Jean-Pierre as
well as arc stubs. Thank you for all of the microwave tools!

-Vesa


___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Bob Gustafson


On 11/26/2014 01:16 PM, Nick Østergaard wrote:

2014-11-26 20:00 GMT+01:00 jp charras jp.char...@wanadoo.fr:

Le 26/11/2014 18:43, Nick Østergaard a écrit :

This is indeed an annoyting bug, it has existed since 2009 at least.
There is already a bug report about it. See:

https://bugs.launchpad.net/kicad/+bug/1035621

Sorry, but I already said why a trapezoidal pad is designed.

The fact an user is thinking something is not expected is not a proof of
a bug.

I do follow you, but I still thought it was worth linking this thread
with the bug report.


Trapezoidal pads have a very specific purpose.
I am pretty sure a copper zone around tracks in a microwave design
breaks the design.

Indeed, IIRC this special behaivour of the trapez pad is not described
in the manual, so one could only assume it was a bug.

On top of my head I can not come up with a design that requires a
trapez pad instead of an oval, rectangular or circular pad. Maybe
wayse could fill in, since he has them in his design.


As the system frequency gets higher, board traces begin to act like 
transmission lines.


Trapezoidal pads use transmission line theory.




So :
First, define the purpose of trapezoidal pads.
After, and depending on this purpose, decide if thermal reliefs are good
or not.


Nick

2014-11-26 18:03 GMT+01:00 jp charras jp.char...@wanadoo.fr:

Le 26/11/2014 17:18, Wayne Stambaugh a écrit :

I just discovered something I never noticed before.  Trapezoidal SMD
pads are not connected to zone files (see attached screen shot).  The
zone fill also does not follow the contour of the pad outline.  Is this
by design or should I file a bug report?


Currently, this is by design.

The primary goal of these trapezoidal pads is microwave applications.

For these microwave applications, they are used to connect a large track
to a narrow pin ( transistor or IC) ( or a narrow track to a large pin)
*without discontinuity* of the copper width.
(A discontinuity between a track and a copper pad area can create signal
integrity issues. In fact any discontinuity on the signal path creates
issues, at very high frequencies )
A trapezoidal shape with an edge having the same size as the track, and
the opposite edge having the size of the transistor or IC pin does not
create discontinuity.
One could use also rectangular pads, and trapezoidal track segments to
avoid discontinuity, but Pcbnew does not know trapezoidal track
segments, mainly because they are not easy to handle in DRC.

When you are using trapezoidal shapes for pads, you are expected connect
a track to these pads using the right edge, and the right track width
(or the right pad size)
If this is not the case, a rectangular pad or an oval pad is better
(more easy to use).

Adding thermal relief to a trapezoidal shape creates discontinuity, and
the shape is no more a trapezoid.
In fact, using a trapezoidal shape for a pad in a copper zone has no
interest.

For these reasons, I did not spent time to code thermal reliefs for
trapezoidal shapes.

Of course, trapezoidal shapes can have now applications outside the
microwave applications, but I don't know these applications.

--
Jean-Pierre CHARRAS

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


--
Jean-Pierre CHARRAS

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp



___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


[Kicad-developers] OSX: Configurable OS X target version for building KiCad

2014-11-26 Thread Bernhard Stegmaier
Hi all,

the attached bzr-bundle/patch (against 5302) removes the hard-coded 
“-mmacosx-min-version=10.5” parameters for KiCad, internal boost, and wxWidgets 
build script.
Instead, it is now required to specify the target version using the existing 
CMake parameter -DCMAKE_OSX_DEPLOYMENT_TARGET=10.x” during KiCad configuration 
(or, wxWidgets build script).
It was done on purpose to force an explicit specification of the target 
version, because if nothing is specified various packages seem to set different 
defaults, which might cause problems.
Compiling documentation has been adapted accordingly.

Note:
Setting -DCMAKE_OSX_DEPLOYMENT_TARGET=10.7” (instead of default 10.5 that was 
hardcoded before) seems to fix the red-blob problem with eeschema on recent OS 
X versions.


Regards,
Bernhard



osx-target.bzr
Description: Binary data


osx-target.patch
Description: Binary data
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] OSX: Configurable OS X target version for building KiCad

2014-11-26 Thread Adam Wolf
Great work Bernhard!

I already rigged up my scripts to compile wx with 10.7 as the min version,
so this is the other half!

I think we're going to be able to give the Mac folks something awesome
before the holidays.

Adam Wolf

On Wed, Nov 26, 2014 at 2:10 PM, Bernhard Stegmaier stegma...@sw-systems.de
 wrote:

 Hi all,

 the attached bzr-bundle/patch (against 5302) removes the hard-coded
 “-mmacosx-min-version=10.5” parameters for KiCad, internal boost, and
 wxWidgets build script.
 Instead, it is now required to specify the target version using the
 existing CMake parameter -DCMAKE_OSX_DEPLOYMENT_TARGET=10.x” during KiCad
 configuration (or, wxWidgets build script).
 It was done on purpose to force an explicit specification of the target
 version, because if nothing is specified various packages seem to set
 different defaults, which might cause problems.
 Compiling documentation has been adapted accordingly.

 Note:
 Setting -DCMAKE_OSX_DEPLOYMENT_TARGET=10.7” (instead of default 10.5 that
 was hardcoded before) seems to fix the red-blob problem with eeschema on
 recent OS X versions.


 Regards,
 Bernhard


 ___
 Mailing list: https://launchpad.net/~kicad-developers
 Post to : kicad-developers@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~kicad-developers
 More help   : https://help.launchpad.net/ListHelp


___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread LordBlick

In response to a message written on 26.11.2014 18:03, from jp charras:

Currently, this is by design.

Adding thermal relief to a trapezoidal shape creates discontinuity, and
the shape is no more a trapezoid.
In fact, using a trapezoidal shape for a pad in a copper zone has no
interest.

For these reasons, I did not spent time to code thermal reliefs for
trapezoidal shapes.

Of course, trapezoidal shapes can have now applications outside the
microwave applications, but I don't know these applications.

Trapezoid pads can be useful for control distribution of melting solder paste. 
For each pad, you can set its connection to the copper zone to None and still 
it will be microwave compatible.


--
Best Regards,
LordBlick

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] OSX: Configurable OS X target version for building KiCad

2014-11-26 Thread Andy Peters

 On Nov 26, 2014, at 1:10 PM, Bernhard Stegmaier stegma...@sw-systems.de 
 wrote:
 
 Hi all,
 
 the attached bzr-bundle/patch (against 5302) removes the hard-coded 
 “-mmacosx-min-version=10.5” parameters for KiCad, internal boost, and 
 wxWidgets build script.
 Instead, it is now required to specify the target version using the existing 
 CMake parameter -DCMAKE_OSX_DEPLOYMENT_TARGET=10.x” during KiCad 
 configuration (or, wxWidgets build script).
 It was done on purpose to force an explicit specification of the target 
 version, because if nothing is specified various packages seem to set 
 different defaults, which might cause problems.
 Compiling documentation has been adapted accordingly.
 
 Note:
 Setting -DCMAKE_OSX_DEPLOYMENT_TARGET=10.7” (instead of default 10.5 that 
 was hardcoded before) seems to fix the red-blob problem with eeschema on 
 recent OS X versions.

I will give this a go tonight :)

-a
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Trapezoidal pads and planes.

2014-11-26 Thread Wayne Stambaugh
On 11/26/2014 3:14 PM, LordBlick wrote:
 In response to a message written on 26.11.2014 18:03, from jp charras:
 Currently, this is by design.

 Adding thermal relief to a trapezoidal shape creates discontinuity, and
 the shape is no more a trapezoid.
 In fact, using a trapezoidal shape for a pad in a copper zone has no
 interest.

 For these reasons, I did not spent time to code thermal reliefs for
 trapezoidal shapes.

 Of course, trapezoidal shapes can have now applications outside the
 microwave applications, but I don't know these applications.

 Trapezoid pads can be useful for control distribution of melting solder
 paste. For each pad, you can set its connection to the copper zone to
 None and still it will be microwave compatible.
 

This is sort of what I was using the trapezoidal pads for.  Someone
created them in the Resistors_Universal library.  It appears they were
designed to handle multiple SMD resistor sizes.  I only use them when
prototyping since most of my in house passive components are 0805 and
1206 sizes so it's handy to have these footprints rather than using an
oversized rectangular pad which can make soldering more difficult.  I
was trying to save etching away large amounts of copper by using a
ground plane instead of traces.  This is the first time I tried this so
I was caught off guard by the fact the the zone did not connect to the
pads.  I checked the zone documentation in Pcbnew and I didn't see
anything about trapezoidal pads not connecting.  It's not a big deal but
we should document the fact that trapezoidal pads are not connected to
zone fills.  Although, it would be useful to allow trapezoidal pads to
connect to zones as long as the can be excluded for microwave purposes.
 Thanks JP for the information.


___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


[Kicad-developers] [MERGE] kicad3dmodels

2014-11-26 Thread Cirilo Bernardo
I have no idea who launchpad notifies of merge requests, so for the

benefit of everyone else on this list I am reproducing my merge

request from my kicad branch lp:~cirilo-bernardo/kicad/vrml_tools
https://code.launchpad.net/~cirilo-bernardo/kicad/vrml_tools

=

These changes merge the Kicad3DModels VRML parametric modeling project to
provide users with python scriptable tools for creating 3D models. The
code has been cleaned up so that the public members comply with the KiCad
coding style.

To build, pass CMake the flag -DKICAD_SCRIPTING_VRML=ON

So far this has only been tested on Linux; I would appreciate any help
in testing (and fixing) on OSX and Windows.

Example use, assuming a local installation in ${HOME}/usr/local

export PATH=${HOME}/usr/local/bin:${PATH}
export LD_LIBRARY_PATH=${HOME}/usr/local/lib/kicad/kc3d
export PYTHONPATH=${LD_LIBRARY_PATH}

export KC3DDIR=${HOME}/usr/local/shared/kicad/kc3d/appearances
(required for binary execs to find material appearances)

(now to set execution on some shell scripts)
cd ${HOME}/usr/local/shared/kicad/kc3d/scripts
find . -name *.sh | xargs chmod u+x

cd  mkdir test_vrml  cd test_vrml
${HOME}/usr/local/shared/kicad/kc3d/scripts/batch/conn/generate.sh
${HOME}/usr/local/shared/kicad/kc3d/scripts/batch/diode/generate.sh
${HOME}/usr/local/shared/kicad/kc3d/scripts/batch/dipn/generate.sh
${HOME}/usr/local/shared/kicad/kc3d/scripts/batch/resistor/generate.sh

Some of those bash scripts make use of parameter-driven binary
executables and others invoke python scripts. At the end 6583
models are generated in about 40s and with a total disk usage
of 1.7GB. The source code itself is about 12KSLOC at 1.3MB.
The huge number of models generated in a short time demonstrates
the power of parametric modeling.

- Cirilo
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Gerbview enhancements.

2014-11-26 Thread Lorenzo Marcantonio
On Wed, Nov 26, 2014 at 09:54:50PM +0100, Thiadmer Riemersma wrote:
 A question (because I am confused): are attributes not exclusive to X2?
 
 I understand the Gerber specification as: X1 = RS274X = Gerber without
 attributes, and X2 = RS274X + attributes.

RS274D (pieces of)
Legacy gerber (external aperture definitions), officially obsolete and not 
conforming

RS274X rev I4 
Standard gerber until last year, some commands are deprecated

RS274X rev J1 a.k.a. RS274X2 
Current standard gerber with attributes. Also drill tapes can
'officially' be done in gerber, now (if someone supports them is
another story...)

In short, yes, you are right :D

-- 
Lorenzo Marcantonio
Logos Srl

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp