Mail Archives: geda-user/2017/02/19/15:37:48
--94eb2c124f348594d20548e81a5f
Content-Type: text/plain; charset=UTF-8
Technically it is possible to use the GL rendering code with Lesstif, but
I'll admit it has been a while since I tried it. (Had a quick go as a proof
of concept).
The rendering API changes do facilitate other hids to improve too, but
honestly - the GDK backend is slow, non transparent and is really a legacy
fallback for the increasingly rare cases where GL can't be used for
whatever reason. I have little interest in sending a lot of effort there
myself.
All this said - there is again no reason we cannot take a more physical
approach to whether or not pads are shown in any gui hid.. could switch
them off when the relevant layer group(s) are hidden. (Now expect cries
from people who want to view the pads, but not the tracks on the top /
bottom layers)
On 19 Feb 2017 20:28, "Chad Parker (parker DOT charles AT gmail DOT com) [via
geda-user AT delorie DOT com]" <geda-user AT delorie DOT com> wrote:
Sure, I'm happy to help review.
This still wont solve the problem for the lesstif or gdk HIDs though, only
the GL hid.
--Chad
On Sun, Feb 19, 2017 at 3:02 PM, Peter Clifton (
petercjclifton AT googlemail DOT com) [via geda-user AT delorie DOT com] <
geda-user AT delorie DOT com> wrote:
>
>
> On 19 Feb 2017 19:51, "Chad Parker (parker DOT charles AT gmail DOT com) [via
> geda-user AT delorie DOT com]" <geda-user AT delorie DOT com> wrote:
>
> The problem with the code right now is that pins and pads don't actually
> exist on layers, they are separate. So, turning off a layer has no effect
> on pins or pads. Does your 3D stuff also rework that aspect of
> pins/pads/vias and the layer code?
>
>
> No, but it reworks the rendering stack extensively... for 3D view, it
> renders much more like a physical exporter than the traditional "gui"
> pipeline imposed by draw.c
>
> Any guess as to when it will arrive?
>
>
> Not exactly, but it is certainly high on my non - work related goals list
> for this year to see that code merged. (It has been out of mainline way too
> long).
>
> I can probably point you to a few bits and pieces where I could use help
> and code review if you like. I'm mid-attempt at rebasing the old patch
> series against HEAD, and one area which could really help (as I'm currently
> not designing any boards), is test-driving the result to validate the
> rebase.
>
> Peter
>
>
> --Chad
>
> On Sun, Feb 19, 2017 at 2:25 PM, Peter Clifton (
> petercjclifton AT googlemail DOT com) [via geda-user AT delorie DOT com] <
> geda-user AT delorie DOT com> wrote:
>
>> Hi Chad,
>>
>> (Moving to geda-user as this is not directly answering the OP question)
>>
>> The later 3D stuff will help when it lands. There is a more physical view
>> mode where turning off individual layers also hides any features on that
>> layer such as pads etc..
>>
>> Otoh... the fact OP is having issues makes me wonder if they have GL
>> enabled or not. Transparency helps.
>>
>> Peter
>>
>> Peter
>>
>>
>> On 19 Feb 2017 19:19, "Chad Parker (parker DOT charles AT gmail DOT com) [via
>> geda-help AT delorie DOT com]" <geda-help AT delorie DOT com> wrote:
>>
>> It took me a sec, but I figured out what he means. He's turning off the
>> pins/pads so that they're not in the way while he's routing on inner
>> layers.
>>
>> The problem is that when you turn off the pins/pads, it turns them off
>> for all layers. On the inner layers a plated thru-hole will (often) have a
>> circuilar pad around it, not to mention that there is actually a hole
>> through the board. With pins/pads off, there is no indication of this this
>> at all and you can route a trace straight through a through-hole with out
>> realizing it (or on the top or bottom layer too, actually).
>>
>> Hence the feature's he's asking for. If you could turn off surface mount
>> pads without turning off pins, then this wouldn't be a problem. Or, if we
>> had an "online-DRC" that would prevent you from drawing things that
>> violated design rules, it would also prevent you from doing this.
>>
>> The quickest fix for this, I think, is to change things slightly so that
>> pins and vias are grouped together for visibility and pads are their own
>> separate "layer". The right way to do this probably involves reworking the
>> layer system, but this is a pretty quick hack that should work for now.
>>
>> I opened a bug report on LaunchPad: https://bugs.launchpad.net/pcb
>> /+bug/1666052
>>
>> --Chad
>>
>> On Sat, Feb 18, 2017 at 3:25 AM, Smilie (smilie AT posteo DOT de) [via
>> geda-help AT delorie DOT com] <geda-help AT delorie DOT com> wrote:
>>
>>> I think, PCB need some additional features:
>>>
>>>
>>> 1. split PIN/PAD Layer in two Buttons PIN and PAD.
>>> 2. make it possible, that the "auto enforce DRC clearance" works also
>>> while moving lines.
>>>
>>> 3. the algorithm for "auto enforce DRC clearance" could be modified,
>>> that the line will be lay around the barrier?
>>>
>>>
>>>
>>>
>>> Am Wed, 15 Feb 2017 15:28:44 +0000
>>> schrieb "Peter Clifton (petercjclifton AT googlemail DOT com) [via
>>> geda-help AT delorie DOT com]" <geda-help AT delorie DOT com>:
>>>
>>> > I think I'll have to try that to understand properly; but if you mean
>>> > - you accidentally shorted against pins which you had switched off
>>> > visibility for... Not sure how to avoid that fully.
>>> >
>>> > Perhaps need to separate the view controls a little more. In my later
>>> > 3D stuff (available on repo.or.cz), in 3D view turning off any layer
>>> > also hides things like copper pads on that layer.
>>> >
>>> > Peter
>>> >
>>> > Peter
>>> >
>>> > On 15 Feb 2017 15:19, "Smilie (smilie AT posteo DOT de) [via
>>> > geda-help AT delorie DOT com]" <geda-help AT delorie DOT com> wrote:
>>> >
>>> > > Am Wed, 15 Feb 2017 10:48:02 +0000
>>> > > schrieb "Peter Clifton (petercjclifton AT googlemail DOT com) [via
>>> > > geda-help AT delorie DOT com]" <geda-help AT delorie DOT com>:
>>> > >
>>> > > > Any chance you can post a screenshot of what you describe?
>>> > > >
>>> > > > (Imagebin.ca or similar if the files are big?)
>>> > > >
>>> > > > On 15 Feb 2017 10:45, "Smilie (smilie AT posteo DOT de) [via
>>> > > > geda-help AT delorie DOT com]" <geda-help AT delorie DOT com> wrote:
>>> > > >
>>> > > > > Hello guys,
>>> > > > >
>>> > > > > if i have an Jumper connected through all layers.
>>> > > > > But if i disable the view of all layers, apart from a inner
>>> > > > > signal-layer, the contacts of the connected through is not
>>> > > > > visible. This is in my opinion a bug.
>>> > > > > Or what is this feature for?
>>> > > > >
>>> > > > > lg
>>> > > > >
>>> > >
>>> > >
>>> > > OK, i see the Failure, but i will explain the situation.
>>> > >
>>> > >
>>> > > Step by step:
>>> > >
>>> > > 1. create a new pcb.
>>> > >
>>> > > 2. Load a Jumper1 and place it to the board.
>>> > >
>>> > > 3. Take also some SMD-footprints an place it free over the board.
>>> > >
>>> > > Now you wont draw a line in a (signal) middle layer, but the
>>> > > SMD-Footprints are worry. Because this you make some layers
>>> > > invisible.
>>> > >
>>> > > 4. Now you make the pins/pads invisible. Or what?
>>> > >
>>> > > But, you can not free draw the middle layer, because you have
>>> > > collisions with footprints like Jumper1.
>>> > >
>>> > >
>>> > >
>>>
>>>
>>
>>
>
>
--94eb2c124f348594d20548e81a5f
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<div dir=3D"auto">=C2=A0Technically it is possible to use the GL rendering =
code with Lesstif, but I'll admit it has been a while since I tried it.=
(Had a quick go as a proof of concept).<div dir=3D"auto"><br></div><div di=
r=3D"auto">The rendering API changes do facilitate other hids to improve to=
o, but honestly - the GDK backend is slow, non transparent and is really a =
legacy fallback for the increasingly rare cases where GL can't be used =
for whatever reason. I have little interest in sending a lot of effort ther=
e myself.</div><div dir=3D"auto"><br></div><div dir=3D"auto">All this said =
- there is again no reason we cannot take a more physical approach to wheth=
er or not pads are shown in any gui hid.. =C2=A0could switch them off when =
the relevant layer group(s) are hidden. (Now expect cries from people who w=
ant to view the pads, but not the tracks on the top / bottom layers)</div><=
div dir=3D"auto"><br></div><div dir=3D"auto"><br></div></div><div class=3D"=
gmail_extra"><br><div class=3D"gmail_quote">On 19 Feb 2017 20:28, "Cha=
d Parker (<a href=3D"mailto:parker DOT charles AT gmail DOT com">parker DOT charles AT gmail.=
com</a>) [via <a href=3D"mailto:geda-user AT delorie DOT com">geda-user AT delorie DOT co=
m</a>]" <<a href=3D"mailto:geda-user AT delorie DOT com">geda-user AT delorie=
.com</a>> wrote:<br type=3D"attribution"><blockquote class=3D"quote" sty=
le=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div d=
ir=3D"ltr"><div><div>Sure, I'm happy to help review.<br><br></div>This =
still wont solve the problem for the lesstif or gdk HIDs though, only the G=
L hid.<font color=3D"#888888"><br><br></font></div><font color=3D"#888888">=
--Chad<br></font></div><div class=3D"elided-text"><div class=3D"gmail_extra=
"><br><div class=3D"gmail_quote">On Sun, Feb 19, 2017 at 3:02 PM, Peter Cli=
fton (<a href=3D"mailto:petercjclifton AT googlemail DOT com" target=3D"_blank">pe=
tercjclifton AT googlemail DOT com</a><wbr>) [via <a href=3D"mailto:geda-user AT delo=
rie.com" target=3D"_blank">geda-user AT delorie DOT com</a>] <span dir=3D"ltr"><=
;<a href=3D"mailto:geda-user AT delorie DOT com" target=3D"_blank">geda-user AT delor=
ie.com</a>></span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"=
margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D"=
auto"><span><div><br><div class=3D"gmail_extra"><br><div class=3D"gmail_quo=
te">On 19 Feb 2017 19:51, "Chad Parker (<a href=3D"mailto:parker.charl=
es AT gmail DOT com" target=3D"_blank">parker DOT charles AT gmail DOT com</a>) [via <a href=
=3D"mailto:geda-user AT delorie DOT com" target=3D"_blank">geda-user AT delorie DOT com</=
a>]" <<a href=3D"mailto:geda-user AT delorie DOT com" target=3D"_blank">ge=
da-user AT delorie DOT com</a>> wrote:<br type=3D"attribution"><blockquote clas=
s=3D"m_7142895127585882925m_-1976021762577280436quote" style=3D"margin:0 0 =
0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div>T=
he problem with the code right now is that pins and pads don't actually=
exist on layers, they are separate. So, turning off a layer has no effect =
on pins or pads. Does your 3D stuff also rework that aspect of pins/pads/vi=
as and the layer code?</div></div></blockquote></div></div></div><div dir=
=3D"auto"><br></div></span><div dir=3D"auto">No, but it reworks the renderi=
ng stack extensively... for 3D view, it renders much more like a physical e=
xporter than the traditional "gui" pipeline imposed by draw.c</di=
v><span><div dir=3D"auto"><br></div><div dir=3D"auto"><div class=3D"gmail_e=
xtra"><div class=3D"gmail_quote"><blockquote class=3D"m_7142895127585882925=
m_-1976021762577280436quote" style=3D"margin:0 0 0 .8ex;border-left:1px #cc=
c solid;padding-left:1ex"><div dir=3D"ltr"><div> Any guess as to when it wi=
ll arrive?<font color=3D"#888888"><br></font></div></div></blockquote></div=
></div></div><div dir=3D"auto"><br></div></span><div dir=3D"auto">Not exact=
ly, but it is certainly high on my non - work related goals list for this y=
ear to see that code merged. (It has been out of mainline way too long).</d=
iv><div dir=3D"auto"><br></div><div dir=3D"auto">I can probably point you t=
o a few bits and pieces where I could use help and code review if you like.=
I'm mid-attempt at rebasing the old patch series against HEAD, and one=
area which could really help (as I'm currently not designing any board=
s), is test-driving the result to validate the rebase.</div><span class=3D"=
m_7142895127585882925HOEnZb"><font color=3D"#888888"><div dir=3D"auto"><br>=
</div><div dir=3D"auto">Peter</div></font></span><div><div class=3D"m_71428=
95127585882925h5"><div dir=3D"auto"><br></div><div dir=3D"auto"><div class=
=3D"gmail_extra"><div class=3D"gmail_quote"><blockquote class=3D"m_71428951=
27585882925m_-1976021762577280436quote" style=3D"margin:0 0 0 .8ex;border-l=
eft:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div><font color=3D"#=
888888"><br></font></div><font color=3D"#888888">--Chad<br></font></div><di=
v class=3D"m_7142895127585882925m_-1976021762577280436elided-text"><div cla=
ss=3D"gmail_extra"><br><div class=3D"gmail_quote">On Sun, Feb 19, 2017 at 2=
:25 PM, Peter Clifton (<a href=3D"mailto:petercjclifton AT googlemail DOT com" tar=
get=3D"_blank">petercjclifton AT googlemail DOT com</a><wbr>) [via <a href=3D"mail=
to:geda-user AT delorie DOT com" target=3D"_blank">geda-user AT delorie DOT com</a>] <spa=
n dir=3D"ltr"><<a href=3D"mailto:geda-user AT delorie DOT com" target=3D"_blank=
">geda-user AT delorie DOT com</a>></span> wrote:<br><blockquote class=3D"gmail=
_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:=
1ex"><div dir=3D"auto">Hi Chad,<div dir=3D"auto"><br></div><div dir=3D"auto=
">(Moving to geda-user as this is not directly answering the OP question)<b=
r><div dir=3D"auto"><br></div><div dir=3D"auto">The later 3D stuff will hel=
p when it lands. There is a more physical view mode where turning off indiv=
idual layers also hides any features on that layer such as pads etc..</div>=
<div dir=3D"auto"><br></div><div dir=3D"auto">Otoh... the fact OP is having=
issues makes me wonder if they have GL enabled or not. Transparency helps.=
</div><div dir=3D"auto"><br></div><div dir=3D"auto">Peter</div><div dir=3D"=
auto"><br></div><div dir=3D"auto">Peter</div><div dir=3D"auto"><br></div></=
div></div><div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On 19 F=
eb 2017 19:19, "Chad Parker (<a href=3D"mailto:parker DOT charles AT gmail DOT co=
m" target=3D"_blank">parker DOT charles AT gmail DOT com</a>) [via <a href=3D"mailto:g=
eda-help AT delorie DOT com" target=3D"_blank">geda-help AT delorie DOT com</a>]" &l=
t;<a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">geda-help AT delo=
rie.com</a>> wrote:<br type=3D"attribution"><blockquote class=3D"m_71428=
95127585882925m_-1976021762577280436m_-1478200450418684039m_666050898519535=
7959quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-le=
ft:1ex"><div dir=3D"ltr"><div><div><div>It took me a sec, but I figured out=
what he means.=C2=A0He's turning off the pins/pads so that they're=
not in the way while he's routing on inner layers. <br><br>The
problem is that when you turn off the pins/pads, it turns them off for=20
all layers. On the inner layers a plated thru-hole will (often) have a=20
circuilar pad around it, not to mention that there is actually a hole=20
through the board. With pins/pads off, there is no indication of this=20
this at all and you can route a trace straight through a through-hole=20
with out realizing it (or on the top or bottom layer too, actually).<br><br=
></div>Hence the feature's he's asking for. If you could turn off s=
urface mount pads without turning off pins, then this wouldn't be a pro=
blem. Or, if we had an "online-DRC" that would prevent you from d=
rawing things that violated design rules, it would also prevent you from do=
ing this.<br><br></div>The quickest fix for this, I think, is to change thi=
ngs slightly so that pins and vias are grouped together for visibility and =
pads are their own separate "layer". The right way to do this pro=
bably involves reworking the layer system, but this is a pretty quick hack =
that should work for now. <br><br></div><div>I opened a bug report on Launc=
hPad: <a href=3D"https://bugs.launchpad.net/pcb/+bug/1666052" target=3D"_bl=
ank">https://bugs.launchpad.net/pcb<wbr>/+bug/1666052</a><br></div><br>--Ch=
ad<br></div><div class=3D"m_7142895127585882925m_-1976021762577280436m_-147=
8200450418684039m_6660508985195357959elided-text"><div class=3D"gmail_extra=
"><br><div class=3D"gmail_quote">On Sat, Feb 18, 2017 at 3:25 AM, Smilie (<=
a href=3D"mailto:smilie AT posteo DOT de" target=3D"_blank">smilie AT posteo DOT de</a>) =
[via <a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">geda-help AT d=
elorie.com</a>] <span dir=3D"ltr"><<a href=3D"mailto:geda-help AT delorie DOT c=
om" target=3D"_blank">geda-help AT delorie DOT com</a>></span> wrote:<br><block=
quote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc=
solid;padding-left:1ex">I think, PCB need some additional features:<br>
<br>
<br>
1. split PIN/PAD Layer in two Buttons PIN and PAD.<br>
2. make it possible, that the "auto enforce DRC clearance" works =
also<br>
while moving lines.<br>
<br>
3. the algorithm for "auto enforce DRC clearance" could be modifi=
ed,<br>
that the line will be lay around the barrier?<br>
<br>
<br>
<br>
<br>
Am Wed, 15 Feb 2017 15:28:44 +0000<br>
<div class=3D"m_7142895127585882925m_-1976021762577280436m_-147820045041868=
4039m_6660508985195357959m_602141869714698789HOEnZb"><div class=3D"m_714289=
5127585882925m_-1976021762577280436m_-1478200450418684039m_6660508985195357=
959m_602141869714698789h5">schrieb "Peter Clifton (<a href=3D"mailto:p=
etercjclifton AT googlemail DOT com" target=3D"_blank">petercjclifton AT googlemail DOT c=
om</a><wbr>) [via<br>
<a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">geda-help AT delori=
e.com</a>]" <<a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_bl=
ank">geda-help AT delorie DOT com</a>>:<br>
<br>
> I think I'll have to try that to understand properly; but if you m=
ean<br>
> - you accidentally shorted against pins which you had switched off<br>
> visibility for... Not sure how to avoid that fully.<br>
><br>
> Perhaps need to separate the view controls a little more. In my later<=
br>
> 3D stuff (available on <a href=3D"http://repo.or.cz" rel=3D"noreferrer=
" target=3D"_blank">repo.or.cz</a>), in 3D view turning off any layer<br>
> also hides things like copper pads on that layer.<br>
><br>
> Peter<br>
><br>
> Peter<br>
><br>
> On 15 Feb 2017 15:19, "Smilie (<a href=3D"mailto:smilie AT posteo DOT de=
" target=3D"_blank">smilie AT posteo DOT de</a>) [via<br>
> <a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">geda-help AT d=
elorie.com</a>]" <<a href=3D"mailto:geda-help AT delorie DOT com" target=
=3D"_blank">geda-help AT delorie DOT com</a>> wrote:<br>
><br>
> > Am Wed, 15 Feb 2017 10:48:02 +0000<br>
> > schrieb "Peter Clifton (<a href=3D"mailto:petercjclifton AT goo=
glemail.com" target=3D"_blank">petercjclifton AT googlemail DOT com</a><wbr>) [via=
<br>
> > <a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">geda-h=
elp AT delorie DOT com</a>]" <<a href=3D"mailto:geda-help AT delorie DOT com" tar=
get=3D"_blank">geda-help AT delorie DOT com</a>>:<br>
> ><br>
> > >=C2=A0 Any chance you can post a screenshot of what you descr=
ibe?<br>
> > ><br>
> > > (Imagebin.ca or similar if the files are big?)<br>
> > ><br>
> > > On 15 Feb 2017 10:45, "Smilie (<a href=3D"mailto:smilie=
@posteo.de" target=3D"_blank">smilie AT posteo DOT de</a>) [via<br>
> > > <a href=3D"mailto:geda-help AT delorie DOT com" target=3D"_blank">g=
eda-help AT delorie DOT com</a>]" <<a href=3D"mailto:geda-help AT delorie DOT com=
" target=3D"_blank">geda-help AT delorie DOT com</a>> wrote:<br>
> > ><br>
> > > > Hello guys,<br>
> > > ><br>
> > > > if i have an Jumper connected through all layers.<br>
> > > > But if i disable the view of all layers, apart from a i=
nner<br>
> > > > signal-layer, the contacts of the connected through is =
not<br>
> > > > visible. This is in my opinion a bug.<br>
> > > > Or what is this feature for?<br>
> > > ><br>
> > > > lg<br>
> > > ><br>
> ><br>
> ><br>
> > OK, i see the Failure, but i will explain the situation.<br>
> ><br>
> ><br>
> > Step by step:<br>
> ><br>
> > 1. create a new pcb.<br>
> ><br>
> > 2. Load a Jumper1 and place it to the board.<br>
> ><br>
> > 3. Take also some SMD-footprints an place it free over the board.=
<br>
> ><br>
> > Now you wont draw a line in a (signal) middle layer, but the<br>
> > SMD-Footprints are worry. Because this you make some layers<br>
> > invisible.<br>
> ><br>
> > 4. Now you make the pins/pads invisible. Or what?<br>
> ><br>
> > But, you can not free draw the middle layer, because you have<br>
> > collisions with footprints like Jumper1.<br>
> ><br>
> ><br>
> ><br>
<br>
</div></div></blockquote></div><br></div>
</div></blockquote></div><br></div>
</blockquote></div><br></div>
</div></blockquote></div><br></div></div></div></div></div>
</blockquote></div><br></div>
</div></blockquote></div><br></div>
--94eb2c124f348594d20548e81a5f--
- Raw text -