Seth (and/or everybody to whom this may concern),
Today I was weighting pros and cons for me to follow up here with more
comments on Multichannel tool. I'm sorry I cannot put them through
gitlab, but the other alternative - silence - doesn't look much better,
so here are my observations.
1. Since Sunday 15th, the "topology mismatch" does not occur on my pcb.
I tried to locate the code update that did that, but failed; still I
cannot believe that my miniscule changes to the board did it. That's a
puzzle.
2. Although I dont have the "topological mismatch" now, the layout
replication tool masses up with component placements. see video
https://www.electric-sheep.eu/6TP2ChMP5cFyS6934FLI3fzo6EkmdOmIYkLlRKllo.mkv,
(4MB) where replication zones are highlighted before replication, and
after replication you can see that components from OUTSIDE of their
definition were moved around - component mapping has a bug.
2.1 I think the mapping should use the "instance" uuid (as present in
schematic files) for components-in-zone identification; Unfortunately
I'm not able to find myself around the source code to suggest a patch (I
don't understand the code).
3. It looks to me, that "rule zones" have two distinct (and ultimately,
contradictory) purposes. One being exactly what the name states for; the
other is "replication". This is because:
3.1 rule zone set areas on pcb, where "exceptions" would apply. In those
regions, the declared "exceptions" will apply to every component one
"puts there" - by chance or by design.
3.2. "replication" (almost by definition) will not have predefined
"areas on pcb", since "areas" on pcb where "replication touches" will be
"defined" during replication based on "replication rules" (like pinned
footprint). In my use cases I use "generate zone" tool using
sub-schematic as bases for "zones". This lead up to multiple zones being
generated on the same sub-schematic (despite ticking the "replace"
option). This a nuisance.
4. in consequence "replication" and "rule" zones already have (sort of)
different configuration options. (like "anchor footprint" ain't useful
for "prevent vias" in rule zone).
My conclusion is that "rule zones" should be more separated at user
interface (event if they share the back-end engine). Those are the
reasons I've earlier opted to put an "option tick" at schematic level to
indicate, that "this sub-sheet will be replicatable".
Ultimately, for the purpose of layout replication, it will be useful to
have "grid/matrix replication" - with explicit grid dimensions and
row/col counts; "circular replication" - with center of rotation, number
of instances and angular step; or more exotic replication path
definitions. Things useless in configuration of true "rule zones". Then
again, since there is an "anchor footprint", for SMD components, the
replication should "flip" replication (on back-side) whenever the
"anchor" for currently replication "zone" is flipped with respect to
anchor. (so replication will propagate to backside, instead of resetting
replication to the initial size the anchor is located at).
I hope those observations will be useful to somebody. It's a shame I
cannot normally put them into the gitlab. (And don't understand the
sources enough to contribute actual code).
Regards,
-R
On 16.06.2025 23:21, 'Seth Hillbrand' via KiCad Developers wrote:
Thanks but without a GitLab report we will be unable to address this.
Please consider that if everyone submitted an issue this way, the team
would be unable to handle any of them.
You may find end users willing to help you with understanding how to use
multi channel tool better at the forum (https://forum.kicad.info
<https://forum.kicad.info>) if you do not wish to register with GitLab.
Seth
KiCad Services Corporation Logo
Seth Hillbrand
*Lead Developer*
+1-530-302-5483
Long Beach, CA
www.kipro-pcb.com <https://www.kipro-pcb.com/> [email protected]
<mailto:[email protected]>
On Mon, Jun 16, 2025 at 2:01 PM Rafał Pietrak <[email protected]
<mailto:[email protected]>> wrote:
Hi,
https://www.electric-
sheep.eu/0WPPnXhN8uAVGNnNzGBaIISw6z7U84p7srSE411BTQ.mkv <https://
www.electric-sheep.eu/0WPPnXhN8uAVGNnNzGBaIISw6z7U84p7srSE411BTQ.mkv>.
It's 16MB and I'll leave it there for a couple of days. Regrettably
gitlab refused my registration, so I cannot use proper channel to hand
it over to you. I don't mind handing over the project files, too. Just
tell me it'll help.
no audio there, so pls follow the mouse moves.
I didn't mean "exclude from multi=channel", quite the opposite. My
intention was to suggest "here multi-channel starts" marker at
schematic/sub-schematic level. This comes from the fact, that currently
implemented generation of zones based on sub-schematic creates areas,
which reflect "currently" (that is "random") placement of components -
because it's done before placement replication action. So in my work I
had to re-generate zones after every attempt to replicate their
configuration. Having a single flag at schematic level fixes that.
-R
On 16.06.2025 19:22, 'Seth Hillbrand' via KiCad Developers wrote:
> Hi Rafał,
>
> If there is a case where sub sheets can't find their matches in the
> multichannel tool, this sounds like it might be a bug. Can you
post a
> minimal example to a bug report?
>
> It is not quite clear to me what additional benefit comes from
> specialized markers for 'exclude from multi-channel'. When
creating the
> multi-channel design, you can already pick and choose which rules
areas
> and/or subsheets are included. However, if you could post a
video of
> your process along with what you are trying to achieve to the bug
> tracker, this would be helpful in either fixing the tool or
fixing the
> documentation
>
> Seth
>
> KiCad Services Corporation Logo
> Seth Hillbrand
> *Lead Developer*
> +1-530-302-5483
> Long Beach, CA
> www.kipro-pcb.com <http://www.kipro-pcb.com> <https://www.kipro-
pcb.com/ <https://www.kipro-pcb.com/>> [email protected]
<mailto:[email protected]>
> <mailto:[email protected] <mailto:[email protected]>>
>
>
>
> On Sat, Jun 14, 2025 at 10:40 PM Rafał Pietrak <solver@electric-
sheep.eu <mailto:[email protected]>
> <mailto:[email protected] <mailto:solver@electric-
sheep.eu>>> wrote:
>
> Hi all,
>
> For some time now I'm using-and-testing the multichannel
tool. During
> that time I've stumbled upon cases where zone mapping of
identical
> copied of sub-schematics fail to find isomorphism complaining of
> "topological missmatch". But those sub-shcematics ARE identical.
>
> So I was wondering, it it would be prudent and along the
design roadmap
> if the "rules zones" based on subschematics were indicated as
sub-sheet
> marker/flag available just like "exclude from simulation" or
"exclude
> from board" are. An additional flag would be something like: "pin
> rule-zone to schematic" (or "schematic also defines role
zone") and
> replace the "generate rule zone" step of defining area to
replicate.
> Currently this becomes possible, as an option to indicate one
part
> withing sub-sheet, that will "govern the placement and
orientation" of
> replicated zones.
>
> I still haven't worked out how exactly and where in the kicad
code the
> "Generate zone" maps the identity of parts involved in the
bug/failure
> resulting in topological mismatches I've experienced during
> replication.
> I'd appreciate a helping hand in understanding the code.
>
> Regards,
>
> -R
>
> --
> You received this message because you are subscribed to the
Google
> Groups "KiCad Developers" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to [email protected]
<mailto:devlist%[email protected]>
> <mailto:devlist%[email protected]
<mailto:devlist%[email protected]>>.
> To view this discussion visit https://groups.google.com/a/
kicad.org/ <https://groups.google.com/a/kicad.org/>
> d/msgid/devlist/b78ce4e8-6f6a-419a-bcde-948fc9f6ed6b%40electric-
> sheep.eu <http://sheep.eu> <https://groups.google.com/a/
kicad.org/d/msgid/devlist/ <https://groups.google.com/a/kicad.org/d/
msgid/devlist/>
> b78ce4e8-6f6a-419a-bcde-948fc9f6ed6b%40electric-sheep.eu
<http://40electric-sheep.eu>>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "KiCad Developers" group.
> To unsubscribe from this group and stop receiving emails from it,
send
> an email to [email protected]
<mailto:devlist%[email protected]>
> <mailto:[email protected]
<mailto:devlist%[email protected]>>.
> To view this discussion visit https://groups.google.com/a/
kicad.org/d/ <https://groups.google.com/a/kicad.org/d/>
> msgid/devlist/CAFdeG-osU9g9xNdF8Eqjb-gP-
> xDDGSbfWsBYUcnTj82BV5AugA%40mail.gmail.com
<http://40mail.gmail.com> <https://groups.google.com/a/ <https://
groups.google.com/a/>
> kicad.org/d/msgid/devlist/CAFdeG-osU9g9xNdF8Eqjb-gP- <http://
kicad.org/d/msgid/devlist/CAFdeG-osU9g9xNdF8Eqjb-gP->
> xDDGSbfWsBYUcnTj82BV5AugA%40mail.gmail.com
<http://40mail.gmail.com>?
> utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google
Groups "KiCad Developers" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected]
<mailto:devlist%[email protected]>.
To view this discussion visit https://groups.google.com/a/kicad.org/
d/msgid/devlist/931a9545-bdca-4c3c-bfb3-d67ad1d84b70%40electric-
sheep.eu <https://groups.google.com/a/kicad.org/d/msgid/
devlist/931a9545-bdca-4c3c-bfb3-d67ad1d84b70%40electric-sheep.eu>.
--
You received this message because you are subscribed to the Google
Groups "KiCad Developers" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to [email protected]
<mailto:[email protected]>.
To view this discussion visit https://groups.google.com/a/kicad.org/d/
msgid/devlist/CAFdeG-
odPUoQgkgPtshRVe_J1h49msuiV5TEOrrizS9mSfFBig%40mail.gmail.com <https://
groups.google.com/a/kicad.org/d/msgid/devlist/CAFdeG-
odPUoQgkgPtshRVe_J1h49msuiV5TEOrrizS9mSfFBig%40mail.gmail.com?
utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "KiCad
Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion visit
https://groups.google.com/a/kicad.org/d/msgid/devlist/72c26c23-835c-46b8-b31a-6c8c8e32cb01%40electric-sheep.eu.
