in the next topic about importing printed circuit boards from altium, I feel like a fool … because it seems that no one needs a working function except me) the errors were very gross and it was difficult not to notice them, but there were no messages output?)
I think most people don’t expect importers to be perfect, so they just fix the errors manually, and the errors don’t get reported. And I assume not many people use Altium footprint libraries directly in KiCad.
It helps if issues are reported properly on GitLab, with minimal reproducible examples.
Such little work on a simple feature . . . so why don’t you code up the solution and submit it to be added to the build ? after all KiCad is open source so you have access to all the code.
I’ll do that if things are at this level.
Problem is that a developer already working on the software would do it in 1/100th of the time, because it would take me some time to learn how things are done in this particular project. I can code (been coding for almost 30 years)
I am willing to donate as soon as it is done.
Altium doesn’t appear to allow customizing the overbar, it overlaps with capital accented characters on common fonts… So it’s not clear what you’re trying to say with this.
I don’t really understand what panel we are talking about … if it’s about the desktop, then everything can be made custom if it’s about the design of the scheme, then there is support for all design standards, including for specific countries with their fonts …
can we conclude that the developers at least did not test this function at all in the release version? I have a hard time imagining how it is possible to manually track the entire silkscreen on, say, 300 components or determine that some of the 1000 footprints are not transferred to the solder paste layer … it is expected that there will be a message about the impossibility of converting a particular place or an error will appear … otherwise this the converter is suitable only for small projects … and if the user does not have the opportunity to compare with the original, then he will not know about the errors at all and will send these boards to production)
Place a Net Label in Altium schematic with text \ŽČÓČĎÉŠŮÚŤŽÝ
using “Times New Roman”, “Arial” or “Tahoma” font (the back slash might need adjustment to get the overbar).
Does it look correct to you?
No one should expect importers to be perfect. There’s always going to be differences. Importers are not guaranteed to know about all other software’s features. You should always check boards before sending to production. Especially on imported boards, check every square cm of it.
then why is it needed at all? conversion in the opposite direction starting from version 5 did not have such problems and does not have … if it is not perfect, then at least 99 percent coincides with the source …
If it’s not needed, don’t use it? If KiCad to Altium importer works better on your boards, that’s great.
conversion in the opposite direction (kicad–>altium) starting from version 5 did not have such problems … if it is not perfect, then at least 99 percent coincides with the source …
Hard to believe. I only have experience with eagle–>altium import, and the result was not usable at all. Altium did a very bad job at that attempt. (admittedly it was 6 years ago).
And my personal opinion on importing from a different CAD-package (frequent forum users can skip this, I have written it enough times): nice to have, but I never would recommend everyone to rely on that. It’s impossible to implement a 100%-conversion from different CAD-packages, where the CAD-packages have different feature-sets. Some features still can’t be “emulated”.
I can give you examples for Eagle–>Kicad (because I know both good enough).
And because Altium has much more features there are much more possibilities something is imported different compared to the original.
So my conclusion is: a 100% import is not possible, even if all kicad-developers work with 110% (==24hours a day, and additionally in the night). So if this ultimate goal is not reachable at all → better spend more of the restricted development-time on kicad-features than on improving the import filter. Import is only necessary once, but more/better features; more reliable kicad program is useful for all.
(And because someone will always understand it falsely: I don’t say to stop work on import. I only want to show where I see the priorities from my point of view)
the question here is not 100 conversion … it would be clear that some trace or component or part is simply missing, but when after the conversion you change components or layers, you cannot determine errors … the whole point of the converter is the ability to work with projects from other systems … and for Linux users this is generally the only option
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.