The forum is mostly just users talking to other users. When improvement ideas are raised on GitLab and they are done so with reasonable tone, it generally goes well I think. Sometimes people get offended when their ideas are not prioritized, though.
1 Like
Iām just throwing Some Spaghetti observations of Kicad & Crashingā¦ on the wallā¦
All related to my usage on Mac from v5 through v7.6
I can Force it to Crash if I do any of these:
ā¢ Rush to Open PCBnew after starting Kicad
ā¢ Rush to Open PCBnew after Closing Eeschema (Schematic)
ā¢ Rush to Open PCBnew after Closing PCBnew
Does not matter if I delete or Not-delete the .ICK file
Iāve learned to always count āSlowlyā to 10 before Opening any Kicad Projects and filesā¦ Video below shows this Crashing is Repeatable, especially with Pcbnew.
Did not bother to show it Not-crashing after counting to 10.
I just accept it as part of Open-Source usageā¦
1 Like
Please report theseā¦ videos of it happening doesnāt help at all. I use KiCad on my Mac all the time and donāt see this problem, so it may be specific to your design somehow.
Ok, now I understand, Iāll only use the gitlab issue site checking there if my particular issue of the moment has been already entered.
Goodbye info forum.
Design does not matter. It can be project with empty Schematic and empty PCB. Very repeatable and has happened repeatably on v5, v6 and v7.
Also, this week, I cleaned out my Mac system and Re-Installed kicad v7.06ā¦
I stopped reporting Bugās on all software - it became more work than I want to bother with. Iād rather be bothered by the crashing than reporting Bugās that may/may-not get fixedā¦ (though, Iām grateful to those that do report themā¦)
If it happens on V5-V7 I suspect a problem with your system, to be honest. There is too much that has changed since V5. If you share the macOS problem report it might be useful.
It could be my Mac, I definitely accept that and, it probably is!
I built a Harley (Chopper) from scrap parts in 1969. It would only run if the Timing (Distributor shaft) was 180Āŗ out of phase. At least one dozen of my (former) Hellās Angel brothers tore the bike apart to figure it out and None could. But, the bike ran and ran wellā¦ Now, thatās real Open-Source (because all the parts were no-doubt Stolen before they gave them to me)ā¦
1 Like
Exactly, I waited years to start thinking seriously about using it and now it is up to par, but still with a lot to fix/add.
sorry for the criticism, but releasing a release with a large number of declared but not working features is not acceptable ā¦ testing on users is a good idea, but not for stable builds ā¦ remember how the stable release of version 7 appeared ā¦
Not sure which features you are referring to in particular, but in general we rely on users to test what will become stable builds. If nobody reports a problem, it might not be found.
In other words: if somebody never tries a release candidate, and then downloads the stable release and is surprised when something doesnāt work for them: I would encourage them to download the release candidate next time and try it.
1 Like
the problem is not in new functions, but in the fact that the old ones stop working, from plugins to built-in ā¦ the consumer needs a stable release that works with the declared functionality and not a testing ground in a stable version
Where did I say I was talking about new functions? We rely on users to test new and existing functions in release candidates.
I think frequent releases is good, but of course the broken already working things have to be corrected first thing.
you must understand that engineering software will primarily work on win and not on mac and linux, and this is related to market share and applies not only to kickads
They normally are, when people report them. People seem to sometimes assume that all the developers use KiCad the same way they do, and so would obviously notice the same things broken as them. This is not the case, because KiCad has thousands of functions and different ways of using the tool.
Built-in functions and plugins are different. It is also up to plugin developers to ensure their plugins continue working with newer KiCad versions, because KiCad does not have a stable plugin API right now.
Agreed.
They must focus more and more on developing and testing on Win because we are āforcedā to use it (sad but true).
The more professional the software becomes the more carefully they must thread through their software.
This is why I though this forum was made forā¦ I throw my issue and see who āsticksā to itā¦ instead I received criticism for criticizing. Now I know better.
I think you are being too harsh on the users who were suggesting workarounds to your issues. They arenāt developers, they canāt actually implement the thing that you want, so instead they are suggesting that you change how you are working in order to accomplish things with the software in its current state. I think this is normal for a user forum. The way you replied saying that their workarounds were unacceptable to you had a pretty negative tone. If you had said āthank you for the suggestion, but unfortunately that option isnāt good for me becauseā¦ā I think the conversation would have gone better.
1 Like
when downloading a stable release, the user expects that at least nothing from the previous version has broken, but it turns out the opposite, and the errors are very gross and it is immediately clear that the programmers simply did not check ā¦ for example, in version 7x1 it worked in version 7x2, it stopped working although both I donāt understand how stable releases could go unnoticed ā¦ I conclude that the developers do not check the stable version ā¦ the second point that annoys thousands of users is the change in the library format ā¦
Please read my reply above again. There is a very clear explanation for why āthe programmers did not checkā, and why you as a user can help by installing release candidates.