I frequently come upon a feature, that i think that is great, but very soon realize that is actually useless. Just now i tried open an DRC marker with doubleclick, it worked, but it was a usless feature, because the entry wasnt marked, so it is as good as nothing…
There is probably put quite a bit of work into things, so please dont stop a few meters before the finish line, as this wouldnt bring a success. Just like building a car with a nice suspension, but putting wooden wheels on it.
Another one: net ties: those can be really helpful, but if you can not place them on an inner layer, it again becomes useless.
The grid dots are also pretty helpful, until you have activated more than 2 layers, because then, they become invisble.
I can work with a few good features, but i cant work with a lot of halfgood features.
Also the basic function of laying down traces is still pretty annoying, if you work in tight spaces and a grid of 0.05mm.
I appreciate that this a vent, but I’ll just point out it is hard to want to help get these things fixed when the tone implies (to me) “why didn’t you anticipate every use case before releasing a (volunteer created) product.” I’m guessing you don’t mean it quite like that, just letting you know how one random fellow user would read your message. It sounds like you’ve come across a couple of bugs and maybe some missing features, which is very frustrating! Some of these things do sound familiar, which may or may not be reassuring.
To confidently address your concerns, we’d need a little more information. Can you include your version info? (About KiCad > Copy Version Info)
For either of these Gitlab issues, you can show support by giving a “thumbs up” (requires a free Gitlab account). The developers as a group do consider the user interest in selecting bugs or features to work on.
I’m not sure what to make of the grid dots concern or the track laying concern. More info/screenshots would be helpful here.
Yes i dont mean it like that. Its just the fact, that there is put much effort in it, but then abandoned before its really finished (as it seems to me). Also when there are bugs that are not fixed, but on the other hand all the symbols are changed, so one has to adapt again, is really annoying.
Just look at the net tie issue: Added 1 year ago…
I cant really say, that it is easier to work with 7.* than with 5.*, which is a shame.
When i exclude a DRC marker, then the next one is selected, but in the window this one is not shown. Even after i click the very same selected marker, it does not jump to it. I have to click another one, and then back to the previous selected one, so its jumps to it.
Is this the same at yours?
Nobody has answered for a couple of days, and I know why. Your tone doesn’t draw positive attention. KiCad has both bugs (unintended behavior) and missing features (something wanted which doesn’t exist) but we can talk about them in several different ways.
As for the DRC marker issue: if you document your case step by step (in the level of clicking UI items etc.) and show and describe exactly what happens, preferably with a zipped minimal example project and all other relevant information to reproduce it following your step by step instructions “blindly”, it will probably be fixed soon after it has been reported to the issue database. Due to the cross-platform nature of KiCad there are quite often small bugs in the UI which appear only on one platform. They will be fixed usually quickly when reported properly. Even if it’s not one platform only, this one sounds – even more so – something which should be fixed.
Another lesson about the tone. No, they don’t become useless. They are perfectly useful for the most usual cases where they exist on outer layers. They just can’t be used in the inner layers. If a feature can’t be used in a certain way, it doesn’t suddenly become useless. Just plain logic, but at the same time about human feelings and attitudes going through communication and causing effects which can’t be reduced to plain logic. If you imply, on purpose or by accident, that you despise KiCad or the developers’ or users’ work, skills or attitude, you will not get much positive response in this forum.
And then technical information. You can upvote True padstacks and via stacks with differing geometries on different layers (lp:#1827233) (#2402) · Issues · KiCad / KiCad Source Code / kicad · GitLab by giving a thumb up there. That’s larger in scope but includes this use case. That’s one reason it hasn’t been implemented: the developers have to decide the direction of development, and features aren’t just implemented like that. Often many smaller features are parts of a larger feature, but that larger one needs planning and much work to be useful and work properly. Therefore many interesting and important features haven’t been implemented yet. The development resources are limited and must be spread thinly over all the needs, and even then only after deciding what is strategically the best feature set, all things considered, and what can realistically be implemented next.
There exists a workaround for inner layer net ties. KiCad doesn’t have a UI for placing pads to inner layers, but it actually supports pads in inner layers. This needs editing some file manually. After a NetTie footprint has been placed to a board and the file saved, you can find the footprint in the file. In a 2-net footprint from the official KiCad library I can find the graphic item and two pads. I can just change the F.Cu layer to the wanted inner layer:
Words like useless kinda piss people off, especially the hard-working devs – a softer tone will get better responses.
Regarding net-tie inner layer issue: yes, I also need the net tie on the inner layers where primary ground planes reside, so I made a tie with three through-hole pads, two round and a center rectangular, and placed them so the overlap. This puts pads on all layers. I use it to tie analog and digital planes, or ground plane and esd ring. Still throws drc exceptions which I ignore.
Alfred. If you need or want perfection or close to it, get yourself a a full Altium seat. Its not expensive either, compared to Cadence, Mentor etc.
It’s an amazing tool, but not also without its frustration at times . Like Kicad. Nothing in complex tools is ever perfect. All you can do is stay cool and try a workaround, ask the community or report bugs.
It’s sort of fortunate for Altium their forums are private, no one would buy it otherwise LOL.
In Kicad, everything is out in the open , and if you have a real problem, you can fix it, or influence to get it fixed. The influence keyword I used here is king .
So, sometimes it takes some effort to write a persuasive argument (especially if your 1st language is not english ! ) . well it does for me.
If you have a problem in Altium, the company can just tell you its fine and go F**k yourself. and that’s what happens over there.
