At this zoom level KiCad 8.0.8 (Win11) runs perfectly smooth:
However, once I start zooming in the performance degrades a lot, like 10fps max:
Can someone confirm this? Or is this just my harware config? I’m not sure if this is report-worthy on gitlab.
If you can attach the project on GitLab, that would be useful, with your hardware configuration info.
You can also try disabling netname updates in settings.
Disabling net names fixes the performance issues
I’ve tested it on a PC with marginally better specs (nVidia RTX A2000 vs my AMD RX580).
The nVidea GPU only has 30% more performance in theory. Not sure why it performs so much better with KiCad. Is AMD known to have problems with openGL? Or is it due to the nVidia being a Quadro?
What’s your CPU+GPU configurations?
i7 8700
Radeon RX580
The CPU usage is pretty low on all cores though, only the GPU usage is ~95%
And what’s the CPU on the “nVidia RTX A2000” PC?
an i7 from the 11th gen, I can later check the exact model number if you need it
There’s been similar reports when using AMD CPU + Intel ARC GPU and AMD CPU + AMD GPU.
I think it’s worth contacting AMD about this issue.
See:
Hm, at home I have an ancient GPU (GTX 970) with a 12 core Ryzen and it runs perfectly fine. The issue seems to be the AMD GPU. Since this is at work, I’ll probably just get a new GPU.
If I can provide anything that helps you to understand the issue, I’m happy to help.
I don’t develop the drivers though, so it’s best to provide all the info to AMD.
I googled a bit and it seems to be a “known thing” since at least 10 years. I don’t think AMD will bother doing anything here
Maybe they’ll reconsider if there’s enough reports.
I’ve upgraded to an RTX 3060 and now KiCad runs perfect. So it was indeed the graphics card
I conducted an experiment on the performance of kicad on different platforms. The fastest option was flat pack and the slowest was win 11. The hardware itself is the same, the only difference is that the drivers in the operating systems are all fresh. I was surprised that the system package version 9 from the pp repository works slower than flat pack. And there is no linear dependence on which functions work faster, or rather, there is no noticeable difference, but some are obviously and noticeably faster.
What CPU+GPU configuration? What was KiCad was showing in Platform and OpenGL lines of version info? X11 or Wayland?
Application: KiCad PCB Editor x86_64 on x86_64
Version: 9.0.0, release build
Libraries:
wxWidgets 3.2.6
FreeType 2.13.3
HarfBuzz 9.0.0
FontConfig 2.15.0
libcurl/8.11.1 OpenSSL/3.3.3 zlib/1.3.1 libidn2/2.3.7 libpsl/0.21.5 nghttp2/1.64.0
Platform: Freedesktop SDK 24.08 (Flatpak runtime), 64 bit, Little endian, wxGTK, X11, cinnamon, x11
OpenGL: AMD, AMD Radeon RX 6400 (radeonsi, navi24, LLVM 19.1.7, DRM 3.59, 6.11.0-17-generic), 4.6 (Compatibility Profile) Mesa 24.3.4 (git-769e51468b)
Build Info:
Date: Feb 21 2025 22:29:54
wxWidgets: 3.2.6 (wchar_t,wx containers) GTK+ 3.24
Boost: 1.87.0
OCC: 7.9.0
Curl: 8.11.1
ngspice: 44.2
Compiler: GCC 14.2.0 with C++ ABI 1019
KICAD_IPC_API=ON
On this machine was tested for comparison and ppa ubuntu system build
cpu 7700 ryzen
