Advanced profile management is a fun sounding new feature for the popular Linux GPU Configuration And Monitoring Tool (LACT) app.
With the 0.8.0 release this allows you to set various conditions to trigger a specific profile. So you could have a few conditions set for specific games and apps, to then have a profile activate. These profiles can be exported and imported to share too.
There's also a new process monitor and for NVIDIA GPUs you can now see GPU voltage, GPU hotspot temperature, VRAM temperature (GDDR6 only). Better support for the power-profiles-daemon to prevent conflicts and improvements for the Flatpak too.
The developer also left some important notes for the release:
- RDNA4 GPUs now require kernel 6.14 or later for overclocking to work. Older kernels had broken clockspeed control behaviour that LACT contained a workaround for, and this workaround has now been dropped. See #599 for more info.
- There is a new workaround for RDNA3 constantly reporting thermal throttling, it should now be reported only when it's actually affecting performance.
- If multiple Vulkan drivers are available (such as both RADV and AMDVLK), LACT will now let you select between them on the software page.
Source: GitHub
Some you may have missed, popular articles from the last month:
All posts need to follow our rules. For users logged in: please hit the Report Flag icon on any post that breaks the rules or contains illegal / harmful content. Guest readers can email us for any issues.
Huh, interesting. I wonder if the Thermal Throttling bug has anything to do with mangohud constantly reporting thermal throttling for RDNA3
0 Likes
It seems the update has some issues, myself and at least 2 other people on the github were unable to get the daemon running on the Flatpak version (and some other person on the deb). I tried uninstalling, reinstalling, disabling, enabling, removing the daemon… And now LACT is completely borked for me. I don't get prompted anymore to reinstall the daemon, so honestly no clue what to do now. Dev already said in github that they cannot reproduce the problems
0 Likes