Universal X86 Tuning Utility V3 - Adaptive Mode Rewrite #2

[JamesCJ60]

Hi all, it's been some time since I last updated you all on the progress towards UXTU V3. So today, I am pleased to share the first demo video of this new lightweight ML-based adaptive mode that will be featured within UXTU V3.

This video shows this new system working alongside the newly released Indiana Jones and The Great Circle on a Framework Laptop 16 (Ryzen 9 7940HS and RX 7700S) that has been kindly provided by Framework, which utilises AMD's SmartShift technology.

As stated in the previous post, this new system can intelligently adjust its TDP limits based on the performance detected in games using parameters you set. In this example, a max power limit of 120w was set with a minimum power limit of 35w.

It is still early, and we're still a good while off from this being released to the public, but I am pleased with the progress this new system is making and how light the model is on system resources.

You may find this video here (for completely free) as it exceeds GitHub's 10MB upload limit: https://www.patreon.com/posts/universal-x86-v3-118319876

This in the video is just one element that will go alongside other functions/tasks, such as helping schedule workloads across big.LITTLE designs more effectively (ensuring that games have full access to the big cores while background tasks are forced to the ecores), dynamic GPU OCs/UVs, dynamically UVing CPUs/APUs, closing bloat background tasks that aren't important, looking at the possibility of installing DXVK in older games for improved performance, etc.

I'm working on it, one problem at a time, to ensure that everything works as I expect before moving on to other things, which is why it’s still a way out.

Again, thank you to everyone for their continued support of this project. It means a lot!