LivingLinux

It will only work for h265, as that is the only hardware decoder available in the Pi 5. So it won't help you with for instance YouTube, as they use VP9 and AV1 (or you can force h264 with a browser plugin). You can try Firefox. https://bugzilla.mozilla.org/show_bug.cgi?id=1833354 Or change some of the flags in Chromium, but it feels as if they keep changing, so you might find other suggested flags all over the internet. You can set the mentioned flags in: chrome://flags/ https://bbs.archlinux.org/viewtopic.php?id=244031

My PBP died, so I can no longer test it myself. You can start by testing the PPSSPP aarch64 AppImage. https://github.com/hrydgard/ppsspp/releases A couple of years ago I had performance issues with the PPSSPP Flatpak on my Phytium D2000, so I built it on Ubuntu 22.04. I'm not sure if I followed the instructions for Qt, or for Linux. So you need to install the prerequisites and I guess it doesn't hurt to install the additional packages mentioned for Ubuntu. https://github.com/hrydgard/ppsspp/wiki/build-instructions#building-with-cmake-other-platforms-eg-linux

PSP emulation with PPSSPP should work for a lot of games. It's a pity we never got a fully working Vulkan driver. God of War is probably too much. At least it was when I tested it many years ago.

I followed the instructions and have OpenGL ES 3 working now. I wasn't able to build Yamagi Quake 2 because of a version issue with one of the packages. I copied a build from the Muse Pi Pro and now I can run it at 1080p, and sometimes it even hits 60fps!

I tried the nightly build, and I'm happy with the result. Vulkan and OpenGL are still running on the CPU, but I'll try newer images in the future. I did a quick test with Yamagi Quake 2. OpenGL 1.4 doesn't even get 60fps at 640x480, but it looks playable. Should run a lot faster once we can run Vulkan on the GPU. https://github.com/yquake2/yquake2/blob/master/doc/020_installation.md This the make command I used: make WITH_GLES1=yes WITH_SDL3=no -j4 I also tested Stable Diffusion XL Turbo with OnnxStream. Image generated in 7m27s. Looks like the CPU performance is as expected. You do need gcc-14 and g++-14. They are available in the repository. https://github.com/vitoplantamura/OnnxStream You need to change the architecture string in CMakeLists.txt. -march=rv64gcv