iiinova — where your digital self keeps evolving

pixelcluster @pixelcluster.bsky.social · 17/08/2026

The GPU memory management saga continues! Is the physical VRAM size of your GPU really a hard limit, or rather a suggestion? What if you could run games requiring more VRAM than your card even has, and still get okay performance? All this and more in my new blogpost: pixelcluster.dev/VRAM-Overcom...

pixelcluster.dev

VRAM Management Part 2: Beyond the Limits of Physical VRAM | pixelcluster's GPU blog

030

pixelcluster @pixelcluster.bsky.social · 09/04/2026

New blogpost! This time we're fixing VRAM spilling for VRAM-constrained GPUs, especially 8GB VRAM and lower! If you see your game FPS ends up degrading over time due to your GPU running out of VRAM, this should provide a really noticeable improvement: pixelcluster.github.io/VRAM-Mgmt-fi...

pixelcluster.github.io

Fixing AMDGPU's VRAM management for low-end GPUs | pixelcluster's GPU blog

A: You need some kernel patches as well as additional utilities to make use of the kernel capabilities properly.

163

pixelcluster @pixelcluster.bsky.social · 31/01/2026

I wrote up a blogpost! There's been a lot of #raytracing work that went into RADV for Mesa 26.0. In my post, I'll try to detail the "how"s and "why"s of some of the major driver compiler work I did over the last years. You can read it here: pixelcluster.github.io/Mesa-26/ #gpu #vulkan

pixelcluster.github.io

Inside Mesa 26.0's RADV RT improvements | pixelcluster's GPU blog

051