I took a break because I was burnt out. I'm working on it again! I haven't posted any updates lately.
I've been fixing the code to work on 3 different integrated GPU computers I have and the 1 discrete GPU computer I have. I also recently added emulated variable shading rates back in for GPUs that don't support that feature (for a performance option on older GPUs). I'm currently working on improving voxel bricks raymarching of the 64tree (or tetrahexacontree) using floating point math "tricks" (not my article: https://dubiousconst282.github.io/2024/10/03/voxel-ray-tracing/) and fixed point integer math DDA brick stepping for speed. I'm still planning on implementing dynamic pyramid-traced (similar to cone-traced) global illumination with perhaps some illumination data caching mechanism if that's too slow. I also plant to implement dynamic / sparse uploading of voxel bricks to GPU in order reduced GPU RAM requirements (ie only voxel bricks uploaded to GPU that have been seen). Probably more information than you wanted or needed to know. :)
Thank you for your interest!