Re:Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs
by dijit on Tuesday January 07, 2025 @12:12PM
(#42624744)
You’re only thinking of the final raster framebuffer, there are multiple raster and shader stages. Increasing the native output has an nearly exponential increase in memory requirements.
Re:Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs
by Macha on Tuesday January 07, 2025 @12:01PM
(#42624625)
My understand is between double buffering and multiple sets of intermediate info for shaders, you usually have a bunch of screen size buffers hanging around in VRAM, though you are probably right that these aren't the biggest contributor to VRAM usage in the end.
Re:Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs
by symaxian on Tuesday January 07, 2025 @06:27PM
(#42628994)
Shadow maps are a good example, if the final rendered image is 4k you don't want to be rendering shadow maps for each light source which are only 1080p else your shadows will be chunkier.
I think the textures and geometry would have the same resolution (or is that not the case? but in 4K if you walk closer to the wall you'd want higher texture resolution as well anyway, if the graphics artists have made the assets at that resolution anyway)
8K screen resolution requires 132 megabytes of memory to store the pixels (for 32-bit color), that doesn't explain gigabytes of extra VRAM
I'd be curious to know what information I'm missing