The difference in game is not that great. You keep the textures you see the most at 4k and downsample the rest to 2k. Or you can compress the textures. I believe you can use the tensor cores to compress the contents of the vRAM. I believe the tensor method is lossless.
Snip
The difference between 4k and 2k is based on the size (and shape to an extent) of the object, the amount of screen real estate it is occupying and resolution that it is being viewed on. it doesn't matter if it is an animation or video game or a still image. Edit: such a blanket statement is wrong. I'm also pretty certain that they already use 4k textures for some models in some games.
This is the part where someone chimes in that they don't walk up to walls to inspect the textures they just play blah blah blah.
Do you not interact with items in video games? Walk up to doors to open them. hide behind objects, Get close to an enemy to melee them etc
Or do you just sit afar and admire the scenery?
Regarding the second part of your post are you implying that the devs for this game don't know about downsampling textures or using mip maps or compressing textures? Or did you just feel the urge to explain that to me?
Last edited: