Deep Learning Super Sampling: Difference between revisions

Revision as of 15:40, 7 April 2020

Revision as of 22:06, 7 April 2020

Line 101:

==Architecture==

DLSS is only available on [[GeForce 20 series|GeForce RTX 20]] series GPUs, in dedicated [[AI accelerator]]s called '''Tensor Cores'''.<ref name="nvidia20"/><ref name="tensorcoretensorcore1">

{{cite web

| url=https://www.nvidia.com/en-us/geforce/news/nvidia-dlss-2-0-a-big-leap-in-ai-rendering/

Line 107:

| publisher=[[Nvidia]]

| accessdate=2020-04-07

}}</ref>

===Tensor Cores===

Tensor Cores are available since the Nvidia [[Volta (microarchitecture)|Volta]] [[graphics processing unit|GPU]] [[microarchitecture]]. Their specificity is that each Tensor Core operate on 16 bits [[Floating-point arithmetic|floating point]] 4 x 4 matrices.<ref name="tensorcore2">

{{cite web

| url=https://www.anandtech.com/show/12673/titan-v-deep-learning-deep-dive

| title=The NVIDIA Titan V Deep Learning Deep Dive: It's All About The Tensor Cores

| publisher=[[AnandTech]]

| date=2018-07-03

| accessdate=2020-04-08

}}</ref>