-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Xbox Series X: AI Custom Hardware Support Offers Unparalleled Intelligence for Machine Learning
- Thread starter bcatwilly
- Start date
It's cool but still a trade off right? It's using the existing shaders to carry out the operations. So if you're doing ML you aren't doing shading in those CUs
That works well for something like a DLSS equivalent as long as the cost of ML time is less than the time saved by the GPU shading less stuff. It might also be necessary if RT really needs to run at lower resolutions to be efficient (as we've seen on PC already)
there are so many new tools for devs it'll be fascinating to see how engines develop to leverage them. ML, VRS, mesh shaders are really big ones that could have huge impact
That works well for something like a DLSS equivalent as long as the cost of ML time is less than the time saved by the GPU shading less stuff. It might also be necessary if RT really needs to run at lower resolutions to be efficient (as we've seen on PC already)
there are so many new tools for devs it'll be fascinating to see how engines develop to leverage them. ML, VRS, mesh shaders are really big ones that could have huge impact
That's my takeaway, but to balance that, you're working with a lower base resolution to balance. But you might not get all the visual effects
No, not presently. Devs prefer to handle denoising with general shaders, possibly because this makes their RT implementation fully DXR compliant instead of relying on NV's proprietary NGX or DirectML with unknown performance on h/w without tensor arrays.
OP
OP
Just a little interesting tidbit that Phil Spencer mentioned in the IGN Unlocked podcast interview with Ryan McCaffrey that the use of DirectML and AI was something that they thought about with the Project Scarlett (Xbox Series X) hardware in terms of its use in the cloud too.