• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Microsoft Makes DirectSR API Available to Developers as a Preview

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,771 (7.42/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
Microsoft on Wednesday made the DirectSR API Preview available to game developers through the DirectX GitHub. The API aims to simplify implementation of super-resolution technologies with games. A super-resolution technology renders your game at a lower resolution that your display resolution, and upscales it to your display resolution using intelligent upscalers that attempt to reconstruct details lost to the process. This yields a significant gain in performance from the lower render resolution, and lets you max out game settings—something you could only do if your hardware was up to it. By default, DirectSR provides an upscaling technology based on AMD FSR 2.2, but the API looks for the best available technology available with the graphics drivers that are supported by the game. All three discrete GPU makers responded positively to DirectSR.



View at TechPowerUp Main Site | Source
 
Good. We need a unified standard. The DLSS-FSR tug of war is tiring.
 
Good. We need a unified standard. The DLSS-FSR tug of war is tiring.
It's not a standard algorithm though, it's just a way to automatically choose one of the existing ones, with a fallback to FSR 2.2 by the sounds of it. My concern is how it determines what the "best available technology available with the graphics drivers that are supported by the game" is. Does that mean it'll always choose DLSS on an Nvidia card? Will I still be able to change it manually if I prefer FSR in that particular game?
 
It's not a standard algorithm though, it's just a way to automatically choose one of the existing ones, with a fallback to FSR 2.2 by the sounds of it. My concern is how it determines what the "best available technology available with the graphics drivers that are supported by the game" is. Does that mean it'll always choose DLSS on an Nvidia card? Will I still be able to change it manually if I prefer FSR in that particular game?
Considering that 99.9% of Nvidia users keep going on about how much better DLSS is than anything else, I don't see a problem if "the best available technology" means DLSS on Nvidia, FSR on AMD or Nvidia Pascal and XeSS on Intel.
 
Does that mean it'll always choose DLSS on an Nvidia card? Will I still be able to change it manually if I prefer FSR in that particular game?

This is a developer choice. The API support enumerating the variants, and the developer chooses the one to activate
 
If developer makes the call to select the most appropriate upscaling technology, then I really don't see any purpose for this extra layer of software. FSR runs on most GPUs anyway, so I am thinking what sorts of value add we are getting here. I rather they reduce software bloat to lessen the load on the CPU.
 
If developer makes the call to select the most appropriate upscaling technology, then I really don't see any purpose for this extra layer of software. FSR runs on most GPUs anyway, so I am thinking what sorts of value add we are getting here. I rather they reduce software bloat to lessen the load on the CPU.
Software bloat? I wish you all the strength to survive the wrath of TPU's Nvidia fanbase. :D *grabs popcorn*
 
so I am thinking what sorts of value add we are getting here
Developers don't have to invest resources to add, test and maintain multiple proprietary upscalers
 
its about time
 
This is a developer choice. The API support enumerating the variants, and the developer chooses the one to activate
Ah excellent, that's very sensible and kind of an obvious way to go about it actually...tyvm for the info!
 
Developers don't have to invest resources to add, test and maintain multiple proprietary upscalers
Now If could only get a mGPU implementation like this from Microsoft for DX12. >.>
 
Back
Top