• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Question RTX audio


Diamond Member
Oct 24, 2005
Hey folks,

Two questions:

Does AMD have an equivalent?
Can you use a 1650, and another card like a 2080, and ensure the 1650 does the processing?


Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
Im pretty sure the audio follows on the HDMI cable.
So the monitor you want audio played out with must be connected to the HDMI port on the card.
So you can not have a monitor connected to 2080 and expect the 1650 to do the audio.

The only real reason also to have 2 different videocards in 1 system is also because you require more HDMI or DP ports.
My case for example, i have 3 monitors, and a Occulus Rift.
So my main gaming monitor and the rift are connected to my 2080ti, while my other 2 monitors are connected to a 1060GTX, although i could of probably went lower videocard which didn't require a PCI-E power plug, but i was worried my 4K would be too much for it even for youtube.


Senior member
Mar 15, 2011
You mean ray traced audio processing?

That's a focus point on the next gen consoles, but not really on PC. But even it's not on the focus, the technology is there. Steam Audio 2.0 is mostly can do what the new consoles can. Not a perfect match, but it can do the job. Now Steam Audio 2.0 support three ray tracing engines: the default built-in solution from Valve, the multi-threaded CPU-optimized Embree engine from Intel, and the GPU-optimized Radeon Rays from AMD. The drawback here is the real-time results. Embree works very well, but you need a lot of CPU cores to get good performance. The Radeon Rays works on any hardware with OpenCL support, but it only gives good results if the hardware support resource reservation, and only Radeons can do that.

NVIDIA also have a VRWorks Audio solution, but nobody really care about it. The industry is focusing on Steam Audio 2.0, and Valve don't want support VRWorks Audio.

There are some ongoing research on the usage of DXR for audio processing, but Steam Audio 2.0 is a far better overall solution. So most devs are just implementing this, Unity and Unreal Engine also support it.
  • Like
Reactions: Magic Carpet


Golden Member
Nov 13, 2016
The RTX Voice app doesn't appear to have a GPU selector. Shame, unlike NVENC in OBS.

If it is named RTX audio wouldn't it require at least a RTX 2060, not a GTX model like the 1650?
On initial release, yes. Some time later, RTX Voice was updated to run on any NVIDIA "GeForce, Quadro or TITAN GPU" (from the official setup guide).


Diamond Member
Jan 7, 2007
atenra posted this video in the dlss2 thread. it is a presentation from a microsoft dev on DirectML api. it's for machine learning development and shows an example with an anti-aliasing ai processing lower res game render output to generate higher res looking real time frames.

in the presentation the dev states that tensor/vector array units are only needed for the early teaching phase when the ai is processing the examples. once the ai has finished figuring out the algorithm the tensor/array units aren't needed to run the processing.
this generally suggests that tensor/rt cores arent absolutely needed for things like rtx audio or dlss, as demonstrated by the hacks to enable both on old gtx and radeon cards. so much like physx it will likely be a vendor lock in feature.

the good news is this isnt some magical "can only be created by nv" thing, other companies can and are training their own sharpening/aa/audio/etc tools. the only question is if it will be a general use app like rtx audio or locked in to their software programs. amd may not come out with an equivalent but someone probably will.