Nvidia DLSS 1 and 2 antialiasing discussion *spawn*
- Thread starter DavidGraham
- Start date
-
- Tags
- nvidia
The model is the same yes. The amount of hardware required to execute the model is not.
Benetanegia
Regular
Surely, you're speaking relatively to the amount of total units? Cause otherwise, I'm pretty sure that statement would be wrong.
As in the number of tensor cores between varying cards is not the same.
Benetanegia
Regular
Yes, but I don't think more tensor cores are required on larger chips? More TCs just finish the task faster, I'm sure. Or am I missing something?
EDIT: Like, hyphotetically, the number of tensor cores could be limited by drivers to be the same on all cards and the model would then execute the same and not somehow worse.
EDIT: Like, hyphotetically, the number of tensor cores could be limited by drivers to be the same on all cards and the model would then execute the same and not somehow worse.
Generally speaking yes more tensor cores would finish the job faster. The more tensor cores you have the faster you can get through the density of the network. If it’s thin, it’s not going to matter much, but if it’s deep, then having more cores will result in faster performance than less. Bandwidth is also a large factor.
So what I meant was exactly this topic with respect to DLSS being the same across the cards. The answer is yes, it is the same model being run across all GPUs, and that the answer is that more hardware runs it faster (generally speaking).
but to answer your question; I have no clue if larger chips require more tensor cores
. That’s beyond me lol
Last edited:
Benetanegia
Regular
Generally speaking yes more tensor cores would finish the job faster. The more tensor cores you have the faster you can get through the density of the network. If it’s thin, it’s not going to matter much, but if it’s deep, then having more cores will result in faster performance than less. Bandwidth is also a large factor.
So what I meant was exactly this topic with respect to DLSS being the same across the cards. The answer is yes, it is the same model being run across all GPUs, and that the answer is that more hardware runs it faster (generally speaking).
but to answer your question; I have no clue if larger chips require more tensor cores
That makes sense. The wording of your first post made me think you were implying that there was indeed a difference.
Generally speaking yes more tensor cores would finish the job faster. The more tensor cores you have the faster you can get through the density of the network. If it’s thin, it’s not going to matter much, but if it’s deep, then having more cores will result in faster performance than less. Bandwidth is also a large factor.
So what I meant was exactly this topic with respect to DLSS being the same across the cards. The answer is yes, it is the same model being run across all GPUs, and that the answer is that more hardware runs it faster (generally speaking).
but to answer your question; I have no clue if larger chips require more tensor cores
That is not what I asked though. Would a 2080 do DLSS faster/better than a 2070...?
Yah, it should do it faster because it has more tensor cores.
Yes. It has more tensor power. It should complete the network faste
D
Deleted member 2197
Guest
Missed this when the DLSS reviews came out. Google translation ... 
https://www.pcgameshardware.de/Nvid...arrior-5-DLSS-Test-Control-DLSS-Test-1346257/
The only thing I've ever seen NVIDIA write about is FP16 Tensor Ops with FP16 accumulate.
When they presented DLSS at gamescom 2018 they had a slide about DLSS where they say that they have "~500 Billion FP16 Ops". But that might be for training and not for execution. The number 500 occurs in the volta whitepaper, "DGX Station is a whisper-quiet, water-cooled workstation that packs four NVIDIA Volta-powered Tesla V100 GPUs delivering up to 500Tensor TFLOPS for deep learning applications."
From "NVIDIA DLSS: Control and Beyond" :
"Our next step is optimizing our AI research model to run at higher FPS. Turing’s 110 Tensor teraflops are ready and waiting for this next round of innovation."
From "NVIDIA DLSS 2.0: A Big Leap In AI Rendering"
"With Turing’s Tensor Cores delivering up to 110 teraflops of dedicated AI horsepower, the DLSS network can be run in real-time simultaneously with an intensive 3D game"
In the Turing whitepaper they say that they have "113.8 Tensor TFLOP of FP16 matrix math with FP16 accumulation". So i guess that is number that they reference.
But I mean, if someone has proof that it's only using INT8 I guess everyone would be happy to see it?
When they presented DLSS at gamescom 2018 they had a slide about DLSS where they say that they have "~500 Billion FP16 Ops". But that might be for training and not for execution. The number 500 occurs in the volta whitepaper, "DGX Station is a whisper-quiet, water-cooled workstation that packs four NVIDIA Volta-powered Tesla V100 GPUs delivering up to 500Tensor TFLOPS for deep learning applications."
From "NVIDIA DLSS: Control and Beyond" :
"Our next step is optimizing our AI research model to run at higher FPS. Turing’s 110 Tensor teraflops are ready and waiting for this next round of innovation."
From "NVIDIA DLSS 2.0: A Big Leap In AI Rendering"
"With Turing’s Tensor Cores delivering up to 110 teraflops of dedicated AI horsepower, the DLSS network can be run in real-time simultaneously with an intensive 3D game"
In the Turing whitepaper they say that they have "113.8 Tensor TFLOP of FP16 matrix math with FP16 accumulation". So i guess that is number that they reference.
But I mean, if someone has proof that it's only using INT8 I guess everyone would be happy to see it?
Turing added int8 and int4 modes to tensor cores. int8 is double fp16 throughput and int4 is double int8 throughput.
Man from Atlantis
Veteran
Is there anyway to launch Minecraft Windows 10 on Nsight?
I ran Control but it pauses anytime i alt-tab. So I can't capture an ingame frame on Nsight to analyse.
I ran Control but it pauses anytime i alt-tab. So I can't capture an ingame frame on Nsight to analyse.
Last edited:
Maybe I was a bit unclear, but I meant in the context of DLSS (and hence why I only named DLSS examples). I will try to be more clear going forward.
