Tensor cores are a Nvidia thing , and only since the rtx2000 series.
My gtx1060 without any tensor cores works fine on deep prime :).
Cuda cores are calculations , tensor cores are ai acceleration.
If DxO would ever benefit from ai acceleration it would be a nice thing to implement it. Most Intel CPU cores have hardware for it for the last generations , heir arc gpus have ai acceleration cores (xmx engines ) , apple m1/m2 chips have ai acceleration hardware…
I just don’t know about amd gpus , i just don’t know then (i now realize )
The GTX 1060 is a bit outdated. It is (was) one of those graphic adapters, that don’t get too hot. Even the fans seem to shut down when in idle mode.
There Which Video Card? - #30 by Wolfgang I listed some newer models & specs.
A faster one is something like a RTX 3070.
But they are exorbitant expensive, produce more heat, need a strong power supply and in my case also a new motherboard w/ PCIe 4.0 – so not for me.
I have a 35watt rtx 3050ti laptop chip , that in games at least seems to match my older desktop gtx1060. Shows the gains that have been made at the lower end :).
Ting why i ask i PL6 deep prime XD generates artifacts for me on the 3050ti while the normal deep prime seems to work fine like it always has in PL5.4.
Deep prime XD works fine on the same laptop when set to cpu only mode. But… Even the monster laptop cpu i have takes just under 2 minutes for a file
I just downloaded the latest Nvidia driver for my GTX 1050ti. I was shocked to see that my average processing time for DeepPRIME XD fell from around 50 seconds to around 35 seconds. That around a 30% improvement. The speed of processing with DeepPRIME went from around 21 seconds to 15 seconds, also a 30% improvement. That seems too good to be true, but your results after downloading you latest driver seem to indicate a similar 30% improvement when processing those two files with DPXD. This is very exciting, but how is it possible that we both are seeing such a huge processing improvement? I wonder if the new drivers for other Nvidia cards are also having a similar effect.
it was @jorismak who found out about the new driver and details after coming up with questions … Otherwise I wouldn’t have checked the driver then in use and the developing times for DP XD …
(had always OpenCL enabled, the same like in PL5 – no idea if the former driver didn’t make use of it)
For the past year or so (beginning with driver version 5.11), NVIDIA has been working to implement a vastly-improved OpenCL/CLANG compiler internal to the drivers. This was an end-user opt-in post-install configuration choice for “power users” until the most recent release of the driver packages.
There is some (minor) information available on this in the NVIDIA Driver release notes, section 2.5.3.
The new internal compilers also improve/speeds up CUDA operations.
On other apps (Davinci Resolve Studio) I have seen a 25-40% improvement in GPU-specific rendering and video stream manipulation work.
Beginning this week with the release of driver pack 5.22.30, the use of the new compiler became standard for Windows and Linux users. There is no indication when/if the improved driver sets will be available for Mac.
That’s a good question. I am not home right now but will test this later today to see whether I get the same export improvement for DeepPRIME in PL 5 as I’m now getting in PL 6 after updating the Nvidia driver. I’m on Windows 10.
My point about the new drivers is not that there are faster… But that they fixed the nasty corruption and artefacts that DeepPrimeXD was giving. So it turned PL6 from a disaster into something actually nice to toy with.
I think it’s just as fast before on my win11 rtx3050ti.