As of today, your folding GPUs just got a big powerup! Thanks to NVIDIA engineers, our Folding@home GPU cores—based on the open source OpenMM toolkit—are now CUDA-enabled, allowing you to run GPU projects significantly faster. Typical GPUs will see 15-30% speedups on most Folding@home projects, drastically increasing both science throughput and points per day (PPD) these GPUs will generate.

 
Full announcement, details, and performance charts can be seen at https://foldingathome.org/2020/09/28/foldingathome-gets-cuda-support/
 
This is a public release. Nothing needs to be done as long as you are running recent or current NVIDIA drivers. If you have an NVIDIA GPU then the FAHClient will automatically download an updated Core 22 when you receive a CUDA-enabled project. It's worth noting that if you are using a computer dedicated entirely to NVIDIA GPU folding then CUDA performance under Linux is significantly higher than under Windows 10. 
 
Seeing one person's 1080 Ti getting up to 2.5m PPD already under Windows, which is insane given that card would be lucky to get 1.2m PPD a year ago. It will likely take a few months before CUDA projects replace most of the standard OpenCL projects, so the higher performance will take time to stabilize.

Thanks for the post Kougar, good info! 

I had been running the beta WUs and they had been giving my 2080S a 33% uplift and the 2080ti a 38% uplift in Linux

turbomadman
Wow this is great news. May be worth checking the "B" Stock for older cards to fold. 
Thanks a bunch kougar.

Sounds like a plan :D

Aye, the beta flag is for people willing to closely monitor hardware and invest serious time into tracking down issues. I was very surprised F@H progressed to a public release so fast.
 
As a side note, the new CUDA projects are utilizing more of the core hardware, people are reporting higher power draw/temps. So if anyone experiences instability I'd recommend some stress testing with OCCT. It's free and includes a built-in artifact scanner: https://www.ocbase.com/

Yay, now I got a shot to get 30 mil on my 2060. 

Getting about 2.4 mil. from two Gtx 1070's. So I'm not that mad about not finding a Rtx 3080.(Yes I am but this lighten the pain)

Okay, this is very preliminary as the variations are still pretty wide across projects. Best case I'm seeing 3.5m, worst case 2.5m with my 1080 Ti. Kind of crazy as a year ago 3.5m would've been 2080 Ti territory.  The lowest I've seen is 2.5m PPD which is still a nice jump from 2.1m last month, or 1.2m from a year ago. Going to have to wait and see what the weekly/monthly average works out to be, hopefully F@H can keep those CUDA projects rolling out in heavy supply!

I went from 1.5 mil/day to 2.5-3 mil/day. Free performance!

Went from 1.1 million PPD on a RTX 2060 to 1.7 million so all good. Got another 1060 3GB to add which I think was getting 250k PPD so looking forward to see that one.

Wondering what a pair of GTX 760s would get.  Might fire them up.  

Bummer I'm stuck with two 780 Classified Hydro Coppers.  Would have been nice if they were also on air.
 
Either way, this is great news and my 2080 Ti Black is currently singing to this one Core 22 project. lol

Caffeinated_Sarcasm
I don't know if it's the update or if FAH took off the kid gloves for me, but I've noticed a bit more GPU utilization, a fair amount more heat, and more complicated WUs.
PPD went through the roof, too.

When NEVIDA unlocked the cuda cores it made a big jump in folding production . Have you got your pass key in order to get the bonus points?

Caffeinated_Sarcasm
I don't know if it's the update or if FAH took off the kid gloves for me, but I've noticed a bit more GPU utilization, a fair amount more heat, and more complicated WUs.
PPD went through the roof, too.

That would be the CUDA projects. They are the most demanding workloads I've seen. My 1080 Ti is hitting power limits even at 120% and won't maintain 2050MHz because it can't get enough power to do so... never happened before with OpenCL projects or games.
 
These CUDA projects will really test if people's cards are stable or not.  Alas a bit of bad news, since my last post five days ago my PPD average has dropped to a smidge over 2.5m PPD. Still pretty nice, but not seeing anymore of those insane 3.5M projects I occasionally received at the start.