Folding Forum

No, the CPU is not folding. But i could not reproduce the Problem, since i got no p17311 anymore.
I got one p14906 and one 17421 with no problems, the others were just 13428.
So maybe it was a Problem or not...it is gone.
But i realized that there is a huge gap between the different Projects on RX6800xt. Some go for 3.7 M PPD and others are down to 2.4 M PPD.
On some Projects the RX6800xt performs better then a RTX3070, on others worse then a 2080.
Maybe we need some support from AMD to get the best out of this GPU

It was expected, and there is nothing we can do for the time being. So enjoy what you are getting. This is the best it can do for a moment

Now i got a p14907 with the same problem. I made a screenshot to explain it.
https://ibb.co/cCBvCGG

This might be normal sanity checks or whatever those are called.

Yes, every few percent of progress a checkpoint is written, that is logged in the log file. At the same time a sanity check is run on the data generated to that point using your system's CPU. How long that will take depends on how fast the CPU is, and what else is running on it. If the progress passes the sanity check processing continues on the GPU, if not processing resumes using the previous checkpoint.

There is no Problem with the Checkpoints in the log. I had that issue with failed checkpoints on a 2080Ti and this is not the same problem.
It only appears on some of the projects, P17309 has also this problem.
As soon as i change the Mhz or even the Fan Speed in the Radeon Software, the card "wakes up" and return to normal speed. Or i wait 10 min and it "wakes up" on his own.
It must be a Problem with the Radeon Adrenalin, maybe some power saving issue....

Skajaquada wrote:There is no Problem with the Checkpoints in the log. I had that issue with failed checkpoints on a 2080Ti and this is not the same problem.
It only appears on some of the projects, P17309 has also this problem.
As soon as i change the Mhz or even the Fan Speed in the Radeon Software, the card "wakes up" and return to normal speed. Or i wait 10 min and it "wakes up" on his own.
It must be a Problem with the Radeon Adrenalin, maybe some power saving issue....

You could submit this issue to AMD driver team, that is if they even know these cards exist
We can't do anything on our end, since there are no cards to buy. Earliest chance of the lottery might be Dec 8th.

muziqaz wrote:Is anything running on CPU folding wise?
Navi tends to act weirdly with certain projects of there is something happening with the CPU.

Perhaps this is a limitation with the AMD plan to share main RAM as an extention to VRAM. We can hope they identify it and fix it.

It looks like the Navi 21 [Radeon RX 6800/6800 XT / 6900 XT] is another case where FAH need to know the revision as well as the basic pci code.

bruce wrote:

muziqaz wrote:Is anything running on CPU folding wise?
Navi tends to act weirdly with certain projects of there is something happening with the CPU.

Perhaps this is a limitation with the AMD plan to share main RAM as an extention to VRAM. We can hope they identify it and fix it.

It looks like the Navi 21 [Radeon RX 6800/6800 XT / 6900 XT] is another case where FAH need to know the revision as well as the basic pci code.

Indeed. Hopefully Nicholas will use my suggestion, and we will have some on hands testing done

The new 17xxx projects are very nice for my RX6800xt, much more speed than a RTX3070 up to 4 million PPD.
But i think there are only 2 or 3 guys folding with this unicorn GPU, there are only 209 samples in the database.
And i did one third of them...
So i push it a bit and do some more full power tests the next days after rebuilding it, because the watercooling frame just arrived

Yes those huge projects are a treat. And stock levels of the GPUs are non existant

Yep, my lucky acquired 6800XT is running a 17320 with around 3,6 Million PPD.

I guess with enough atoms, this card can be put to good use.

Now my RX6800xt runs a 17321 with 4.5 Million PPD. That is almost the performance of a 2080Ti on the same project.
I think with more atoms the RDNA2 perform better then Nvidia with CUDA.

Skajaquada wrote:Now my RX6800xt runs a 17321 with 4.5 Million PPD. That is almost the performance of a 2080Ti on the same project.
I think with more atoms the RDNA2 perform better then Nvidia with CUDA.

Now now, let's not run away to fantasy land
2080ti is 3rd fastest nVidia GPU

Well, on 17320,17321 and 17322, my R VIIs under win 10 have higher PPD than my Titan XPs under Linux (2.8 Mppd versus 2.5).