Oh, right on! Got 5 TFLOPS on it…
1 Like
Well I’d assume they have other things to worry about then writing text descriptions 
1 Like
I really should upgrade my folding rig’s CPU though. That poor i5-2500 is maxed out 24/7 and the GPUs are only at 75-80%. Textbook bottleneck. Didn’t have that problem before adding that third 1070.
Time to look for a 2600K.
Sure.
Make sure that you have opencl-nvidia and cuda installed for nVidia GPUs, and opencl-mesa for AMD GPUs.
Yeah that’s what I thought as well, but:
[tarulia@localhost ~]$ dnf search opencl mesa
Last metadata expiration check: 11 days, 15:41:54 ago on Fri 28 Feb 2020 20:33:47 CET.
========================================= Name & Summary Matched: opencl, mesa =========================================
mesa-libOpenCL.x86_64 : Mesa OpenCL runtime library
mesa-libOpenCL.i686 : Mesa OpenCL runtime library87
mesa-libOpenCL-devel.i686 : Mesa OpenCL development package
mesa-libOpenCL-devel.x86_64 : Mesa OpenCL development package
[tarulia@localhost ~]$ sudo dnf install mesa-libOpenCL
[sudo] password for tarulia:
Last metadata expiration check: 1:21:26 ago on Thu 12 Mar 2020 16:07:12 CET.
Package mesa-libOpenCL-19.2.8-1.fc31.x86_64 is already installed.
Dependencies resolved.
Nothing to do.
Complete!
So that’s not it unfortunately 
I do have an idea of what is happening though… So in the System Info Panel it says:
OpenCL: Not detected: Failed to open dynamic library 'libOpenCL.so': libOpenCL.so: cannot open shared object file: No such file or directory
And I can see why:
[tarulia@localhost ~]$ ldconfig -p | grep libOpenCL
libOpenCL.so.1 (libc6,x86-64) => /lib64/libOpenCL.so.1
libOpenCL.so.1 (libc6) => /lib/libOpenCL.so.1
I’ve had this issue with libgamemode before and the fix was essentially a package update providing this:
[tarulia@localhost ~]$ ls -l /usr/lib | grep libgamemode
lrwxrwxrwx. 1 root root 20 Jan 23 15:54 libgamemodeauto.so -> libgamemodeauto.so.0
lrwxrwxrwx. 1 root root 24 Jan 23 15:54 libgamemodeauto.so.0 -> libgamemodeauto.so.0.0.0
-rwxr-xr-x. 1 root root 15192 Jan 23 15:54 libgamemodeauto.so.0.0.0
lrwxrwxrwx. 1 root root 16 Jan 23 15:54 libgamemode.so -> libgamemode.so.0
lrwxrwxrwx. 1 root root 20 Jan 23 15:54 libgamemode.so.0 -> libgamemode.so.0.0.0
-rwxr-xr-x. 1 root root 19448 Jan 23 15:54 libgamemode.so.0.0.0
So I could probably create a symlink libOpenCL.so -> libOpenCL.so.1, but I’m not sure I really wanna fuck around with system libs…
two questions from a first-time folder:
-
I’m going to get my secondary gpu (normally for gaming VM) set up to fold. it’s an nvidia gpu, while the host is amd. would installing opencl-nvidia and cuda (as mentioned above) cause issues?
-
fah is making good use of my cpu, but i have a lot more ram than it’s using. can i configure anything to allow more usage, or does it just not need it?
thx
Ah. Nice to see a level1 team. I usually switch around teams, so I’ll stick my TR3960X / 2070 Super on this team for a while!
Screw human malware.
1 Like
Should work fine. Each driver is only attached to the cards that can use them.
Doesn’t need a lot of RAM. The workunits themselves are pretty small and saved on interval anyway.
1 Like
Yea, enable advanced mode. There are no guarantees but you will be available for larger projects if you do that.
I think all you need to do in the driver area is use the latest version to get the best utilization.
1 Like
just put my compute to team enjoy~
3 Likes
Just joined the team. I don’t game much these days, so my gaming rig has been bored. Now it’s working hard
And my GPU picked up the corona project as the first thing.
2 Likes
hmm …
Just installed the client on my main rig as well. It started just fine, but I shut it down to disable the CPU and enter my identity and passkey.
Now it refuses to start, probably because it thinks the collection server is 0.0.0.0
EDIT : sorted, the old “have you tried turning it off and on again?” trick worked.
Currently doing 1.7 to 1.75 million PPD on the folding rig, 1.4 to 1.43 million on the main one.
2 Likes
Wow, do you have like a zillion cuda cores or something? Mine don’t move nearly as fast, lol.
The room temp dropped noticeably, so I checked the folding rig and found that it’s only running on one GPU right now, with the others having the same problem I mentioned above.
Just checked the folding forum. Apparently the F@H servers just can’t keep up with the amount of requests. I guess that’s a good thing.
If it happens to you too, just wait. You’ll get new WUs eventually. One of my 1070s picked up a new WU as I was typing this post, the other one still hasn’t.
2 Likes
I’ve been seeing that periodically all week. Not enough work to go around, which as you said, is probably a good thing.
1 Like
Mine is having the same problem.
After it finished the first gpu job, it never got a new one. Restarting the client haven’t improved things.
That is probably a good thing.
I’ll let it wait for a job to pop up. In the meantime the cpu can keep up the work
Is it possible to see somewhere how many projects the cpu or gpu has worked on?
1 Like
Not really sure, the best info I know of is on the team stats page, the far right column. Looks like yours has done 14 WUs when the list was last updated.
https://apps.foldingathome.org//teamstats/team232084.html
Just search for your username, or scroll down to team rank #144.
1 Like
Neat.
Well, now the gpu got a work server at least.
But the log says:
WARNING:WU00:FS01:Failed to get assignment from ‘18.218.241.186:80’: No WUs available for this configuration.
It seems kinda weird to me that they could be “out” of work?
Or are they changing the jobs they are sending out, so there aren’t multiple systems doing the exact same calculations?
Some variation to what a work includes seems like something that might make sense?
Not sure what’s going on. I’m guessing that either
- a lot more processing power has recently become available (chinese mining farms temporarily doing F@H to help combat Covid-19 ?)
- they are having server issues or
- we found cures for almost everything
My money’s on #2
2 Likes