I haven't been to F@h for a couple of years. DartMouse wrote a series of Linux scripts to deal the various bugs. He hasn't earned CURE in over a year. I don't know what they do or how they work. I merely offer them for your perusal, caveat utilitor:

https://drive.google.com/drive/folders/1Vvckgu5SK_hVM7NlCfCXZDDs2vaVI99o

P.S., I would think that this monthly chart would convince Chodera et alia that they need to change their ways, guess not:
https://folding.extremeoverclocking.com/team_summary.php?s=&t=224497

...
P.S., I would think that this monthly chart would convince Chodera et alia that they need to change their ways, guess not:
https://folding.extremeoverclocking.com/team_summary.php?s=&t=224497

I would guess the massive decline in CureCoin's value is mostly responsible for that. There's certainly not a decline in total compute power at present.

The recent influx of users has brought with it a lot of developer interest in the project, hopefully they can take advantage of that and fix bugs plus improve performance.

---

Team USA forum | Team USA page
Join us and #crunchforcures. We are now also folding:join team ID 236370!

on theregister.co.uk there is a notice saying that Folding@home has broken the exaflops barrier and is now the fastest computer on the Earth. Is this possible?
Tullio

---

on theregister.co.uk there is a notice saying that Folding@home has broken the exaflops barrier and is now the fastest computer on the Earth. Is this possible?
Tullio

It is possible, that according to their benchmarks the connected hosts computing power reached that level. However I couldn't get a single workunit for 1,5 week before I abandoned it, so many GPU can be counted towards computing power without getting a hold onto a work unit.

Of course I do not know, how they exactly measure computing power, but after the announcement of Nvidia, and the joining of many Seti guys, a single user with a somewhat outdated GPU (Rx480) I really felt useless there. Now I upgradet to 2070 OC, but never tried the app again. It was really annoying to pause-start-pause-start to reset the counter, just to see, there aren't any WUs.

I have been running SETi and using GPUGRID as my backup project for quite some time.

Since the demise of SETI I have been running GPUGRID for my GPUs and rosetta on my CPUs and all has been fine. I have been running 2 WUs (one each) on my 2 nvidia 1070s and 9 wus on my 12 CPU AMD computer (leaving 1 CPU and 12 Logical CPUs idle). That gave me 2 GPUs busy and 2 WUs "in the wings" ready to run. BOINC 7.14.2 Win-10.

BUT

Since the introduction of ACEMD 2.10 I am only running 1 WU (on GPU 0) nothing on GPU 1 and 3 WUs "in the wings".

I have plenty of memory (32Gb total) plenty of page file (64Gb) and plenty of free disk space (232Gb) as well as "13 process slots". I have scheduled 4.5 days of work plus .1 day of "additional' work".

As far as I know the only thing to change is the move to V2.10 .

any ideas on where to look to get both GPUs running would be appreciated

Ed Frybarger

That is how I am rolling

CPU to rosetta, and GPU to GPUGRID. seems to be working well. rosetta is very memory hungry, eats pretty much all my 16GB

[quote

any ideas on where to look to get both GPUs running would be appreciated

Ed Frybarger[/quote]

What flavor of linux are you running on the Dual Nvidia GTX 1070's system.

I am running Ubuntu 18.04 (for now), with dual Nvidia GTX 1070 ti's. Both running well.

You are also running an old Nvidia Driver (390), I recommending upgrading your driver.

For the Linux hosts and the 2.10 app, the minimum driver level is 410 for the 1070's.
https://www.gpugrid.net/forum_thread.php?id=5002&nowrap=true#53704

running win-10 with most current nvidia driver 445.75.

BOINC is set to use 100% memory and 75% page/swap space.

also ...

GPUGRID will run 2 NVIDIA GPU's (0 and 1) IFF I suspend rosetta. (now there's a twist ... some sort of memory shortage?? I shut rosetta down to 4 cpu's and it didn't change the GPUGRID problem at all)

Ed F

oops double post

Ed F

Would you please consider discussing non-Folding@home stuff in a new thread? Thanks.

so many GPU can be counted towards computing power without getting a hold onto a work unit.

That's not how it works. Like with BOINC, the processing power estimate is based on actual returned work, not registered hardware in the system. So yes, they did break that barrier and if they managed to get enough servers up to completely saturate every connected host, it should be considerably higher still. See https://stats.foldingathome.org/os

I've never experienced that an OpenCL app could draw as much power as a CUDA app does.

It's true only if you use Nvidia gpu...

Ed, You may need to add this line to your cc_config.xml file:

<use_all_gpus>1</use_all_gpus>

Where can I find information on Folding@home resource usage in BOINC terms? For example, how many virtual CPU cores does it use? Does this vary with the Folding@home Power setting?

So far, it seems that running both BOINC and Folding@home at the same time slows down the BOINC work quite a bit, due to the number of background processes being much more than the number of virtual cores.

So far, I'm only running CPU work for Folding@home, and plan to continue this until they have a steady enough supply of GPU WUs that there will be little variation in the fan noise from my computer.

For example, how many virtual CPU cores does it use? Does this vary with the Folding@home Power setting?

Yes. Moreover you can configure it manually if you select Configure -> Slots tab -> CPU -> Edit -> set/select the number of CPU threads.
It shouldn't be a prime number, except for 2 and 3 (so it shouldn't be 5, 7, 11, 13, 17, 19, 23 etc. Fahcore will reduce the number of its threads by 1 if you set one of the forbidden numbers.)

So far, I'm only running CPU work for Folding@home, and plan to continue this until they have a steady enough supply of GPU WUs that there will be little variation in the fan noise from my computer.

You can set up BOINC to stop GPU work when the FahCore_22.exe is running by adding it to the exclusive GPU apps (lower pane) in the options menu.
FahCore_22.exe is located at

%AppData%\FAHClient\cores\cores.foldingathome.org\v7\win\64bit\Core_22.fah\FahCore_22.exe

I've never experienced that an OpenCL app could draw as much power as a CUDA app does.

It's true only if you use Nvidia gpu...

It's obvious. If I used other GPU, I couldn't run a CUDA app at all, so my comparison couldn't be done.

Is there any config available for Folding@home to enable a Turing card?

Is there any config available for Folding@home to enable a Turing card?

I am running both a GTX 1660 Ti and an RTX 2060 on Folding (Ubuntu 18.04.4). No special configuration is necessary, just a recent driver.

PS - On Ubuntu, you have to install the OpenCL driver. It is not included in the Nvidia driver. That isn't necessary for Windows of course (unless it is).
sudo apt install ocl-icd-opencl-dev

Is there any config available for Folding@home to enable a Turing card?

I couldn't add the GPU though the graphical user interface with the v7.6.9, as it always gave number -1 for the GPU slot.
So I edited the %AppData%\FAHClient\config.xml file, I added the following line:

  <slot id='1' type='GPU'/>

before the last line:

</config>

Now it works.