Folding Forum

https://foldingathome.org/2020/07/08/ci ... -covid-19/

In my case a Threadripper crunching with 15 cores it took 3.055 hours to crunch a wu with 221911 atoms and 250000 steps.

The GPU, RTX 2080, is crunching a wu with 182699 and 1000000 steps in 1.99 hours.

Based on the information in this thread.
If all the CPU cores running folding, 21418365 cores, and all the 306000 GPU, number are taken from here https://stats.foldingathome.org/os
had the same mean values as in my case for CPU and GPU I end up with total of 1 protein covering 0.1 seconds is being crunched in 6.77 days on all the GPUs and 1 protein crunched in 10.2 days on all the CPU cores.

During 4 months folding this ends up with finishing 13.46 proteins on all the CPU combined and 17.7 on all the GPU combined so in total 31.16 proteins for the whole Folding@home community.
This numbers is also based on the assumption that all the computer power is up and running all the time but maybe 50% is more realistic number here.

I guess that there is no such thing as enough computer power when it comes to folding?

It's funny to talk about.
For the time you start running your first WUs on a new GPU, to the time, where you retire the GPU, it will probably have folded about the amount of time it takes to type in a character and for it to appear on your screen.

Hopfgeist wrote:...Given the total number of projects I'm not sure that is right, but I cannot find statistics about which project has been finishing how many work units...

There is historical information that is stored on the database but there's no public access. Reason is that it would fully saturate the stats system.

Moreover, in some (rare) cases, Project numbers can be re-used but the trend is to use unique project numbers. Thus, it would be an interesting challenge to figure out if that information even becomes public

BTW, those stats don't count Internal Projects which may be used for data generation to help create Public Projects which means that the true number of Project would be a bit higher than the calculated one.