Yes, I believe the output of Folding@home is way above that of WCG. Boincstats claims WCG to do 507 TeraFLOPS, whereas Folding@home claims almost 500,000 TeraFLOPS. Comparing FLOPS can be a tricky thing and can be like comparing apples to oranges, but still, Folding@home's output is far larger. Part of that may be explained by higher user numbers. But Folding being done almost exclusively on GPUs is the most important part of the answer, as these are way more powerful than CPUs. To anticipate the question why WCG does not run on GPU: not all problems are suitable to be solved effectively by GPUs.


Also that has to do with CPU vs GPU: there are 3 million CPU cores on folding, but also almost 300,000 GPUs. The CPU cores only provide an insignificant part of Folding's output.


I do not remember, if the average Folding WU takes longer than the average WCG WU. But it surely takes a lot more FLOPS to finish one. Again, also this has to do with CPU vs GPU. So don't expect yourself in the top 25...


People often assume, you can just throw computing power at a problem and then a vaccine or cure comes out. In reality, doing computer simulations is only a small part of a process. It can help, but many more things are needed. Apart from that, IBM does not design projects themselves. They only support projects from scientists who are in need of computing power and approach them.
Besides that, I believe there is such a focus on the Coronavirus (rightly so), that funding is certainly not the bottleneck. If some institution has a good idea for simulations to fight the virus, I believe getting the computing ressources is less of a problem. However, the problems WCG supports, often have trouble to get funding and in many cases could never obtain such large computing resources. Either, because they are ground research, not likely to lead to a major breakthrough soon, or because they are about diseases affecting mostly the 3rd world, and therefore unlikely to draw the attention of big pharma.


Not sure what you mean by that. But if you mean that all the work WCG volunteers are doing could also be done on supercomputers, you are right. It is only a funding problem, many researchers cannot afford expensive super computer time. And I think giving people who are interested in helping, an easy way to do so, is a great idea. Some people here are donating serious money in the form of electricity bills and hardware. And they are also spending much time in building and maintaining their hardware. For me it is a good way to donate money to a good cause - maybe better and certainly more interesting (for techies at least) than donating to a charity.

In specialized calculations, and they are specialized, kind of RISC, a GPU core can do 20 fold that of a CPU core, so Article 1 states that the top 7 supercomputers have around 27 million cores and article 2 states folding@home has around 3 million cores has a certain amount of credence to be more powerful. IBM's world #1 supercomputer has thousands of GPU's in it though. A number of supercomputers have.

Remember HCC... jobs that run multiple hours and done when ported to GPU in minutes, so much so that the setup time on the CPU and wrap up was often longer than the GPU time?

That is nice, but F@h does not have (as far as I know) GPU & CPU temp monitoring, which is essential in the "long run" if you want to do calcs. All my past here & on SETi@home (with GPUs, where I'm top contributor from Croatia) gives me some knowledge about it & credibility to speak about it.

I have fried several GPUs with not monitoring temps. Imagine the loss of the GPUs in this time of shortage of computer parts? So NO, I'm not going to F@h by far.

Sticking up to BOINC related projects.

"Thanks to our AMAZING community, we’ve crossed the exaFLOP barrier! That’s over a 1,000,000,000,000,000,000 operations per second, making us ~10x faster than the IBM Summit!"

If I'm reading the OS (https://stats.foldingathome.org/os) page right, it's actually 1.5 exaFLOPs. I'm assuming they mean the x86 TFLOPS figure.
https://twitter.com/foldingathome/status/1242918035788365830