Folding Forum

I generate via solar, so I don't worry about being frugal. Me too :!: ... but the Electric company still charges me more between 1600 and 2100 (local) in an attempt to get me to offload things during their peak hours, depending on my annual total generated vs. used. They're trying to discourage roo...

If you have a CPU with AVX and you get an assignment you cannot complete when running 24x7 please report it on the forum along with the thread count. The project owners are interested in COMPLETED WU as much as you are. They can easily limit assignments to low thread count CPUs. That may mean you ge...

tcphillips wrote:Re:
Spending on gear -- who on earth can afford a new GPU board these days..?

I seem to have been lucky. I have a number of half-height systems with 4-way CPUs and GTC1050Ti s. They've been running windows 7 or 10 for years and years. I wonder how long that can last...

I have not run 18436 so I can only guess. (There might be something unique about 18436)

How many % of the current WU have you completed? (The viewer accesses the information YOU have completed...perhaps in 5% or 10% segments.)

And if you're running Windows, you still have to run all GPUs from a single logged-on user. (blame Microsoft)

FAH used to be able to us a proxy server. If you have (written) permission from the owner of the machine you might be able to convince somebody to configure a proxy server that allows connections, but FAH is licensed for "AT HOME" installations when you can open your firewall to allow uplo...

It has been tried, unsuccessfully. I do not think either software has changed enough to make them more compatible. I agree, and it would still require a lot of effort for essentially zero benefit. Running both concurrently is still possible if you figure out how to get a reasonable balance for you ...

This might be the answer to your question: The FAHCores were originally complied for single precision. It was accurate enough to tiny proteins. Scientists need to study larger proteins but needed to study larger proteins. Recompiling for double precision and using more RAM fixed one problem but prot...

FAH was designed to use the same protocols as http:\\ (probably to avoid blocked connections on win4 or win5). I'm going to guess that using TCP/IP would fix this problem but introduce a lot of new problems, making it expensive to build and test. The original version was intentionally made to look l...

It seems that something isn't working correctly unless the policy on assignments has been changed. Many policy decisions are made at the server level. Many of them can be tweaked at the project level but that depends both on reports from those members of the community who happen to have the particu...

See also viewtopic.php?p=353959#p353959

If the only projects available at the time you need do download a new assignment don't happen fully utilize your hardware would you rather it be idle or inefficient? It's hard to find a single answer that will satisfy everybody.

FAH needs a blend of projects for fast and slow hardware but that depends on what the scientists are studying.

Too much of the wrong type of hardware doesn't dictate what the scientists need to study.

If the 4090 would perform twice or thrice a 2080 Ti, I would buy them. Only if they can be used to the fullest extend. At this time, very few (if any) WUs fully load a 3090, 3080, or a 2080 Ti even... I think for folding and deep learning the low end 4000 series of GPUs will be excellent, just beca...

I can envision the legal teams of AMD and NV in long, long uncomfortable meetings. NV uses CUDA as a powerful proprietary money-maker.

promeneur wrote:I assume the nvidia gpu is more powerful than the intel igpu.

Can you run both GPUs? It makes a simple system into a complex one to manage the primary display and you have to contend with more power/heat.