Devs, is it possible to fork the f@h software and make it more GPU efficient so we can really get our protein on?
The F@H software has some closed source parts so it's not possible and I think it's GPU efficient already. The F@H project started 9 years before Bitcoin was released so they should have a good idea how to design software.
I see, if it wasn't closed source though, I wouldn't be surprised if someone from the crypto community could manage a faster more gpu intensive way though. Let's be reasonable, if our gpu's aren't maxed it's unlikely it's fully optimized.
Maybe - but how would you get the work units and deliver the results - that's all from stanford and through their software. GPUs are running in max but it may not for example use a certain portion of your GPU that causes it to heat up a lot - a portion that is just not used for protein folding.