Folding Forum

You don't understand my approach, I'll try to figure it out on my own. Your only recourse is to not shut down your computer while the client is running. You must shut down folding first, since the exact GPU state isn't saved when you shut down the OS, so if you have the client still running when yo...

Do we have any preliminary ideas yet of possible speed increases...?

Anything new on the AMD/HIP front...? It's been some moons now since the last update.

After a bunch of searching online, it looks like a lot (maybe most? not all...) AM4 motherboards shipped from the factory with CSM enabled, which also means ReBAR is disabled. It also looks like most (maybe not all?) AM5 boards shipped with ReBAR fully enabled. On my really rather old core i9 7900X...

The client is in the process of doing just that when Windows comes along and just terminates the processes. That leaves files open and partly written, or corrupted by the time the restart is done. There's an easy fix to that. Ideally you'd keep two sets of checkpoints and write to them in an altern...

arisu wrote: ↑Wed Jun 25, 2025 1:38 am You must be using the older v7 client. Try the newer v8 client

Righto... I discovered this myself just now. Am reading up on the new client as of this moment!

Thanks!

Create 2 resource groups, both having just that GPU ticked Just pretend I have no idea what a resource group is or how I create one. ;) Using Windows client btw. It uses "slots", and when the GPU is used in one of those, you can't re-use it again in another slot. So I assume you're talkin...

muziqaz wrote: ↑Tue Jun 17, 2025 4:42 am AMD can run multiple WUS on one GPU on both, Linux and Windows, as long as GPU supports hardware scheduling. And obviously it is enabled.

How does one utilize this? Thank you!

That is the reason I always write guesstimate of PPD increase, since 60% of performance increase is not 60% of PPD increase. 60% of performance increase is waaaay more in PPD increase, but we don't know exactly how much, since HIP FAHcore is taking sweet time to be released :/ That sounds really ra...

Joe_H wrote: ↑Sat May 31, 2025 2:09 am Just a guesstimate.

Ah, ok, thanks... So it's just a matter of being patient, then.

Seems to be getting about 5x the ppd of my M1 mini. 540-1040k ppd so far. That's quite impressive for the power consumption and only 10 cores/threads... I'm getting an aggregate 1+ million PPD folding on 14 cores/28 threads out of the 16/32 of my AMD 7950X3D CPU (split into two groups), and burning...

Output ranges from 1m to 2.6m PPD in OpenCL. Probably double that with HIP Is that a guesstimate of what a theoretical HIP core might accomplish, or is there actually some way to enable HIP support in F@H? The way OpenCL is absolutely stone dead, should not F@H do something more to move off of this...

What software can report how much PCIe bandwidth is used by a GPU? Genuinely curious here! Windows task manager will do it for storage drives, but not GPUs...

My AMD Radeon 7900XTX strongly favors a particular series of parkinson WUs - it's the only project which gives even half decent performance for the wattage consumed (over 10M PPD at times with some of these WU - at 300-350+ watts consumed.) Previously I didn't favor any particular cause - it didn't ...

I’ve been playing with BOINC with my CPU’s. It’s cpu control is a much needed improvement that F@H should adopt. I last ran BOINC ten years ago at least, I don't really remember anymore. But at the time, you could set CPU usage as a percentage, and it was actually really accurate. That always fasci...