Folding Forum

Hi! I use my old mining rig for folding, it has 6 P106 GPUs and I can do 1,7M PPD

Welcome to the forums, and happy folding!

https://www.youtube.com/watch?v=_BTPBPUWqXs
Here is my rig

I would remove the CPU slot ... just to get rid of the brown non-spinner on that screen.

bruce wrote:I would remove the CPU slot ... just to get rid of the brown non-spinner on that screen.

Good idea

The browser is using more resources than FAHControl.

The spinning arrows (green) actually use CPU / GPU resources, as well as the browser does.
But if you do want to view the browser client, you can safely block the green spinning arrow elements, with a popup/ad blocker.
That way the entire page remains in tact, but the green rotating arrow will be sent to a 'null-driver', where it won't be processed for display updates.
Saves you a few PPDs.

Or better yet, close both WebControl and FAHControl when you're not looking at them. They don't contribute to the folding process ... they just inform you what's going on.

Does it really make a difference if browser is running with some spinning arrows? I mean like loosing 10% PPD? I can't believe it...

For mining the cheapest celeron processor is enough but for folding it is not enough! First i changed to an i3 and it was still weak, it could only manage 4GPU with good PPD, so now I have an i7-9700 and it's ok I think

Yes it needs one CPU thread per GPU to feed it. RAM usage should be 0.5 GB per GPU + 2 GB for OS. So for your 6 GPUs it is 8 GB.

I see you run Windows which also bottlenecks on x1 risers. If you can switch to Linux like Ubuntu 16.04 that will improve PPD by 50%

foldy wrote:Does it really make a difference if browser is running with some spinning arrows? I mean like loosing 10% PPD? I can't believe it...

On a low end Atom/Celeron processor it makes more of a difference than on a modern Ryzen CPU. But it still makes a difference.

On GPUs, it will just stall a few GPU cores, just like your desktop does; but not as much as when running a screen saver.
It all still uses resources.
The best thing you can do, is set a screen saver to 'blank' out the screen, or even turn it off.
Not only does it save processing resources, but power as well.

Live streaming doesn't help performance either, but it's certainly an alternative to SSH, to see if your PC is still online and folding.
I wouldn't do it just for the CPU overhead, as well as the increase in network traffic...

But to each his own..

foldy wrote:...RAM usage should be 0.5 GB per GPU + 2 GB for OS. So for your 6 GPUs it is 8 GB...

Some Projects with large number of atoms can use up to 2.3 GBs of RAM. My current estimate is that 3 GBs of RAM dedicated to 1 GPU in the system as a starter. Not sure how that will play out with the new version of FahCore_22 (no ETA).

foldy wrote:Yes it needs one CPU thread per GPU to feed it. RAM usage should be 0.5 GB per GPU + 2 GB for OS. So for your 6 GPUs it is 8 GB.

Your numbers are obsolete. The 0.5 GB per WU is on the low side. Yes, there still are some WUs that fit in 0.5 GB but I've also seen some that need 1.5 GB. Notice that we're now seeing projects that have 400K atoms PLUS. I had two of them at the same time. Even with my paging file on SSD, the system was thrashing.

And during the transition between two WUs , which can last for a few minutes, that number will possibly double: viewtopic.php?p=334045#p334045

Project 14201, 453k atoms: