• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

I Need some help

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

||Console||

Member
Joined
Jul 26, 2004
Its been a while from the last time I ran FAH on my own system I had a few clients pumping in some numbers but they seem to have stopped a while ago .
So I have my 4670k running @ 4.5 right now . It auto-ed to use 3 cores on my CPU and 1 for the GPU I guess ? (win10)

FAH.JPG

I also have a old i7 920 @ stock that I have Ubuntu installed and FAh running on . any specific settings I should be using on these setups ?

Im getting 19k ppd off the haswell dont know for the 920just turned it on
 
Last edited:

HayesK

Member
Joined
Oct 11, 2008
The client sets aside a core/thread for each GPU slot. If you are not going to fold on the GPU, you can change the CPU slot to use all the cores/threads.

My GTX670 make 80-90K ppd each with linux client, Ubuntu 14.04. Likely 5-10% less in windows.

Be sure you have a passkey to receive the QRB. Only need one passkey and needs to be entered in all the clients. Need to complete 10 work units to qualify the passkey for the QRB, and maintain 80% success rate for wu (complete without error and return within deadline).

The Advanced Control interface allows addition of parameters for individual slots and remote control/configuration of FAH clients running on other hosts in the same network. Access from the menu on the FAH client icon in the tray.

HFM is good tool for monitoring FAH clients, local host and remote hosts on the same network.
 
OP
||Console||

||Console||

Member
Joined
Jul 26, 2004
That passkey sure makes a diff on the point value , If im goiong to jump 50k ppd .

I might have access to a 660gtx that isnt being used would it be better in the win10/haswell system or in the i7 920 system ?
 

HayesK

Member
Joined
Oct 11, 2008
My preference for the extra GPU would be linux, but, you need to be aware that the linux core_17 was never patched to fix a slowdown issue which affects Fermi and Kepler cards with drivers later than v327. Linux driver v304 works well on these older cards and is easily installed.

The GPU folding does not use much CPU resources, but fully loads one thread, and runs fine on a hyperthreaded CPU. The 4 core / 8 thread CPU could then run SMP 4 or 6. My old 920/930/950 hosts have 2 or 3 GPUs folding in linux. My 660Ti make 75-80K ppd, and 650Ti-boost make 40-45K ppd. Suspect the 660 will be somewhere in between.