• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

12 core ppd?

T

twoodcc

Limp Gawd
Joined
Nov 25, 2009
Messages
265
so what kind of ppd would i expect with dual intel 2.66 ghz 6-core processors at stock speed?
 
SazanEyes will know the answer to this one. I believe he was running that exact setup before he had to RMA his SR-2 this week.

Edit - from SazanEyes in a different thread:
Good news: I just started bigadv on my SR-2 and the first WU is a 2685. I'm seeing seeing 69K PPD while also GPU folding. For reference, running standard SMP I was seeing about 35K with GPU folding (four clients) and 50K with idle GPUs. I'm running two standard clocked X5650s.
Click to expand...
 
The P2685 on a substantially OCed SR-2 should output over the 100k PPD mark. The P2684 will likely be under that threshold but this will largely depend on the clock frequency, and what if anything else is consuming CPU cycles. If you run a GPU client, the only way it will be worth the power and lost CPU cycles is if the card produces very high PPD. Otherwise, it's probably best to run the system purely SMP. A dual Fermi would be worth it for the additional ~22k PPD, but will consume much more power. The upshot is such a system would be equivalent to a mid-sized farm in total production. These are just guesses on my part.
 
The rig in my signature gets 70K PPD on 2685s, and 58K PPD on 2684s.
 
The "with GPU" PPD numbers above refer to four F@H GPU clients running on two GTX 295s. It seemed like they used about 5% of total CPU. Power usage was about 700-725W if I remember correctly. I tried pausing the GPU clients and the bigadv PPD went up to 74K, but because it takes so long between steps I'm not sure how accurate that number is.



Yes, because of the bonus structure, you have to balance the PPD hit against the overall greater PPD of adding GPUs. My goal was maximum PPD over the weekend before I had to RMA the board. I'm still planning to test more configs as soon as I get the SR-2 back from EVGA (another week or two).
 
SazanEyes said:
Yes, because of the bonus structure, you have to balance the PPD hit against the overall greater PPD of adding GPUs. My goal was maximum PPD over the weekend before I had to RMA the board. I'm still planning to test more configs as soon as I get the SR-2 back from EVGA (another week or two).
Click to expand...
Again this is just guesswork without actually owning one, but I wouldn't run more than two GPU clients on a SR-2. The more GPUs employed, the more CPU cycles are consumed with some WUs like the 353-pointers being very bad in that respect. This affects -bigadv bonuses drastically. Even then, with an SR-2 system I'd probably only run Fermi GPU architecture because the production yields better PPD vs processor cycles consumed.

Much will depend on what processors one installs on a SR-2 and what max frequencies are attainable. All of these factors play a role to determine acceptable trade-offs in lower SMP bonuses vs. added GPU production and the increased power draw. Each setup is different and there will be variations on this theme.
 
Last edited:
Yeah, I realize what I previously posted was ambiguous and reworded what I meant to say...
 
thanks for the replies.

so if i didn't run any GPU folding, and kept the processors at stock 2.66 ghz, i would roughly get 50K ppd with regular smp, and roughly 75k ppd with bigadv units?
 
ill say 55,000ppd regular smp,87,000 for a 2685, 59,000ppd for a 2684, 40,000ppd for a 6701. A 2.66 dual 6 core has double the power of a i7 4 core@4.0 I cut my fram times in half and used the smp Calculator. http://www.linuxforge.net/bonuscalc2.php though the potential of this rig is much better than this.
 
Last edited:
I never got much over 50K PPD on regular SMP. Also, I wasn't overclocked but I did have turbo mode enabled, so 2.79GHz.

lehmann's numbers for bigadv are probably close, but as you can see they're all over the place. It's really dependent on the specific WU, even for regular SMP. You just notice it more with bigadv because they take so long to run.
 
Didn't Parja post that he saw close to 100k PPD with his SR-2 and hex-cores, or am I thinking of A2 WUs...?
 
From another thread:

Parja said:
Not a [H] Folder, but I'll post this just to illustrate what's possible with dual hex cores on an SR-2.

Folding machine in sig munching on a 2685...

141,000 PPD @ 590W = 238.98 PpDpW
Click to expand...

His is at a little over 4GHZ with X5650s.

Edit: His sig rig:
Folding box: Dual Xeon X5650@4.07GHz - EVGA SR-2 - 6x2GB G.Skill Ripjaws DDR3 2000 - Enermax Galaxy Evo 1250W - Dual Prolimatech Megahalems each with dual Yate Loon D12SH-
 
musky said:
From another thread:

141,000 PPD @ 590W = 238.98 PpDpW

His is at a little over 4GHZ with X5650s.
Click to expand...
Yep, I thought I saw something to that effect. So yeah, hex-cores and SR-2 is a great -bigadv combo... :D
 
You must log in or register to reply here.
Back
Top