Anyone have Phenom II Folding numbers?

Parja

Parja

[H]F Junkie
Joined
Oct 4, 2002
Messages
12,670
I've got a K9A2 Platinum system that I'm considering throwing a Phenom II X4 into for some more PPD. Anyone have any PPD info on any of the Phenom II CPUs? I can't seem to find much of any info on them anywhere.
 
I've got an X2 550 unlocked to 4 cores. I haven't OC'd it, though. So 4 cores @ 3.1GHz puts out a hair under 5000ppd.
 
phenom II 940 @ 3.5ghz w/ notfred client on 2 VM's = 2800-3000 PPD per VM..
 
That's a lot better than the X2 numbers.

Humm..... if q6600's where not so cheap now.
 
I've got 1 VM running on my Phenom 2 720 x3 at 3.45 Ghz. It makes right around 3k ppd on the VM. It's also supporting a couple 4890s.
 
Kendrak said:
That's a lot better than the X2 numbers.

Humm..... if q6600's where not so cheap now.
Click to expand...


they are only better then the x2 numbers because of the overclock.. the x2 is a phenom II 955 with 2 cores disabled.. if they were running identical clocks it should actually do better then my 940 because at the time i was running a slower NB/HT clock but now thats fixed..
 
They were fine, in their day, you can't really expect a four year old processor to be that fast today. The only thing they're good for now is in GPU boxes, and probably only single/dual GPU boxes.
 
My laptop kicks out about 1100PPD running LinSMP... Nothing spectacular, but it's still a decent amount of points. It's a 31W CPU, and ~36PPD/W isn't too shabby.
 
I've got a 940BE @ 3.0 in a K9A2. It is good for 2300 PpD per VM (Ubuntu 8.04). I would crank it up higher, but heat is a killer.
 
Zero82z said:
My laptop kicks out about 1100PPD running LinSMP... Nothing spectacular, but it's still a decent amount of points. It's a 31W CPU, and ~36PPD/W isn't too shabby.
Click to expand...

My Opteron 170 (overclocked to 2.4GHz) pulled about 1200-1300ppd, with the Windows client. I might configure a Linux VM on my Opteron boxes, but I might not. It depends on how well the GPU clients perform with/without them.
 
natermeister said:
My Opteron 170 (overclocked to 2.4GHz) pulled about 1200-1300ppd, with the Windows client. I might configure a Linux VM on my Opteron boxes, but I might not. It depends on how well the GPU clients perform with/without them.
Click to expand...
Your CPU is almost 1.5 times faster than mine, so it would probably put out a decent amount more points with LinSMP. Probably 1500-1600PPD.

As for VM overhead, you'll have to see that for yourself. As long as you use WinAFC or something similar to automatically raise the priority of the GPU client, you shouldn't see any decrease in performance there, but you might notice slower general system performance since the extra overhead from running the VM will definitely tax your PC more.
 
the real performance kill with the VM's is just the amount of ram you have to leave dedicated to it.. that was the problem for me running 2 VM's on 2 gigs of ram.. once i added the extra gig.. i couldnt even tell the VM's were running..
 
sirmonkey1985 said:
the real performance kill with the VM's is just the amount of ram you have to leave dedicated to it.. that was the problem for me running 2 VM's on 2 gigs of ram.. once i added the extra gig.. i couldnt even tell the VM's were running..
Click to expand...
CPU usage can be a problem as well, since VMs don't relinquish CPU time as easily as native F@H clients do.
 
Zero82z said:
CPU usage can be a problem as well, since VMs don't relinquish CPU time as easily as native F@H clients do.
Click to expand...

never really noticed anything with that.. especially since the most cpu usage i ever saw running both VM's was 97%(95% was the norm).. heck i even left the VM's running when i was playing QW which is a multi-threaded game.. and never had any problems.. i guess it just depends on the processor and main OS..
 
I guess I never really understood why people insist on using something like WinAFC to control priorities when it can be done right in the VM configuration file itself.

Just add...

priority.grabbed = "idle"
priority.ungrabbed = "idle"

...to each of your VM configuration files and set each of the GPU clients to "Slightly higher" and it's a done deal.
 
You must log in or register to reply here.
Back
Top