• 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.

Doing something wrong????!????

FrEaKy

FrEaKy

[H] Movie and TV Show Review Guy
2FA
Joined
Jan 31, 2003
Messages
14,173
3x E8600's, getting only about 2-2.3K PPD each
my Q9550, only running 1 SMP with 4 cores, only getting 1800 PPD....

I feel like im missing something here, like they are not pushing like they should.

Any ideas?

Windows 7 64 for the Q9550
Win XP 64 for the E8600's.
 
VM's or Windoze ??

If your running the client in a VM, you should be getting well over 300 PpDpGhz.
Mine are running at ~380 PpDpGhz on my Q6600's.

If your running in Windoze then all bets are off.
If you get a slow A1 work-unit you may be only doing ~150 PpDpGhz.
A2 work-units in Windoze fold at only ~250-300 PpDpGhz.
Hence the reason to run VM's, you can get ~25-100% points boost depending on work-units for no extra power.
It the reason I've used VM's for ~2 years now.

Luck ............ :D
 
Its a notfred VM client on Windows
 
2677 and 2669 on my E8600's
2662 on the Q9550
 
My Q6600@3.2 does an average of 5300PPD on the 2662 so id say something is wrong.
 
It sound like you may be running without advanced methods boost.
Check in your FAHlog.txt file at the start of the work-unit for the line "Assembly optimizations on if available." or "Extra SSE boost OK".
If either is NOT there then your running without any SSE boost.
You need to shut the client down correctly and restart it to restart the SSE boost.
But the SSE boost is hard coded into the client, so I don't think its that.

Check in the FAHlog.txt files how big the work-unit is.
It should have 250,000 steps.
Stanford went through a patch where some work-units had auto-generated with a lot more steps than normal, so they folded slow.
If yours is too big again then you may have picked up a bad work-unit.
There are some recent reports of a p2665 with 350,000 steps which folds at half speed.

If you do shut the client down then you could try deleting the FahCore files and let the client redownload new ones.
There could be a minor glitch in the core.

Luck ............ :D
 
I'm also getting horrific performance on one of my Clovertown systems, and I mean horrible. Never could find the reason why there is such a disparity. There seems to be great variation with Intel quad-core performance and the SMP client. I have 4 Intel systems with core based processors. They all perform very differently from each other even though they are very similarly clocked. The performance variation can be enormous and extremely difficult to explain. :confused:
 
You must log in or register to reply here.
Back
Top