Page 1 of 1

Project 5102 on SMP Notfred's PPD cut way back

Posted: Wed Oct 21, 2009 7:40 am
by cyclometric
Hi,

The past week or so my SMP Notfred's VMWare Player client has been averaging around 2200 PPD. At some point today I downloaded Project 5102, which has a huge total point value, but my PPD are down to 1400. Here are the other symptoms and my environment stats:

Other symptoms:
vmware-vmx.exe process is only using ~40% of CPU
GPU client (details below) is now using a couple more % points of CPU than it was previously (was 4% or 5%, now is 6% to 7%)

Client & Machine stats:
Memory on the VM 1024 GB
Total machine RAM=4GB (2x2 @ 840mhz DDR2)
Vista 64 Home on Phenom 940 BE OC'd to 3675
Running 8.6 drivers, as have been for days without issues.
SMP Core 1.24

Running side by side a GPU Console 6.23 on a stock 3870
Envirovariables:
BROOK_YIELD=2
CAL_PRE_FLUSH=1
FLUSH_INTERVAL=64
(can't use CAL_NO_FLUSH at all or I get UNSTABLE_MACHINE)
GPU Load 95%, CPU Load of FahCore_11.exe ~7% (up a point or two from when SMP had the higher PPD)

Is there a problem? because it would be nice if the VMWare client used 2 cores fully, along with getting higher PPD. I noticed other issues with this WU, but those threads were a while back.

Thanks,

Re: Project 5102 on SMP Notfred's PPD cut way back

Posted: Wed Oct 21, 2009 7:50 am
by toTOW
This is normal. 5102 uses the A1 core which is older and slower than A2 ... (if you search on the forum, you'll find many threads about it).

Re: Project 5102 on SMP Notfred's PPD cut way back

Posted: Wed Oct 21, 2009 8:00 am
by cyclometric
Thanks, OK.

BTW, I did look up the WU before posting, but the last post was in January, and before that, Mid-2008, so I thought there might have been more recent news. :-)

Re: Project 5102 on SMP Notfred's PPD cut way back

Posted: Wed Oct 21, 2009 9:55 am
by uncle fuzzy
I pick up the occasional 5101 in VM, which does the same thing.

Re: Project 5102 on SMP Notfred's PPD cut way back

Posted: Wed Oct 21, 2009 9:55 pm
by cyclometric
Oh. If it's good for the cause, that's more important than me getting a F@H postbit on Overclock.net, I was just looking forward to it cuz I'm getting close. :-)