It is possible to avoid core 17 WU?

[OVV]

Hi,
my system is:

Xeon E1650 (6C/12T)
Nvidia quadro 2000
8 Gb RAM
Win 7 Pro 64 bit.

Due to low performance of the GPU (is basically a downclocked GTS450) when running core 17 (ie p8900) i have a big loss of PPD, here some example:

SMP 12 (32-35K) + GPU 762x (12K) = 44-47K
SMP 12 (32-35K) + GPU 8018 (9K) = 41-44K
SMP 10 (26-28K) + GPU 8900 (6K) = 32-34K

So, starting from "equal point means equale science produced" for me it's better to run only SMP instead of fold also on GPU if i receive core 17 WU.

There is some option to force the system to accept only core 15 WU? (at least until they exist or i can upgrade my GPU)

Probably the answer is no...

Thanks in advance

[Zagen30]

Well, the most drastic option I can think of is to remove the GPU slot and install GPU client 6.41 from here. Core 17 requires v7, so you won't get it with v6.

At one point early in core 17 beta testing, adding the flag max-packet-size:big would prevent you getting core 17 WUs, probably due to them being set to size of normal. I don't know if that's still true, but you could try that.

[7im]

That is only a temporary delay tactic. All GPU fahcores, except core 17, have been announced to be End Of Life. There is no way to avoid them going forward.

[OVV]

Tried max-packet-size:big flag with no luck.

[7im]

Tried max-packet-size:big flag with no luck.

That work initially, when that core was still in testing, but it stopped working when released to the public.

[OVV]

Today received 8018 WU, probably luck draw because i haven't changed nothing.

[bruce]

Core 17 will not run on the V6 client so that would force the assignment of an older WU using an older core (if there are any WUs left). With the EOL announcements, the servers will run out of WUs to assign but nobody can predict when that will happen. Also no GPU support on the Linux V6 client, but that's not your problem.

Finding the right version of drivers to run FAH on a Quadro 2000 is questionable.