• 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.

New GPU2 Project 10109 on -advmethods

Tobit

[H]ard|DCer of the Month - March 2010/May 2011
Joined
Dec 12, 2005
Messages
5,145
This GPU2 project for nvidia GPUs is the beginning of a study on how ubiquitin folds.

Points: 494
Timeout: 2 days
Deadline: 3 days

Note to the [H]orde: Don't be surprised if these WUs run 10% slower compared to other GPU2 WUs. The same thing was experienced by me in beta testing. However, these are very stable and I had no issues on a variety of highly overclocked G92 and G200 hardware. If you don't like the 10% slowdown, remove -advmethods.
 
Some of my numbers:

Code:
 Project ID: 10109
 Core: GROGPU2
 Credit: 494
 Frames: 100

 Name: callisto - 9600GSO
 Path: C:\fah\gpu2-1\
 Number of Frames Observed: 300

 Min. Time / Frame : 00:01:53 - 3,777.1 PPD
 Avg. Time / Frame : 00:01:54 - 3,744.0 PPD

 Name: callisto - GTX275
 Path: C:\fah\gpu2-0\
 Number of Frames Observed: 300

 Min. Time / Frame : 00:00:48 - 8,892.0 PPD
 Avg. Time / Frame : 00:00:49 - 8,710.5 PPD

 Name: ganymede - GTX275-0
 Path: \\ganymede\fah\gpu0\
 Number of Frames Observed: 300

 Min. Time / Frame : 00:00:49 - 8,710.5 PPD
 Avg. Time / Frame : 00:00:49 - 8,710.5 PPD
 
How are these with the temps, similar to other WUs?
 
there are some big point non-fermi gpu3 -advmethods projects out there now slow going though (about 5min per 1% on a 9600gt) but worth 1300 pts.

by yslin » Tue Sep 14, 2010 10:04 pm
Hi,

New GPU3 projects 11161-11168 are entering the advanced stage of testing (restricted to non-Fermi boards).

These Amyloid beta peptide simulations will serve as tests to evaluate the performance of a new core (openmm-gromacs) on gpu with Generalized Born (GB) model used as implicit solvent.

Points and deadlines:
project 11161: 1352 points, preferred deadline 20 days, final deadline 29 days
project 11162: 1358 points, preferred deadline 20 days, final deadline 29 days
project 11163: 1349 points, preferred deadline 20 days, final deadline 29 days
project 11164: 1352 points, preferred deadline 20 days, final deadline 29 days
project 11165: 1352 points, preferred deadline 20 days, final deadline 29 days
project 11166: 1241 points, preferred deadline 19 days, final deadline 27 days
project 11167: 1358 points, preferred deadline 20 days, final deadline 29 days
project 11168: 1352 points, preferred deadline 20 days, final deadline 29 days

Thanks!

yslin
 
These WUs are also being assigned to clients without -advmethods. I guess they are general release now?

there are some big point non-fermi gpu3 -advmethods projects out there now slow going though (about 5min per 1% on a 9600gt) but worth 1300 pts.
What PPD do these get on a G92 card? I have had several GPU clients with -advmethods hard set in the client config and didn't receive anything other than the P10109 WUs.
 
there are some big point non-fermi gpu3 -advmethods projects out there now slow going though (about 5min per 1% on a 9600gt) but worth 1300 pts.
Yes, I started a separate thread for those when they were released as well. :p
 
What PPD do these get on a G92 card? I have had several GPU clients with -advmethods hard set in the client config and didn't receive anything other than the P10109 WUs.
The projects mrbigshot referenced requires the GPU3 client. They are fine on G92 cards but they suck ass on G200 cards.
 
These WUs are also being assigned to clients without -advmethods. I guess they are general release now?

What PPD do these get on a G92 card? I have had several GPU clients with -advmethods hard set in the client config and didn't receive anything other than the P10109 WUs.

These seem to be in the wild. I got one on my GTX285 earlier today. About ~9500 PpD on a card that normally gets 10.5K PpD or better depending on the WU
 
^ same here - came back from camping over the weekend, and I've gone from 9700-9800 ppd down to 3 straight days of 8400, 8900 and 8700 ppd, and I'm not running -advmethods. I'm seeing about 3600 ppd on an overclocked 9600 GSO and about 5200 ppd on a stock 9800GTX+, down from 4100/5800 ppd - or about 12% down. May have to postpone some commando work to bring some quads online for the push to 4B to offset this loss! :D
 
they hit mine really hard as well. takes about 9 hours to finish one and i have been getting them nearly non-stop. only 3 regular units in 3 days now. im down about 30% from them.
 
I'm taking about a 1k PpD hit per GTX285 and my 9800GX2's are also affected by this.

These are definitely not on -advmethods as I don't normally run that on my GPU2 clients.
 
Back
Top