• 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.

2 GPU's problem on vista

Vaulter98c

Vaulter98c

[H]ard|DCer of the Month - October 2009
2FA
Joined
May 21, 2008
Messages
5,840
OK, so 5k PPD isnt enough, I gotta get my 9600 GSO goin too!

Problem is, I cant seem to figure out how to get two running

Ive read the multi-GPU guide here and followed it to the T, three times now
Ive redone my drivers with three pairs (18something, 182.06 WHQL, 185 something betas)

Still, no luck, whenever I start up GPU one its fine, then when I click two I get an error, something like

"A potential conflict was detected"
process blah blah is currently running and may also be a client with Mach. ID ##. Program will exit now. Check to makes sure no client is using path BLAH BALH

Ya, thats about it, Im stumped, no clue what I did wrong, the targets and start in are fine, but I only ever use my 8800 GTX. When I start the GPU1 shorty first then click the shorty to GPU2, It has one unit, and says BLAH BLAH folder1 . When I click shorty GPU2 first, then GPU1, its a different unit (diferent percentage) and the error says BLAH BLAH folder2, so it is reading from 2 seperate folders, and it is 2 seperate WU's, but no mater which I click first, the 8800GTX takes that one, and the next one errors out

Any help, Im stumped....

Im running Vista 32, all drivers have been moved enough I dont think its a driver issue, and Im using the latest GPU client from FAH (the one with the Nvidia Viewer) webpage

Please help, I need more points!!!
 
Well first off it is normaly not a good idea to mix cards with different shader numbers.

Just checking.....
You have a dummy plug and desktop extended?
 
ya, Ive got the dummy, Its my Physx card right now

I thought that only applied to big differences, like my 88 and my 85
 
Post your logs from both clients it seems like you didn't change the macine id when you set it up.
 
I did, but Im not sure how to post the logs, Im a folding newb, actually, Ive been doing it for 3 years, but that was in the day an age of run folding app, carry on with life

I do know I changed the ID numbers, and they always change themselves back, to whatever number that I start (ex: GPU1=6 , GPU2=7 , whatever one I click first, thats the number they both change too...)

It would really suck if I have to have all the same cards because I cant afford to make a whole new box, I just wanted to get my Physx card rolling
 
What happens if you try to restart it after the "A potential conflict was detected" error.
Sometimes folding only thinks something else has the same ID of the client.
When you restart it, then it runs fine.

But the first ten lines or so from each log file would help a lot.

Luck ........... :D
 
OK, long post will be LOOOOONGGG

here we go, Ive closed them all, then I click GPU 1 link on desktop, and let it run for a minute to see that its working

# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.23

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Users\Chomp\AppData\Roaming\Folding@Home-gpu
Arguments: -gpu 0

[14:58:07] - Ask before connecting: No
[14:58:07] - User name: Vaulter98c (Team 33)
[14:58:07] - User ID: 323968F618B9DEE7
[14:58:07] - Machine ID: 7
[14:58:07]
[14:58:08] Loaded queue successfully.
[14:58:08] Initialization complete
[14:58:08]
[14:58:08] + Processing work unit
[14:58:08] Core required: FahCore_11.exe
[14:58:08] Core found.
[14:58:08] Working on queue slot 03 [February 26 14:58:08 UTC]
[14:58:08] + Working ...
[14:58:08]
[14:58:08] *------------------------------*
[14:58:08] Folding@Home GPU Core - Beta
[14:58:08] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[14:58:08]
[14:58:08] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[14:58:08] Build host: amoeba
[14:58:08] Board Type: Nvidia
[14:58:08] Core :
[14:58:08] Preparing to commence simulation
[14:58:08] - Looking at optimizations...
[14:58:08] - Files status OK
[14:58:08] - Expanded 98726 -> 492276 (decompressed 498.6 percent)
[14:58:08] Called DecompressByteArray: compressed_data_size=98726 data_size=492276, decompressed_data_size=492276 diff=0
[14:58:08] - Digital signature verified
[14:58:08]
[14:58:08] Project: 5751 (Run 6, Clone 116, Gen 142)
[14:58:08]
[14:58:08] Assembly optimizations on if available.
[14:58:08] Entering M.D.
[14:58:14] Will resume from checkpoint file
[14:58:14] Working on Protein
[14:58:18] Client config found, loading data.
[14:58:18] Starting GUI Server
[14:58:18] Resuming from checkpoint
[14:58:18] Verified work/wudata_03.log
[14:58:18] Verified work/wudata_03.edr
[14:58:18] Verified work/wudata_03.xtc
[14:58:18] Completed 74%
Click to expand...

Then I click GPU2 link.... get error, and this is the log file from second folder (under users)

Umm, WTF, I didnt change a damn thing, and now it started working all of a sudden? Here's log file from GPU2 folder.... I swear, I have no clue WTF is going on

# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.23

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Users\Chomp\AppData\Roaming\Folding@Home-gpu2
Arguments: -gpu 1

[15:01:55] - Ask before connecting: No
[15:01:55] - User name: Vaulter98c (Team 33)
[15:01:55] - User ID: 323968F618B9DEE7
[15:01:55] - Machine ID: 10
[15:01:55]
[15:01:56] Loaded queue successfully.
[15:01:56] Initialization complete
[15:01:56]
[15:01:56] + Processing work unit
[15:01:56] Core required: FahCore_11.exe
[15:01:56] Core found.
[15:01:56] Working on queue slot 02 [February 26 15:01:56 UTC]
[15:01:56] + Working ...
[15:01:56]
[15:01:56] *------------------------------*
[15:01:56] Folding@Home GPU Core - Beta
[15:01:56] Version 1.19 (Mon Nov 3 09:34:13 PST 2008)
[15:01:56]
[15:01:56] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[15:01:56] Build host: amoeba
[15:01:56] Board Type: Nvidia
[15:01:56] Core :
[15:01:56] Preparing to commence simulation
[15:01:56] - Looking at optimizations...
[15:01:56] - Files status OK
[15:01:56] - Expanded 45401 -> 251112 (decompressed 553.0 percent)
[15:01:56] Called DecompressByteArray: compressed_data_size=45401 data_size=251112, decompressed_data_size=251112 diff=0
[15:01:56] - Digital signature verified
[15:01:56]
[15:01:56] Project: 5771 (Run 9, Clone 239, Gen 275)
[15:01:56]
[15:01:56] Assembly optimizations on if available.
[15:01:56] Entering M.D.
[15:02:02] Will resume from checkpoint file
[15:02:03] Working on Protein
[15:02:03] Client config found, loading data.
[15:02:03] Resuming from checkpoint
[15:02:03] Verified work/wudata_02.log
[15:02:03] Verified work/wudata_02.edr
[15:02:03] Verified work/wudata_02.xtc
[15:02:03] Starting GUI Server
[15:03:42] Completed 1%
Click to expand...

FAHMon shows both working, but the NVDisplay thing is all fuxxed up, it only reads a 8800GTX and the same %% for both clients.

What just happened? Here's a shot of my desktop with FAHMon running now 2 clients, dos this mean everything is good in folding land? Obviously, Im assuming I'll only know after each card submits a WU and it gets accepted, but this is crazy, I swear I didnt change a thing


((She made me cover her face lol))

Any ideas why it just started working? Im sorry for wasting yalls time and what not, I really have no clue what happened
 
Looks good to me. The viewer really doesn't work right. The best way to monitor your clients is with FahMon or FahSpy. Let it run for a while then re-check with FahMon. You'll be able to tell if you're making progress.
 
LOL

thanks for the sugestions, sorry I wasted your time when it just randomly worked

love this dc stuff!

also, maybe its jus tthe WU's Ive been getting, but I have noticed a drop on my 8800 GTX from about 5--5.5k PPD to about 3.5-3.8k , but with both cards, Im at 6.8k so I'll leave it here for now. Im at about 7.5k+ PPD with the PS3 and this rig folding, think I should start a CPU client as well??
 
Meh, no problem. I often just keep hitting buttons until something work..... or I have to re-install :p
 
ah, must have been the unit, Im up to 8.7K now, I'm back on track
 
Yep, my 9800GX2's will run anywhere from 2500ppd/core up to 5000+ppd/core depending on the WU.
 
SatTech said:
Yep, my 9800GX2's will run anywhere from 2500ppd/core up to 5000+ppd/core depending on the WU.
Click to expand...

Thats slow.
I've clocked my GX2's shaders to 1728 Mhz.
511's give me ~3,900 PpDpCore, 353's give me ~6,100 PpDpCore.

Luck ....... :D
 
Yeah, they're running stock right now. I was kinda guessing on the range of PPD. It might be more around 3000 to 5500 ppd but I'm nowhere near my box to check those numbers.
 
You must log in or register to reply here.
Back
Top