Overclock.net banner

[SOLVED]Anyone else getting non-stop UNSTABLE_MACHINE errors?

551 Views 16 Replies 11 Participants Last post by  zooterboy
WU 5903 pretty much exclusively, getting it on two of three GTX 260s in the same rig. Won't even start, just constant errors until EUE limit. Even getting them at stock clocks.

Code:
Code:
10:04:46] [email protected] GPU Core - Beta
[10:04:46] Version 1.24 (Mon Mar 2 19:49:32 PST 2009)
[10:04:46] 
[10:04:46] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[10:04:46] Build host: vspm46
[10:04:46] Board Type: Nvidia
[10:04:46] Core      : 
[10:04:46] Preparing to commence simulation
[10:04:46] - Assembly optimizations manually forced on.
[10:04:46] - Not checking prior termination.
[10:04:46] - Expanded 68430 -> 357580 (decompressed 522.5 percent)
[10:04:46] Called DecompressByteArray: compressed_data_size=68430 data_size=357580, decompressed_data_size=357580 diff=0
[10:04:46] - Digital signature verified
[10:04:46] 
[10:04:46] Project: 5903 (Run 1, Clone 254, Gen 4)
[10:04:46] 
[10:04:46] Assembly optimizations on if available.
[10:04:46] Entering M.D.
[10:04:52] Will resume from checkpoint file
[10:04:52] Tpr hash work/wudata_09.tpr:  4227136318 1622523732 3885116843 2446024178 1763735763
[10:04:52] mdrun_gpu returned 
[10:04:52] Going to send back what have done -- stepsTotalG=0
[10:04:52] Work fraction=0.0000 steps=0.
[10:04:56] logfile size=21653 infoLength=21653 edr=0 trr=25
[10:04:56] - Writing 22191 bytes of core data to disk...
[10:04:56] Done: 21679 -> 5273 (compressed to 24.3 percent)
[10:04:56]   ... Done.
[10:04:59] 
[10:04:59] [email protected] Core Shutdown: UNSTABLE_MACHINE
[10:05:02] CoreStatus = 7A (122)
[10:05:02] Sending work to server
[10:05:02] Project: 5903 (Run 1, Clone 254, Gen 4)
[10:05:02] - Read packet limit of 540015616... Set to 524286976.

[10:05:02] + Attempting to send results [March 16 10:05:02 UTC]
[10:05:02] - Reading file work/wuresults_09.dat from core
[10:05:02]   (Read 5785 bytes from disk)
[10:05:02] Connecting to http://171.64.122.70:8080/
[10:05:03] Posted data.
[10:05:03] Initial: 0000; - Uploaded at ~6 kB/s
[10:05:03] - Averaged speed for that direction ~12 kB/s
[10:05:03] + Results successfully sent
[10:05:03] Thank you for your contribution to [email protected]
[10:05:07] EUE limit exceeded. Pausing 24 hours.
[10:17:39] ***** Got a SIGTERM signal (2)
[10:17:39] Killing all core threads

[email protected] Client Shutdown.
See less See more
1 - 17 of 17 Posts
I think we're about to see all of our averages take a nose dive...

Edit: Correction, getting it on 5902 and 5903.
One of my 8600GTs was doing it pretty extensively last few weeks but I haven't seen one since about this time last week. I wasn't getting EUE, I was getting GuardedRun.
Thing is though, I've never overclocked this card and the temps are fine (<50C).
Meh.
Quote:

Originally Posted by XFreeRollerX View Post
driver problems, try a newer/older version
obviously not when alot of people are having problems.
See less See more
Like I said, its a driver problem, software related... When a software has a bug, don't a lot of people get angry over it?

I've gone through this several times and the result is a driver change/update will fix the problem or wait for stanford to update the core_14 for GPU folding.
Yea same issues here with those cores.. I don't believe I'm getting EUE's but my points are not totally as well my queueinfo shows 3 of those 5603's that were downloaded yesterday yet had a FINISH DATE of Jan, 1st


I'm afraid I've lost those points due to them.. All 1680 pt ones.. Maybe Stanford did something to the core and it was bugged. I'm also going to update some drivers tho honestly when the 1680 pt WU's first came out they were great but last 2 days they have been giving many issues to many people.
See less See more
I'm gettting it with my 4870, usually the next WU is a different project and just keeps going.
Quote:


Originally Posted by Powelly
View Post

One of my 8600GTs was doing it pretty extensively last few weeks but I haven't seen one since about this time last week. I wasn't getting EUE, I was getting GuardedRun.
Thing is though, I've never overclocked this card and the temps are fine (<50C).
Meh.

8600GT's are notorious for EUEs. I had one that didn't finish a single WU in 2 weeks time. Underclocked, different drivers, tried everything. I now recommend only 8800GS or better.
See less See more
Does the 4870 have problems with it like the other cards and you delete it or you just keep it running and patiently wait for the other WU to pop up after you finished the 1680?
Do people not realize when you get an EUE it pro-rates your points upon sending the incomplete WU...You really aren't loosing alot of points unless the 24hr Pause kicks in and you don't notice it...

Look at post #2 inside the red box ipod put around the pause..."results successfully sent"..."thank you for your contribution" some points were still awarded

Knock on wood I haven't had an issue with the pause, only a few EUEs every now and then...

Keep 'em Foldin'...
See less See more
2
Quote:


Originally Posted by Inktfish
View Post

Does the 4870 have problems with it like the other cards and you delete it or you just keep it running and patiently wait for the other WU to pop up after you finished the 1680?

I just let it run, and a new WU automatically downloads. I get the EUE before the first frame is processed: total time from EUE to next WU is less than 60 seconds.

Quote:


Originally Posted by wannabe_OC
View Post

Do people not realize when you get an EUE it pro-rates your points upon sending the incomplete WU...You really aren't loosing alot of points unless the 24hr Pause kicks in and you don't notice it...

Must be why I turned in 5 WUs for only 25 points the other day, although I did not get the 24 hour pause.
See less See more
The only time I had that problem I lowered the shader OC by 20 and I've not seen that issue since.
I actually am having much better luck with these newer 59XX WUs. Heat and power consumption is down signifigantly, and points seem about where they should be if not better than when I was getting a large number of 511s.

When you have 50 GPUs running, you notice when things are going well, and when they are not.
34
I don't think I made the OP clear. I'm getting non-stop guarded run errors instantly until it hits the 24 hour pause. I doubt that it's a driver issue since I've been running this driver for a couple weeks now with no problem. Just all of a sudden I have 2 out of 3 cards not working? They're GTX 260s too, never had any problem folding anything with them... Instant fail with every WU, even underclocked...I don't get it. I did all the old standby things, deleting the work folder, log files, queue, etc. Even cloned the one card's folder that's working, still no good.
I guess I'll uninstall and reinstall the driver when I get home...see if that works.
Funny thing is, my 8800s and gx2 are having no problems...must not be the WU.
My rig is probably pissed off at me for gaming with it yesterday instead of folding. LOL

OMG...Guess who's an idiot? Yeah. Does everyone know that you have to extend the desktop to the cards again after enabling and disabling SLI? Just wanted to make sure, because APPARENTLY I DON'T!!!


So, yeah. Uh... Problem solved. I.D.10 T error combined with PEBUAC.

I'm so freaking embarrassed. . .
See less See more
1 - 17 of 17 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top