• 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.

units ended early

F

FireGryphon

n00b
Joined
Sep 20, 2004
Messages
22
about what ratio of units should terminate early?

sorry if this was already posted, lock if it was.
 
Hi, you shouldn’t have any WU terminate early. Have you tried running Prime95 to check the stability of your over clock? You can find it here.
Prime95
If you can this program for 24 HRS you are stable.
Hope this helps. :D
 
I did indeed run Prime95, and it seemed to work fine. I ran it for several hours. Is that not enough?
 
You may get the odd protien finishing early if its a new one straight out of beta testing.
You tend to know when that happens as Stanford pull it for a while as they retest it.

More than one failure and I would say you have a problem with your box.
Try running Memtest86 to check for a bad memory cell.
I've had Windozes run fine with a bad stick of memory but a 5% failure rate with F@H.

Luck......... :D
u=Tigerbiten.gif
 
Yeah, you might want to also try Priming for 24 hours too. That should let you know of any math errors you could be getting.
 
gromacs core stresses cpu further than prime.

My gigabyte piece of garbage stable in prime for 24hrs+ and can run 2 tinkers at 240 fsb.

Must step back to 235 fsb if running dgromacs and most other gromacs.

With the "large" p13xx wu's, must step back further to 229 fsb to avoid Early_Unit_End.
 
ezee said:
gromacs core stresses cpu further than prime.

My gigabyte piece of garbage stable in prime for 24hrs+ and can run 2 tinkers at 240 fsb.

Must step back to 235 fsb if running dgromacs and most other gromacs.

With the "large" p13xx wu's, must step back further to 229 fsb to avoid Early_Unit_End.
Click to expand...

Exactly right, prime is a good test but gromacs kill. Give prime 24 hours, and then to memtest.

Some times backing down your fsb one notch or two is all it takes.

Luck
 
Another option for memory problem diagnosis is Microsoft's own Windows Memory Diagnostic program. Handy little tool that checks memory too. I just this weekend used it to diagnose my first ever bad sticks of Corsair memory.

Worst part of it all was that I had to remove the Gig of RAM from my personal first Linux build (Fedora Core 2, and the install went just fine thanks) to put into the customer's machine. I didn;t even get Folding installed om the Linux box, and now I will have to wait for Corsair's RMA in order to bring it back on line.

<----Thanks xENo
 
I was going to ask the same thing. This is what I got:

[01:56:29] Completed 186000 out of 200000 steps (93)
[01:57:14] Quit 101 - Fatal error:
[01:57:14] Step 186066, time 372.132 (ps) LINCS WARNING
[01:57:14] relative constraint deviation after LINCS:
[01:57:14] max 0.002165 (between atoms 43481 and 43484) rms 0.000033
[01:57:14]
[01:57:14] Simulation instability has been encountered. The run has entered a
[01:57:14] state from which no further progress can be made.
[01:57:14] If you often see other project units terminating early like this
[01:57:14] too, you may wish to check the stability of your computer (issues
[01:57:14] such as high temperature, overclocking, etc.).
[01:57:14] Going to send back what have done.
[01:57:14] logfile size: 81585
[01:57:14] - Writing 82271 bytes of core data to disk...
[01:57:14] ... Done.
[01:57:14]
[01:57:14] Folding@home Core Shutdown: EARLY_UNIT_END
[01:57:17] CoreStatus = 72 (114)
[01:57:17] Sending work to server
[01:57:17] + Attempting to send results
[01:57:19] + Results successfully sent

The thing is I've had no stability problems with this machine. it started a new WU and it's been chugging since. I rebooted this morning when I came in anyway (this is my work PC) after I saw this but it seems like it's been all good since.
 
The run has entered a state from which no further progress can be made.
If you often see other project units terminating early like this too, you may wish to check the stability of your computer (issues such as high temperature, overclocking, etc.).
Going to send back what have done.

Thats the relevant bit.
If its an only one, then you just got an iffy work unit and you will get some of the work units points depending on how far you got before it shut down.
If you get two or more then I would start to think maybe I'd better check the boxen over to makesure its running ok.

If your not sure then check out the forums over at http://forum.folding-community.org/
Any iffy work units will showup as people question why they finnished early.

Luck....... :D
u=Tigerbiten.gif
 
You must log in or register to reply here.
Back
Top