Wus Not Finishing!

[airman]

I dno't know what to do....it gets to like the 30th step or so (depends) and then it says,

 

[08:43:55] Simulation instability has been encountered. The run has entered a

[08:43:55] state from which no further progress can be made.

[08:43:55] If you often see other project units terminating early like this

[08:43:55] too, you may wish to check the stability of your computer (issues

[08:43:55] such as high temperature, overclocking, etc.).

[08:43:55] Going to send back what have done.

[08:43:55] logfile size: 9536

[08:43:55] - Writing 10218 bytes of core data to disk...

[08:43:55] ... Done.

 

that's 1,

 

here's 2,

 

[05:48:30] Completed 440000 out of 1000000 steps (44)

[05:51:05] Gromacs cannot continue further.

[05:51:05] Going to send back what have done.

[05:51:05] logfile size: 43810

[05:51:05] - Writing 44346 bytes of core data to disk...

[05:51:05]... Done.

 

 

and 3,

 

[07:28:36] Completed 19000 out of 100000 steps (19)

[07:35:31] Gromacs cannot continue further.

[07:35:31] Going to send back what have done.

[07:35:31] logfile size: 9413

 

and 4,

 

[05:17:25] Completed 3500000 out of 10000000 steps (35)

[05:17:41] Gromacs cannot continue further.

[05:17:41] Going to send back what have done.

[05:17:41] logfile size: 104736

[05:17:41] - Writing 105272 bytes of core data to disk...

[05:17:41] ... Done.

 

 

 

also...i've noticed my score has gone down...i had over 4k yesterday

[hardnrg]

do you have the -advmethods switch set?

 

cos all that does is download the beta proteins which might mess up like that...

 

also, i had a load of problems with FAH4Console version right when v5 came out so i upgraded to v5 and all was well...

 

i'd like to get those in there before i suggest the usual possibilty of system instability and asking if you've done a Prime95 24 hour test after your latest oc tweak etc...

[airman]

yeah i was....mainly because a guide told me to. i took it off and i'll see what's goin on.

[hardnrg]

it (-advmethods) used to do something useful like give preference for/against gromacs or double-gromacs... that's what most of the guides are talkin about...

 

since i removed it i haven't had any WUs just die apart from the v4/v5 thing which might have been more because i use EMIII to monitor/start/stop them...

[Nemo]

I agree with hardnrg. I've stopped using -advmethods. I posted an explanation as to why a little while ago if you want to search for it.

[r_target]

Yeah, I don't know if those are errors from instability or not. Looks like they may have sent you some wonky proteins.

[Nemo]

I have seen that occassionally on some of my rigs (maybe twice). None of these rigs are OC'ed so I think it was the protein I got. You'll get partial credit for the protein, just not the full point value you would get if the protein finished.

 

If, after taking off the -advmethods flag, you continue to see these errors, then I would back your OC off a little to see if that would help.

 

One of the comments I remember reading over on the Folding Community forum from one of the mods is that Gromacs are the most stressful procedures you can ask a CPU to perform- even more so than Prime (take that for what it's worth).

 

Edit:

Check out this thread and see if it helps.

[Zeni]

I used to have that issue all the time, prevented me from joining the F@H team much earlier than I had, because I got that message on my first proteins for 3 different installs, it eventually went away on it's own I guess, as the 4th install had no mention of that, and it's been smooth folding ever since.