Here's a new one.. "box exploding" while folding!

[MDC]

Happy New Year everyone! I see nobody noticed that I hit 2 million a few weeks ago yet.

One of my unoverclocked AMD systems got this error a few days ago. It also got an "early unit end" error on the next attempt at p210x, but other than that nothing wrong. Here's the log clipping.


[16:13:17] Protein: p2101_project2101_DKvillin_run0
[16:13:17]
[16:13:17] Writing local files
[16:13:17] Extra SSE boost OK.
[16:13:17] Writing local files
[16:13:17] Completed 0 out of 2500000 steps (0)
[16:21:12] Writing local files
[16:21:12] Completed 25000 out of 2500000 steps (1)
[16:29:07] Writing local files
[16:29:07] Completed 50000 out of 2500000 steps (2)
[16:37:02] Writing local files
[16:37:02] Completed 75000 out of 2500000 steps (3)
[16:44:47] Writing local files
[16:44:47] Completed 100000 out of 2500000 steps (4)
[16:50:19] Quit 101 - Fatal error: Box exploding.
[16:50:19]

[Honsy]

+1

[KJ]

What the.....?

[luxor]

And it's not even April 1st yet

[Lupine]

One of these days I'm going to remember your milestone the day it happens. Cross your fingers for 3mil.

[MDC]

That was really in my fahlog.txt today, and not a joke. And thanks Lupy, I think.

[Lupine]

If it is any consolation, you're not the only one dodging shrapnel:

http://forum.folding-community.org/viewtopic.php?t=14109

And I dug this up from some time back (can't source it):

"Box exploding" doesn't mean your computer.

The FAH process simulates a volume of space containing the protein(s) being simulated, together with a suitable number of molecules of the solvent. One of the possible errors that can happen to a simulation is for the various atoms to move farther and farther apart, until the calculations no longer mean anything. (Its sort of the opposite condition from some of the LINCS WARNING errors.)

You can treat it essentially the same way you do other unstable simulations - - namely, if there is a pattern of repeated errors on your machine, look for unstable hardware; if it's just an occasional WU, it's probably inherent in the WU itself and you should ignore it. (. . . except perhaps if you're ltchong and you understand enough about biochemical simulations to figure out how to correct the simulation, itself. )

meh, the reality of the error msg isn't nearly as entertaining a thought.

[LordPain]

If it is any consolation, you're not the only one dodging shrapnel:

http://forum.folding-community.org/v...ic.php?t=14109



And I dug this up from some time back (can't source it):


meh, the reality of the error msg isn't nearly as entertaining a thought.

The funny thing is, I suspected that was the case since they place the equation in a "box" lol