mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Software (https://www.mersenneforum.org/forumdisplay.php?f=10)
-   -   Same crash, same test, same problem. Bug or instability? (https://www.mersenneforum.org/showthread.php?t=2360)

SlightlyStoopit 2004-04-17 20:25

Same crash, same test, same problem. Bug or instability?
 
Just having a bit of a problem here, and was wondering what you all thing. Been tweakin around with the comp a bit more due to some unstability. Right now im running at 11x215 @ 1.93 Volts. This is on a NF7 -s, 2800+ and 1gig PC3500 Mushy Lvl2. My comp should be able to handle it, but P95 crashes in the exact same spot, every time:

[Fri Apr 16 19:50:52 2004]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Fri Apr 16 20:07:38 2004]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Fri Apr 16 20:28:01 2004]
FATAL ERROR: Rounding was 0.4998058081, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Fri Apr 16 20:49:53 2004]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Fri Apr 16 20:57:13 2004]
FATAL ERROR: Rounding was 0.4953308105, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Fri Apr 16 23:33:59 2004]
FATAL ERROR: Rounding was 0.4997987974, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Sat Apr 17 00:51:09 2004]
FATAL ERROR: Rounding was 0.4999995828, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Sat Apr 17 01:41:43 2004]
FATAL ERROR: Rounding was 0.4999999106, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Sat Apr 17 02:18:29 2004]
FATAL ERROR: Rounding was 0.4999198914, expected less than 0.4
Hardware failure detected, consult stress.txt file.


I know its not the SAME numbers but its the same test. So lemme ask you all, do you think its my hardware, or have I just found some kind of bug?

Ive run memtest up to test 8, not a single error. So I do not believe its faulty ram. Any ideas?

dsouza123 2004-04-17 23:46

With the guess that you are running Prime95 on Windows XP.

What version of Prime95 ? 23.8.1 ?

Does it also crash at lower settings than 11x215 @ 1.93 volts? (CPU unstable at high overclock)
What would the default settings be ?

A slightly elevated voltage at normal clock speed can make it more stable if the PSU (power supply) is marginal.

Is it normally 2.08 or 2.25 Ghz ?

Does it crash in Safe Mode ? (driver subtly interfering with the calculations)

Are you using any line conditioning, UPS, that would remove noise, spikes and under/overvoltages ?
(memory can be corrupted by power issues)

SlightlyStoopit 2004-04-18 01:31

Yea, I tried up to 2v and it still crashed there. 2800's usually run at 2.08 or so, so its not an amazing overclock. The rails seem pretty smooth, but Im gonna be getting a bigger PSU soon- I think it may help out a bit.


Im starting to think it was a bug- went about 3 hours this morning with no crash, so..the Computer Gremlins got me again. Thanks for the response though :showoff:


All times are UTC. The time now is 01:18.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.