Vcore & Prime95

TheUnk

Golden Member
Jun 24, 2005
1,810
0
71
So let's say you get your OC stable at 1.3 vcore, running 10hrs or so of Prime95 with no error. Can any harm be done to your chip by running the same everything except the vcore, you drop that to 1.25v and Prime95 has errors after about an hour. Is your biggest risk just having apps crash or BSOD?

 

mrSHEiK124

Lifer
Mar 6, 2004
11,488
2
0
You won't actually damage the chip, like you said, it just won't be as stable. 1.25V is the next step down from 1.3V on whatever setup you're using? Try somewhere in between like 1.275V.
 

DSF

Diamond Member
Oct 6, 2007
4,902
0
71
Errors can also cause corrupted OS/files. If you know a setting isn't stable, you don't want to leave it there.
 

Machine350

Senior member
Oct 8, 2004
537
0
0
Like DSF said, instability can cause some pretty nasty problems. I'd let prime95 run for 24 hours to really know you're stable. As long as your reported vcore isn't exceeding the chip maker's specs, you don't have to worry too much about damaging the chip.
 

error8

Diamond Member
Nov 28, 2007
3,204
0
76
Originally posted by: Machine350
I'd let prime95 run for 24 hours to really know you're stable.

I'd use linpack for ~ 10 minutes ( 10 tests ) or OCCT for 2 hours, to test for stability. It's amazing what people do to test their hardware. 24 hours of prime95 seems horrible!
 

mrSHEiK124

Lifer
Mar 6, 2004
11,488
2
0
Originally posted by: error8
Originally posted by: Machine350
I'd let prime95 run for 24 hours to really know you're stable.

I'd use linpack for ~ 10 minutes ( 10 tests ) or OCCT for 2 hours, to test for stability. It's amazing what people do to test their hardware. 24 hours of prime95 seems horrible!

10 passes of Linpack isn't exactly enough. I'd do at least 20, I did 50 to test my X3360 overclock to 2.4 GHz, it took an hour and 45 minutes.
 

eternalone

Golden Member
Sep 10, 2008
1,500
2
81
Who came up with the 24 hour theory anyway and who said it cant damage your cpu??? Im sure that it actually can if your messing with high voltages and reaching the limits of your Tcase Temps. That being said I agree 2- 5 hours should be enough, but Im sure people would still say 24hrs then why not 48 hrs just to be sure??? :)
 

Machine350

Senior member
Oct 8, 2004
537
0
0
My experience has been that sometimes I'll get an error in prime95 after about 10 hours. It's kind of weird, but that's happened more than once. So for me, I don't feel my system is stable unless it's stressed 24 hours, but then again, I'm kind of OCD.
 

LOUISSSSS

Diamond Member
Dec 5, 2005
8,771
58
91
your system will be stable enough for ANYTHING you do if you're 2 hours stable with OCCT.

and YES you will have problems after an instable CPU. if you're lucky u won't, but your OS will probably be corrupted already.
 

mrSHEiK124

Lifer
Mar 6, 2004
11,488
2
0
Originally posted by: LOUISSSSS
your system will be stable enough for ANYTHING you do if you're 2 hours stable with OCCT.

and YES you will have problems after an instable CPU. if you're lucky u won't, but your OS will probably be corrupted already.

I like Linpack, it's very quick, 50 passes will take less than two hours and is GUARANTEED stability, 20 passes is enough for anyone who isn't crazy.
 

nerp

Diamond Member
Dec 31, 2005
9,865
105
106
Uhh, running prim95 isn't going to "damage" your CPU just like running F@H isn't going to damage it unless you're overvolting it inappropriately.