Shouldn't Prime95 fail before BSOD?

[Mastiff]

I'm still scratching my head here, trying to get my Prescott to behave, and I thought about something: The pc has a tendency to go straight to BSOD, without Prime stopping. The temperature doesn't matter, it seems, since reducing the VCore makes that BSOD come a lot sooner, and I have seen temperatures around 60 degrees when it happens, while Prime seems to be quite happy chugging along with around 67-68 degrees. The STOP messages are different from time to time as well. Confusing!

[Lancelot]

It might happen that Prime95 causes such a large error the system BSOD's before it even has time to goto yellow...

[Mastiff]

OK, thanks. I have used it on many computers earlier, but never had that happen. It may be because I have never tried a P4 with HT enabled. I will try without that and see what happens.

[aaronrkelly]

HT doesnt have anything to do with it - its related to the type of failure. Does it BSOD if you drop the temps, try removing the side panel to drop the temps and see if its BSODs then - if not its temp related.

[Mastiff]

No, it BSODs totally random. I had it running all night on 68 degrees, and nothing happened. Actually it happens a lot faster when I lower the VCore, which should lower the temp. As of right now I'm running Prime95 on the computer without HT, and with the VCore dropped to 1.5 instead of 1.6. Strangely enough it does work and has worked for...all of four minutes now. But with HT enabled I wasn't able to get into Windows before I saw the BSOD on that VCore. Oh, this is with FSB 255, 3.2 @ 4.0. Looking very much up! Temperature around 62 degrees, which is four degrees down, because of the lower VCore, of course.

[Lancelot]

1.5Vcore is the absolute max. you should give a Prescott. Running above that will also damage your mainboards power management section in the long run. You'll simply have to lower your OC...

[Mastiff]

OK, thanks. That explains a few things. I'm used to those Nortwhoods. They can take a lot more. You know what I mean, you've got the same Abit mobo that I'm running my Northwood on! :)

[Lat]

my northy 2.4c freezes at 261+, without any warning in prime

[Mr.Guvernment]

didnt know 1.5 is max to throw into a pressy - any links to provde more info on that? First time i have heard that.

[Mastiff]

Look above, Lancelot's message. Of course I haven't heard it any other place either. Is it wrong?

[Sjaak]

Look above, Lancelot's message. Of course I haven't heard it any other place either. Is it wrong?

It sounds reasonable, much like the 1.75 limit on early NW and the 1.65 on the later ones.

Though, who are we not to try :D

[hawtrawkr]

id suggest 1.5v or under for really high end air/ avg watercooling and up to 1.6v for high end watercooling/phase with mosfets sinked and cooled.

i just blew up a 4 hour old ic7-g that was sinked and had a fan blowing across the sinks on the mosfets when i was benching with it @ 1.63v blew a mosfet on it and made the vcore droop to 1.1v under a load :-/