• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Prime95

D

dnas

Weaksauce
Joined
Jun 7, 2004
Messages
66
What's an acceptible error rate with Prime95?

I've overclocked a processor, and I get an error after between 20 & 50 sets of iterations.
(between 15 & 30 minutes I guess, long after the CPU has heated to its maximum under load temperature.)
 
It depends.

Technically, you should never get an error at all.

However, Prime95 stresses a CPU way more than any other app that you'll be running. (Including F@H, as far as I have seen) If all of your other programs run without problems, then your overclock is relatively safe. It's not 100% solid, but if it works...
 
I've' just gotten a Celeron D (320) 2.4G, and I'm running it at 3.6G, stock fan. It's a D0 stepping, not the C0 stepping I've put into a few people's PC's.
A C0 I had would run at 3.4G, but it would not run Prime95 for even one iteration without an error.
 
0% error rate is acceptable to me.

A processor should never make an error, and neither should the memory. I would hate to have something like that corrupt important files, as I have learned once.

Perfection > Speed.
 
Talonz said:
0% error rate is acceptable to me.

A processor should never make an error, and neither should the memory. I would hate to have something like that corrupt important files, as I have learned once.

Perfection > Speed.
Click to expand...

QFT...
 
Talonz said:
0% error rate is acceptable to me.

A processor should never make an error, and neither should the memory. I would hate to have something like that corrupt important files, as I have learned once.

Perfection > Speed.
Click to expand...

This may or may not happen. You could lose everything in an unlikely event, or be perfectly fine. YMMV.
 
MemoryInAGarden said:
This may or may not happen. You could lose everything in an unlikely event, or be perfectly fine. YMMV.
Click to expand...

Specifically, that was because of a high PCI bus freq, but the chance of having an error isn't even worth it IMO.
 
0 errors... no other way... seriously.. a CPU is supposed to work correctly 100% of the time.. If it can't handle what you dishout.. IMO it's not good enough..

i have too much to lose..
 
yea, there is no acceptable failure rate.

Bottom line, if it failed P95 at stock speeds, I'd make them RMA the chip, so why should an overclock have an acceptable failure?

Kinda like Memtestx86. I've gotten that question before as well...
 
if you want your processor to be stable, it should do at least 24 hours of prime with no errors, if you want extra assurance, run it for 48 hours. i run prime95 for a few days on every new cpu i get. i have had a processor fail at stock speeds after 30 or so hours and noticed that it did not overclock very well at all and that the replacement processor overclocked much higher.
 
Talonz said:
Specifically, that was because of a high PCI bus freq, but the chance of having an error isn't even worth it IMO.
Click to expand...

It wasn't because of a high PCI bus frequency, which was the normal speed.

It was because of the overclock of the CPU (Celeron D). FSB from 533MHz -> 800MHz.
(2.4G -> 3.6G) It runs fine at 3.5G, but you don't get the 200MHz memory bus.

I guess the errors are hardly surprising given it is a 50% overclock, stock fan stock voltage.

I had did read somewhere that Prime95 taxes the CPU more than any real world application, so the occasional error was acceptable..... however, judging from reactions here, it seems it's better to eliminate them completely.
I'll tweak the voltage to see if I can eliminate the Prime95 errors completely.
I could also get a better fan, but it tops out at 50 degrees C under Prime95 load, with the stock fan.
 
dnas said:
It wasn't because of a high PCI bus frequency, which was the normal speed.

It was because of the overclock of the CPU (Celeron D). FSB from 533MHz -> 800MHz.
(2.4G -> 3.6G) It runs fine at 3.5G, but you don't get the 200MHz memory bus.

I guess the errors are hardly surprising given it is a 50% overclock, stock fan stock voltage.

I had did read somewhere that Prime95 taxes the CPU more than any real world application, so the occasional error was acceptable..... however, judging from reactions here, it seems it's better to eliminate them completely.
I'll tweak the voltage to see if I can eliminate the Prime95 errors completely.
I could also get a better fan, but it tops out at 50 degrees C under Prime95 load, with the stock fan.
Click to expand...

dnas, I realize that. I was referring to the one time I had data corruption that I mentioned above. Yea, a voltage bump should fix that problem for you, have fun ;)
 
You must log in or register to reply here.
Back
Top