What do you consider Prime95 Stable?

What is Prime Stable?

  • Small FFTs only for 24hrs

    Votes: 14 43.8%

  • Large FFTs only for 24hrs

    Votes: 5 15.6%

  • Small and Large FFTs for 24hrs

    Votes: 3 9.4%

  • Small/Large FFTs for more than 24hrs.

    Votes: 2 6.3%

  • If it runs for a little, its stable to me.

    Votes: 8 25.0%
  • Total voters
    32
I

Ingonuts13

[H]ard|Gawd
Joined
Jun 9, 2003
Messages
1,982
So I am curious what people think "Prime Stable" is to them.

1)Small FFTs only 24hrs?
2)Large FFTs only for 24hrs?
3)Small and Large FFTs for 24hrs?
5)More than 24hrs.
4)If it runs any test for a few hours it's stable to me.

I know there is some debate on this issue so I would like to get some discussion going on the subject. My machine runs prime stable on small FFTs for days but it seems that Large FFTs only run about a day or so.
 
_Korruption_ said:
Blend, 24hrs+.
Click to expand...

Hmm...I never really considered that. Seems to me that blend does not really stress the CPU all that much but hits the RAM really hard.
 
Ingonuts13 said:
Hmm...I never really considered that. Seems to me that blend does not really stress the CPU all that much but hits the RAM really hard.
Click to expand...
i don't find that any test other than blend does any good: i've had unstable systems pass small or large FFTs.
 
"4)If it runs any test for a few hours it's stable to me."

that's how i do it. when i was undervolting my dothan, this is how i did it, just keep it running in background and lower voltages. if you change it lower and it fails then obviously you go back up in voltage. i leave my laptop on all the time and it's never crashed during gaming or F@H and i've never run prime95 24/7 on it. i think it differs for you overclockers. ;)
 
blend for around as long as i want it to be...but over 10 hours generally...
 
In-place Large FFTs (max heat, power consumption, and some RAM tested) for 8 hours and a Blend for 8 hours after that...

My reasoning: If the CPU can handle getting romped on for 8 hours... the CPU is stable... If the RAM can handle 8 hours with lots of testing, its stable (these are imo)...
 
Prime in itself is ok to test initially, I like to see 5 to 6 hours minimum.....but to really find out...run some of the most demanding games for a few hours....encode a few multi gig videos....Fold@Home for a couple days straight.....

It's all good after that....
 
I agree with some of the other guys. Generally 8-10 hrs is the limit of my patience anymore, plus any longer lowers my ppd in F@H. Since I let folding run 24/7, if it doesn't bork any work units,it's stable for me.
Add another option for "At least 8 hours of Prime95" and You'll get more votes.
 
I usually do just solid small ftt's to make sure its testing just the cpu for 24-30 hours. I use memtest to stress the memory, thus the 'blend' test is useless to me. :p
 
You must log in or register to reply here.
Back
Top