• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

5.5 hours Prime Stable

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Strages

Member
Joined
Feb 26, 2003
Location
Stockton, CA
Yay! I just got back from a date and noticed that my computer ran Prime95 torture test for a tad over 5.5 hours straight!!! I was quite happy to find this out, but also kind of concerned because it crashed after 5.5 hours (I was gone a total of about 7 hours), it didn't restart or anything, just froze. I restarted the computer and everything was fine. I'm pretty ****in' happy with this thing so far - good overclock results for me (longest running stability I've yet tested). However, is this a bad sign?? I'm thinking I'm comfortable with that overclock, however, should I be?
Just looking for some second opinions. Any would be a great thanks.
 
usually, people are happy when prime runs for more than 12 hours with no errors. stability is what you want it to be though, so however you feel comfortable. but i wouldn't be excited that your computer froze after 5.5 hours.

in my experience, i had my computer running at 3.5, but it wasn't very stable (prime froze after 3 hours). i liked being able to tell people my computer ran at 3.5 (partially to rationalize my purchasing $200 in watercooling equip), but i backed it down now.
 
wasn't that I was excited it froze, I was simply pleased that it ran so long stably - because I honestly didn't expect it to last an hour, lol
 
Sure it would be great to run the stress test for ever. Just to say "hey guys I'm stable!!!!!!"
I think running Prime for that amount of time is plenty. Also there are certain versions of the test that really stress the system and or cpu more so than the others.

To give an idea I ran Prime for some time a while ago. That was good enough for all intensive purposes.. :D

Prime2371.gif
 
I disagree. The way I stress my computer sometimes (for things that can't have errors), 5.5 hours prime stable is not good enough for me. If you just play games a little and don't care if thay have a hardware related crash once in a reat long while, then you'll be fine. If you actually want it to be stable, you should work on it. Try bumping voltages, increasing cooling, and decreasing overclock.
 
To add to my previous comment.

Personally, I can run Progs such as Prime for days on end without errors... Even with the 3GHz+ Barton setup it does this. But that's what it was built to do.

Just remember that Prime can find faults not just in the cpu but also other components like your memory can affect it's ability to function without reporting an error.

Points made earlier are valid concerning voltages, fsb speeds and very importantly cooling.

I'm not certain which Prime you tested with. You should try the version in the pic that was posted (23.7.1) There are options to hit the cpu or ram or a combination. That way you can help pin point where it is your system might be failing....

Good luck, I'm off to do some coding for NASA. ;)
 
5.5 hours stable on prime95 which mean...your system is not fully stable to performance some stuffs.

i usually let it run over night..12hrs+ then next morning :) i see n error i know its stable and go for more. just like right now..i woke up and see my p4 run 2xprime95 stable for 13 hours straight the combination of cpu+a lot of ram testing torture. i'll aim for 3.5ghz today on this 2.6c

w00t!
 
Damn. I usually leave my system running over night, and apparently it froze overnight :(

Still working on this damn o/c, I want 2.4 dammit... my core voltage is already at 1.9v, considering pushing it slightly higher, though I'm thinking of loosening my ram timings or running memtest to see if it's my ram (which is still a possibility). Anyhow, time to get ready for work :rolleyes:
 
consider a better cooling to bring down voltage?

throw in case fans to keep the temp down would help in certain degrees too.

1.9v is pretty high..i never hit over 1.6v yet
 
1.9v is high, that's why I'm only considering pushing it higher, but I don't really know how much higher I *should* take it. I've got an Enermax 460 noisetaker, so my PSU can handle the power fine and the NF7-S can handle 2.1v+ with relative ease, however, I'm dissipating 99w right now, which is already high, but keeping temps at max of 42C, which is really good for air cooling.
Like I mention in my profile, I'm looking into new cases right now, I'm not a n00b like my profile might suggest, been on these foums since '01 (look for SUlfUR-CyCLoNe, lol), but I only had $60 with me and *had* to get a case (for unfortunate reasons), so I got this ****ty one... but I'm looking at a Coolermaster WaveMaster right now, I'm just planning out how I'm going to mod the damn thing... it's cooling isn't quite up to par for me at this present moment, plus I have to design the side panel I want and I have to make sure that I have enough clearance for the massive Tornado (it BARELY clears the 92mm fan on the bottom of the noisetaker, I literally had to remove the grill to get it to fit).... so I'm kind of o/c'ing with a limp so to speak, lol... once these issues resolve I'm hoping for 1) more stability right now and lower temps or 2) ability to hit higher voltages without sacraficing my current good temps. Later.
 
Back