How long is long enough? Prime 95

S

sechkie

n00b
Joined
May 10, 2006
Messages
46
Ok, so how long do you usually try to be "prime stable" for before just going into everyday use?

With my new 8400 and Gigabyte DS3L I've got it set at 4.0 ghz and BIOS Vcore of 1.38785.. the one right below 1.4. (CPU-Z shows 1.36) I get to about 17 hours of Prime before core 1 craps out.

Temps are fine, never goes above 55c on a Noctua in Push/pull. Memory is fine after an overnight run of memtest. I've bumped up Northbridge and FSB voltage by 0.2 in bios.

Should I push for 24hr stability or just go ahead and use it? (Normally just gaming, nothing intensive like video encoding)
 
Small FFTs will give you the highest CPU temps, and I'd say it's most important. Blend will stress the memory more, but Memtest is better for testing the memory anyway. Of course, maybe there are problems that only appear when both the CPU and memory are reasonably stressed, which would be an argument for running Blend too, but I'd be reasonably satisfied if Small FFTs and Memtest are okay.
 
I normally recommend 24 hours @ small FFTs. I have personally seen 12 hours Prime95 and still not totally stable depending on the application and what I am putting it through.

 
It is a personal thing, do what makes you feel comfortable as being stable is in the eye of the be-holder. I personally do eight hours and I have never had a problem. But I do not run my system at 100% very often. Others that fold or encode at lot may be better served by going for a longer period. Others consider there system stable if it can game without crashing. Is that stable, not to me.but if it serves them then so be it. If you are comfortable with 17hrs then let it be, if not keep tweaking.
 
I was just going to post a poll about this very subject...

How long do you let your CPU burn? I understand people can get "hot" (pun intended) when talking about the validity (or lack of validity) of stability testing with these programs. I used to be a "Damn Prime95, my computer runs fine!" person. Truthfully, the only problem I have had with my machine was a mobo failure that the maker (MSI) took full responsibility for and basically admitted it was a QC slip that allowed a board with a bad PCI lock out of the factory. Many of my friends subscribe to this view, which makes logical sense in the following way:

If my system always is able to do what I use it to do when overclocked, and I don't experience any problems, there isn't a problem.
That is, until your system flames out. There's no guarantee that this will happen. In fact, many will never push their system as far as Prime95 does. People complain about "synthetic" benchmarks that reveal very little if anything about how a computer will perform when playing an actual game or running a real application. In a way, Prime95 represents a synthetic system load. No game or application that I know of short of some distributed computing projects (such as F@H or SETI@Home) push a CPU this hard. Even still, a computer can be "Folding Stable" but "Prime unstable." So Prime still requires more cajones than Folding...

If you're not running those applications and you've transcoded movies and played games, you'd think your system is fine in its overclocked state. However, the Prime people (and I am increasingly one of them) come back with this:
If your system can't survive this load, then it is more than likely that a lesser load will be the straw that breaks the camel's back. Of course, Prime95 could be the only straw that breaks the camel's back, but it is possible that some "lesser" task will bring your computer to its knees. However, if your computer can survive Priming then it will definitely survive everything else you throw at it.

I think some people dislike Priming or other testing like it because it takes so much time. With some hardcore overclockers recommending 24+ hours of running an application that chews up all your cores and makes your computer virtually unusable, it's understandable why people would grow impatient. "I didn't build this computer to watch it error-check itself all day!" they say. Some people say that 8 hours is sufficient, that anyone else is crazy. Others opine that tests often fail in the 16-20 hour range and those failures will NEVER show up earlier. Generally, very few people will argue that 24 hours is too little.

Others dislike Prime95 because it tells them something they don't want to hear: your OC is wrong or too aggressive and won't hold.

What do you all do? My Q6600 is currently Priming at 3GHz and has been going for seven hours strong. I personally find 24 hours to be excessive...
 
I usually do 4-8 hrs. I think if it will fail, it will be within 4hrs. It should at least be stable enough to run windows and the things I use it for
 
tuksonrider said:
Blend or Small FFT?

They make a HUGE difference in temps.
Click to expand...

Running blend or small fft actually gives me similar temps according to realtemp..


This is strange... one time I run it, it goes 17 hours and then fails.. another time I run Prime, it goes only 30 mins, another time it goes 4 hours...
 
I did this before declaring my OC stable:

12+ Prime95 Small FFTs
12+ Prime95 Blend + Furmark + Winamp loop running
12+ Orthos Small FFTs
12+ OCCT Blend + Furmark Winamp loop running

Most of the tests went close to 20 hours due to my schedule. I ran some other combination too for a few hours each, so the total test time was over 5 days with all four cores at 100%, no errors. Running Furmark with something that stresses the CPU is also a good idea since it creates more heat in your system and taxes the power supply. You may find that once you cause and extra ~90+ watts to get drained from your PSU, you get some system voltages dropping a tiny bit and that can cause a stability problem. I run Winamp just to keep the sound card busy too.

Once all that passed, then I downclocked my FSB by 4Mhz for some extra head room. All of that testing was done at 370 FSB and I run at 366 now. If at any point during any of this a problem came up, I would have declared the whole OC trash and started over.

Fact is once your system gets up and running, new variables start to creep up like dust on fans, AC going out, and components slowly wearing. Also you end up running lots of different software that may put a slightly different load on a CPU that you would test for with Prime95, and the whole thing craps out.

Everyone runs their system differently and cares about the data on the system to varying degrees. While I have nothing important on my new build right now, I do not want to have a problem months down the road with a bleeding edge OC while I am doing some real work on the system and have some bad data end up on the HD or an app crash when I am just finishing up a project. People have different comfort levels, and I would rather be 99.99% sure that my OC is stable rather than push for an extra 1% performance and increase my chances of running into a problem later.
 
sechkie said:
This is strange... one time I run it, it goes 17 hours and then fails.. another time I run Prime, it goes only 30 mins, another time it goes 4 hours...
Click to expand...

Thats because once the CPU gets up to temp, the first test on Prime95 will be just at likely to fail as the last. Its not like the system load goes up or down over time. Its the same functions getting performed over and over. Running Prime95 for longer periods of time just increases the chances of a problem showing up.
 
Here's what I say. The more the better but you only have enough hours in the day to do certain things. If you are trying to achieve a stable overclock and test it for 24+ hours, well there goes gaming or something else that would have been enjoyable on your computer. Also, what happens if it fails after 10hrs? You change some settings, then check to see if that's stable. And then repeat.

Basically, technology is depreciating every second. The longer you keep testing for a stable overclock, you'll find that your so called top of the line CPU has now been replaced by something better.


But then again, I think this really applies to those who change their FSB by a few Mhz and then tests right afterwards if it was stable. *points to self*
 
SolidBladez said:
Basically, technology is depreciating every second. The longer you keep testing for a stable overclock, you'll find that your so called top of the line CPU has now been replaced by something better.
Click to expand...

Interesting concept, though honestly I do not settle into a new build for a couple weeks anyway, so taking almost a week to make sure the OC is good is fine with me.
 
How do you loop 3dmark06? Can you do it via the basic free version or do you have to buy the advanced version?
 
Under "Settings" click "Change" and then under "Repeat & Loop" select "Loop all selected tests". Took me 5 seconds to figure it out. Although I don't have the basic version.
 
[21CW]killerofall;1032393518 said:
Under "Settings" click "Change" and then under "Repeat & Loop" select "Loop all selected tests". Took me 5 seconds to figure it out. Although I don't have the basic version.
Click to expand...

Yeah, I saw that option, but as I'm using the basic version, I can't get to it. I wondered if there was another way. Guess not.
 
You must log in or register to reply here.
Back
Top