What is a 100 percent stable overclock?

Posted on 19th Feb 2009 at 08:09 by Clive Webster with 12 comments

Clive Webster
I’m not sure why a PC that’s less than 100 percent stable is acceptable, but how many times have you read that someone’s got an incredible overclock that ‘works most of the time’, or ‘is fine unless you put too much stress on the system’? A PC that’s less than totally reliable is as useful as a hammer that only works on weekdays or a car that only starts nine times out of ten. You can probably put up with it, but the thing’s broken and needs fixing.

Of course, the magic ‘100 percent stable’ tag is an elusive one. If the PC complete a run of Cinebench R10 but fails to run Crysis for longer than two minutes, is that PC 100 percent stable in Cinebench but not in Crysis?

Also consider whether a PC that’s been successfully looping a 3DMark test for a year is any more stable than a PC that’s been looping the same test for three days. You’d be harsh to say anything other than that both PCs are 100 percent stable, but should the former be considered as being more stable because it’s lasted longer?

When judging the stability of an overclock (something we’re doing all the time) the best method is to set a barrage of soak and burn-in tests for a system to run. If it passes all these tests, then we can be confident that the overclock is stable. If the system fails even one test, that overclock is unstable and should be disregarded. You’d stop to fix a hammer as soon as it failed to knock a nail in, after all.

What is a 100 percent stable overclock? What is a 100 per cent stable overclock?

I’d say that a PC is only as stable as the last stability test performed on it, and that as soon as it fails one of these tests it should be considered unstable. The importance of a stable overclock can’t be stressed too much; if you know your overclock is stable, then any odd behaviour you experience with your PC can be attributed to something else – a new graphics driver, piece of hardware, the moon changing alignment with Venus or whatever. Having confidence in the stability of your overclock helps problem solving when (yes, not if) problems occur.

For the record, our stress testing of PCs consists of running Prime95 across all process threads (either the 32-bit or 64-bit version) and then looping the Deep Freeze test of 3DMark06 for 24 hours. After this test is passed, we also unleash system-wide tests such as Custom PC’s Media Benchmarks (the Gimp test can be a southbridge killer) and then we load a save game in Crysis and leave ourselves to die and re-spawn for another six hours. We’ve found Crysis to thrash graphics card memory in the past, so this is a good gauge of graphics overclocking and cooling.

Temperatures are monitored with tools such as Core Temp, SysTool and GPU-Z where possible.


Discuss in the forums Reply
-Acid- 19th February 2009, 08:49 Quote
100% stable overclock ....... on what part ?

Should that not read a overclocked computer system that is 100% stable over a the users choice of tasks.

Most folding/crunching overclocked rigs are 4/12 stable (down time is loss of Wu's) they tend to Oc lower due to wanting no down time.

Gaming computers tend to be stable for a fixed amount of time ( few hours worth of gaming).

Most people have different classes of what they call stable too.
Linpack over all cores and fur mark covers most bases and are two of the most stressful test going atm.

One thing that you havent mentioned is that many programs are not stable !!!!!! and we are not even getting into drivers.

The Customs Pc media benchmark erm.... that's hardly a stress test its a benchmark test ( the name gives it away). It can run will failing cores and errors in memory fine (just not as well).

This is almost as funny as the how to apply Tim post few days ago, good topic but not thought through.
Tim S 19th February 2009, 09:43 Quote
Real applications are arguably the best stability tests out there. There's a lot of data being passed across the system buses with the media benchmarks and a 24 hour loop of it can pick up problems almost right across the system. The only part that doesn't get stressed by the media benchmark suite is the graphics subsystem.

The important thing when determining stability is to combine a selection of applications and that is essentially the point Clive has made here.
Psytek 19th February 2009, 11:02 Quote
"100% stable overclock ....... on what part ?

Should that not read a overclocked computer system that is 100% stable over a the users choice of tasks.

Most folding/crunching overclocked rigs are 4/12 stable (down time is loss of Wu's) they tend to Oc lower due to wanting no down time.

Gaming computers tend to be stable for a fixed amount of time ( few hours worth of gaming). "

The article would suggest that all parts should be 100% stable.
Claave 19th February 2009, 19:19 Quote
"The Customs Pc media benchmark erm.... that's hardly a stress test"

Actually, the Gimp and multitasking tests are system-wide tests and so really stress the Southbridge and I/O busses - many's the time an overclock has passed CPU-only soak tests only to fail in these as the Southbridge hasn't enough voltage or the FSB flakes out.
specofdust 19th February 2009, 20:26 Quote
There's no such thing as a 100% stable overclock. Even stable unclocked machines can crash given the right conditions. Everything is a matter of degree, how comfortable are you with your systems stability. The focus shouldn't be on zealous application of every benchmark and stress-test known to man, but rather on ensuring your system is stable enough for your needs and desires. I had my E6300 running at 3.3Ghz for a fair old while. Sure, I could run it at 3.0Ghz and probably not crash in years, and running it at 3.3 meant I suffered a crash every month or two - but so what? I got a slightly faster system which made me happy, and since I only ever gamed on it I never lost anything important in crashes. It was fine for me, and that's what counts.
Nicb 19th February 2009, 20:46 Quote
I like to keep my newly built computer at its manufactured settings when I first build it. Install everything I will use and let it run for a few months to see if there are any issues to begin with. Then later I begin overclocking, and tweak over time If need be. I might take 6 months to finish a "truly" stable overclock. If any overclock can be called that.
crazybob 20th February 2009, 18:14 Quote
I don't overclock, but I do undervolt, which has much the same effect on stability. My definition of a stable computer is exactly as strict as the article describes - if it crashes even a single time throughout its entire life, it's broken and needs fixed. It should not make any difference whether I'm running a distributed computing task, playing games, browsing the internet, or just letting the computer sit idle; if it crashes I will find the problem and fix it.

I've never understood almost-stable overclocks, because dropping speeds by a few percent won't make a substantial impact on performance, but can make a huge difference to the appeal of using the computer. My current computer has crashed only a single time in the year and a half since I built it, and this was because of a video driver issue which I quickly resolved. That's the standard I hold my computers to, and even a monthly or bi-monthly crash seems like too much.
brummie 25th September 2010, 21:36 Quote
never been able to pass that test at stock ever, something is going wrong captain.

leexgx 25th September 2010, 23:07 Quote
1-2 weeks stable seems best test for an overclock i am@ 3.83 and its stable (i think it is at 4ghz as well but its at 3.83 due to me trying to work out why only 1 program (game) was not stable (but it was not due to the OC) just never bothered to put it back to 4ghz as i only have 1 video card so 3.8ghz is overkill any way

was trying to work out why Supreme Commander 2 stutters and takes my PC out (bit random some times video card would stop get blank screen still sound some times as TS still works for about 30secs when i guess system BSOD as the system reboots) this has done it on 2 New systems annoying me really as its Quite random but if i get 3 stutters 4th one is an crash, would love to know how to fix this as the system is fully stable apart from that one game (even when system is all stock settings)
thehippoz 26th September 2010, 01:09 Quote
yeah stability is a key thing in ocing.. you can go balls to the wall but it will usually suffer

transfering a few hundred gigs over e-sata is a good stability test too.. I usually prime blend for a few hours- if it passes that then it's pretty stable

got this rig running right now at 4.24ghz and so far looks good.. thing with prime is.. if you can pass blend on all cores it will be good enough to encode.. you want a rig you can leave on 24/7
dvijaydev46 23rd May 2011, 15:21 Quote
No stability test program is 100% accurate. You may be able to pass 10 hours of Kombustor test for <a href="">GPU stress testing</a> but crash immediately when loading a game. You have to use all apps at your disposal to ensure stability.
vdbswong 23rd May 2011, 16:29 Quote
Necro Much?
Log in

You are not logged in, please login with your forum account below. If you don't already have an account please register to start contributing.

Discuss in the forums