Best prog for stability

P

Pivo504

2[H]4U
Joined
Feb 18, 2005
Messages
3,105
What is the best program for testing stability of an overclocked PC? I used to use Orthos Prime but is there anything better out now?
 
orthos, occt, Intel burn test. 3dmark0x, and using the programs that you use normally. after testing it, use the pc normally as you would and see if anything happens.
 
Folding? :p I ran intelburntest, prime95, and another one for 2 days each application and none of them crashed. 6 hours of folding and pc would crash, I had to go back to stock to fold, I don't know why but no matter what the overclock was (even a measly 200mhz) would crash folding and that was it, games were all fine. Try a bunch because just because one passes doesn't mean it is stable.
 
meh ive done alot of testing recently and in my experience occt is just a mild stress tester, intel burn test is a quick and fairly accurate test but 12 hours of prime 95 is still my final test for stabilty
 
Thanx everyone! Is there a 64 bit version or the 32bit is fine for my 64 bit windows?
 
Other tests have their own purposes they try and serve but the truth of the matter is that LinX will stress and strain your system more than Prime/OCCT/Folding ever will.

http://www.xtremesystems.org/forums/showthread.php?t=201670

IntelBurnTest is more or less a re-skinned form of LinX and should be equally as demanding but if you're able to run 20 passes of LinX using all available memory you shouldn't have issues with much else. I never believe in putting all my eggs in one basket but Prime (Blend) is more of a backup test at this point that I'll run for 12-24 hours after passing 20 runs of LinX just to comfort my paranoia. ;)
 
well my rig is easily passing ITB on some settings 20-60 passes and then fails prime 7-9 hours in. have to set it back to 472 for both to be stable, so for me at least prime 95 seems to be more stressfull than IBT
 
Folding is really good. LinX is also really good. I used to use Prime, but it just seems way too time-consuming for what it's worth; if I find an instability during normal usage I just bump the voltage up another notch. There's no need IMO to run a program continuously for days on end to check for stability.
 
RM_Bulldog said:
well my rig is easily passing ITB on some settings 20-60 passes and then fails prime 7-9 hours in. have to set it back to 472 for both to be stable, so for me at least prime 95 seems to be more stressfull than IBT
Click to expand...

I would try LinX instead of IBT, the settings within IBT do not completely replicate LinX and there have been a number of people who have issues with IBT (including myself) that are not reflected on their system stability. LinX is the much more widely accepted test and I would urge you to give it a try.

20 runs using all available memory and you should be fine, my system does it just fine but continues to have issues with IBT...and it's far from being my opinion alone the IBT isn't as reliable in some cases.
 
Ususally I just play some intense games to make sure I'm stable. But OCCT is my go to stability tester if I need to check it quickly.
 
bloop said:
if your apps run without error or issue, you are stable.
Click to expand...

This is a common school of thought but you really don't know if or when something will happen. I'd much rather spend a few hours stresstesting my system then deal with finding out the hard way when it locks up in the middle of a game or something more important.

Not having to worry is more than worth the time spent ensuring that you're good and stable...don't say you weren't warned. ;)
 
If heat output is any indication, CoreDamage is among the most stressful. More people are starting to use this in combination with Prime 95.
 
Not only does Core Damage appear to be purely overkill...LinX still proves to be one of the most reliable and updated options that's actually recognized by Intel. LinX allows you to load down 100% of your available memory as well and will still produce more of a heat load on your CPU then Prime95. If you use your system to game/benchmark and it'll typically never see the temperatures or stress levels generated by Prime and LinX I'd like to know what the point is in pushing it so far that your overclock is severely gimped to compensate for the artificial scenario.

I'm not one to skimp on stresstesting but with as proven as Prime95 and LinX are and with how much heat/stress they produce there is absolutely no reason to push your CPU to these extents. It'd be different is someone was using sub-standard methods and was being recommended to use Prime95 or LinX but when you're already using one of the most reliable methods available you stand to gain nothing from adding another 10-20 degrees to the equation.

Sorry man, not a person attack but there's no good reason to use anything more than LinX and Prime95...I don't care if my overclock would fail at 90 degrees because it'll NEVER be running that hot. If you can make it through 20 runs of LinX and 12-24 hours of Prime95 "Blend" there's no realistic reason to go any further.
 
You've got a good point-- as long as we're not interested in stressing the cpu to the theoretical limit (wouldn't want that overclock fail) perhaps it is more prudent to run a stress-less test.
 
cornelious0_0 said:
Not only does Core Damage appear to be purely overkill...LinX still proves to be one of the most reliable and updated options that's actually recognized by Intel. LinX allows you to load down 100% of your available memory as well and will still produce more of a heat load on your CPU then Prime95. If you use your system to game/benchmark and it'll typically never see the temperatures or stress levels generated by Prime and LinX I'd like to know what the point is in pushing it so far that your overclock is severely gimped to compensate for the artificial scenario.

I'm not one to skimp on stresstesting but with as proven as Prime95 and LinX are and with how much heat/stress they produce there is absolutely no reason to push your CPU to these extents. It'd be different is someone was using sub-standard methods and was being recommended to use Prime95 or LinX but when you're already using one of the most reliable methods available you stand to gain nothing from adding another 10-20 degrees to the equation.

Sorry man, not a person attack but there's no good reason to use anything more than LinX and Prime95...I don't care if my overclock would fail at 90 degrees because it'll NEVER be running that hot. If you can make it through 20 runs of LinX and 12-24 hours of Prime95 "Blend" there's no realistic reason to go any further.
Click to expand...

+1. Linx is completely fine. People recommend 20 passes, but i've found that I can pass 5 passes, and I've never had any issues, however, I don't fold or do anything necessarily CPU intesive apart from gaming.
 
Wow core damage really puts it to it :eek: fired it up and temps went up to 96C! Cudos to Acer for putting some killer cooling on this i7 lappy

 
You must log in or register to reply here.
Back
Top