Your Stability Testing Routine

Soldato
Joined
16 Nov 2010
Posts
16,513
Location
Swimming in a lake
Title basically says it all, what do you guys all use to test the stability of your system?

I apologise in advance if there is already a thread similar to this around somewhere, but yeah....

Also am quite intrigued as to what, if they're allowed to post it, the stress testing scheme on OCUK's overclocked bundles are - More curious than anything else.

kd
 
  1. Boot to BIOS
  2. Boot to windows
  3. Superpi 1m
  4. 5 passes of Linx (1gig of memory)
  5. 20 passes of linx (ALL memory)
  6. Prime 95 for apx 12hours
  7. Various gaming sessions with HEaven/Vantage benchmarks

Once the above is done I re-install windows/apps/games as corruption may have occured while unstable system was being tested.

Never let me down.
 
I use IBT for quick testing 5 passes on maximum. Then when you think you have your final settings, 20 passes on maximum

If the above passes OK then it gets over night tested with OCCT, (CPU:Linpack set to max and 8 hours set)
If it passes that then I consider it good to go :)
 
You guys think too much. SuperPi 1m, five passes of Prime95, then set it folding. Worst case, I lose a work unit, best case, it folds perfectly and that's a much harder test than any artificial test
 
My own system: current testing method
600 loops of IBT
20 loops of Linx
48 Hrs prime 95 Blend
1hr OCCT gpu test at native res and max shader complexity, 1hr power supply test.
few loops of 3dmark

Might be overkill but its rock solid. I don't depend on the system so i have time to test, i use two pcs mainly.

Clients system: 100 runs IBT
24 hrs prime 95
If its a gaming rig i'll do OCCT GPU for an hour, same for psu test.
Few loops of 3d mark
 
10 hours of Prime95 small FFTs stress test
10 hours of Prime95 Blend test
20 passes of Intel Burn Test


For gpu:

30-35mins of Furmark or MSI Komboster
 
Everest to find cpu temps and game whilst checking gpu temps.

Then...use it like I normally would.

I can appreciate if you are sorting other peoples rigs the below will not apply in quite the same way;

I dont get this whole stress test for days- whenever I have actually bothered to run a few hours stress testing I end up playing a game for ten minutes and getting a crash.

As per using furmark, to my mind when the gpu makers start putting limiters in...thats a sign that something isnt right with that test, given my vrms heat up to 100C (I stop it then) after a few minutes, even with the beefiest vrm cooler money can buy using extreme furmark, it is just a plain bad idea to my mind. God knows how many people are sat their grinning after fifteen minutes of furmark as they see a 70 degree core, whilst their vrms are doing 120C+.

Real life tests the best, you are not just sat there revving the proverbial engine and why have your pc out of action for a day when you can have fun using it at the same time? :)
 
Last edited:
When in the process of overclocking I do 20 IBT runs and 30mins of Prime95 Blend. When I think I have it stable I do 20 IBT runs after a cold boot and after idle for 30mins. I then prime for as long as I can, I have had overclocks show instability after 20hours you see. If that passes I just generally game and benchmark for a few days.

An overclock system for me is something that can do everything an unoverclocked system can. If it fails after 3days of Prime95, then its not stable.
 
Wow @ some of the time taken! I used to OC it then run superpi, then do a few hrs of gaming something intense. Got an s939 3800+ running at 2.8 for a few years like this. :)
 
You can run prime for 7 days and still get a crash during the 5th or 6th day;).

This is true. I run prime for as long as possible to spot these mistakes but usually I want my PC back after a day or two. Usually I have found that if it will last 24hours, it will last anything but there have been some expections to that.
 
Back
Top Bottom