![]() |
|
|
#1 |
|
41×157 Posts |
Hello,
I wanted to know how for how long should i run the torture test to ensure that my system is going to be rock solid ? I ran the test for continuous 7 hours at a stretch and didnt get a single error at all. Does this prove that my system is quite solid ? the specs are as follows. 3 modules of ram - 2 *128 1 module of 256 Amd athlon xp 1800+ Asus A7V266-M motherboard Radeon 8500 128 MB There are quite a few options in prime95 torture test , which one should i be selecting ? thank you. |
|
|
#2 | |
|
Nov 2003
European Union
23×13 Posts |
Quote:
For a non-overclocked system I would accept a 24-hour torture for declaring it stable. For overclocked systems I would require 48 hours or more depending on frequencies, voltages and cooling used. With an 7-hour test, you can be only partially assured that the system is stable. You may be happy with the 7-hour test if you are willing to accept just a 40-60% assurance that the system is stable. Last fiddled with by optim on 2004-08-27 at 17:24 |
|
|
|
|
|
#3 |
|
"Patrik Johansson"
Aug 2002
Uppsala, Sweden
52×17 Posts |
I would run it for about 24 hours, but if you are going to use it for testing prime numbers, that is not enough.
If you want to test prime numbers I would recommend that you first do a number of double-checks (under Test->PrimeNet, takes a week or so for each test), and wait 1-2 weeks until the status files (bottom of this page) are updated. If a test matches the original test, it turns up in lucas_v.zip and your computer made no error during this test. If it ends up in hrf3.zip either your test or the original one had an error. The first time tests take a couple of weeks, so it is important that no errors occur during this time, since a single undetected error spoils the whole test. But the time you should spend on the torture test also depends on other things. If you are overclocking and it fails the test when you have the FSB a few MHz higher, then you should be careful and run it for a longer time. EDIT: Optim was quicker than me, but I see that we agree on 24 hours. Last fiddled with by patrik on 2004-08-27 at 17:27 |
|
|
|
|
#4 |
|
22·52·79 Posts |
I am not overclocking the pc at all. 24 hours should prove it stable i guess then.
|
|
|
#5 |
|
142268 Posts |
I ran mine overnight and it got 1 error 7 hours 2 minutes later ...looks like 7 hours isn't long enough. Guess I'll play with my STOCK RAM settings to correct this.
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Prime95 - Torture Test | doffdoff | Software | 12 | 2012-05-04 19:53 |
| # of h for p95 torture test to be "stable"? | graysky | PrimeNet | 2 | 2008-05-03 22:02 |
| prime95, torture test, and stability | polyestr | Hardware | 6 | 2006-08-12 12:45 |
| Torture Test - System running processor very low compared to other systems | DougTheSlug | Hardware | 5 | 2005-01-27 09:51 |
| Help making system stable | bartok | Hardware | 7 | 2003-07-06 07:05 |