Page 1 of 2 12 LastLast
Results 1 to 16 of 20

Thread: Prime 95 Torture Test

  1. #1
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts

    Question Prime 95 Torture Test

    I've just downloaded prime95 to test the stability of my system after trying an overclock. But, when I go to "run torture test", it comes up with four options. Which one should I run?

    1. Small FFT's (maximum FPU stress, data fits in L2 cache, RAM not tested much)
    2. In-place large FFT's (maximum heat, power consumption, some RAM tested)
    3. Blend (tests some of everything, lots of RAM tested)
    4. Custom - I'd have to change the settings, but I have no idea what they mean!

    Also, am I right that I should run it for 24 hours to be 100% sure it's stable?

    Thanks

  2. #2
    Large Member
    Join Date
    Apr 2004
    Posts
    3,720
    Thanks
    47
    Thanked
    99 times in 64 posts
    It says on each what it stresses the most, so if you want to check the stability of a particular item you can do that, if you want to test general stability, just go with 'blend'. 24 hours is to make sure your system is completely stable, yes.
    To err is human. To really foul things up ... you need a computer.

  3. #3
    Cute & Fluffy GreenPiggy's Avatar
    Join Date
    Apr 2004
    Location
    Cardiff
    Posts
    1,196
    Thanks
    5
    Thanked
    9 times in 8 posts
    3 will give you the more accurate real world usage - recommended if you are looking for general stability. Only use number 1 if you are positive that the cpu will be the only thing that needs testing.
    Knight 1: We are now no longer the Knights who say Ni.
    Knight 2: NI.
    Other Knights: Shh...
    Knight 1: We are now the Knights who say..."Ekki-Ekki-Ekki-Ekki-PTANG. Zoom-Boing. Z'nourrwringmm.

  4. #4
    pre
    pre is offline
    Member
    Join Date
    Aug 2004
    Posts
    171
    Thanks
    0
    Thanked
    0 times in 0 posts
    3 is recommended because it stresses the memory subsystem fully (CPU, memory controller, large amount of the RAM used). It does do Small FFT testing as well but its spread out over the time you run the blend test rather only doing them as test 1 does.

    24hr is recommended but in general if your time is limited 12hr good enough for all but the most stubborn of people.
    Last edited by pre; 06-11-2004 at 02:45 AM.

  5. #5
    HEXUS.social member Agent's Avatar
    Join Date
    Jul 2003
    Location
    Internet
    Posts
    19,185
    Thanks
    738
    Thanked
    1,609 times in 1,048 posts
    Lightweights

    I tend to run 2 copys of prime, one doing option 1, one doing option 2
    Quote Originally Posted by Saracen View Post
    And by trying to force me to like small pants, they've alienated me.

  6. #6
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    Right, I've run number 3 a few days ago, but it fails after a few minutes. So I asked for help on this board. You said change the vcore setting. So I changed it to 1.70 (i think), but it still fails test number 3.

    So just now, I've tried number one - and it worked and it's been running for a few hours now (I've got up to 28K test)! In the description, it says number 1 doesn't test memory so much, so could that be that problem, and if so, how do I fix it?

    I don't know anything about memory timings or whatever though, so all help must be step by step if possible .

    Thanks again.

  7. #7
    Large Member
    Join Date
    Apr 2004
    Posts
    3,720
    Thanks
    47
    Thanked
    99 times in 64 posts
    Sorry, what part of the system have you overclocked?

    Basically if it fails, keep increasing the voltage of whatever you have overclocked until it is stable (within reasonable limits ). My Barton 2500+ isn't happy at 3200+ speeds unless it's at 1.85-1.90 vCore.
    To err is human. To really foul things up ... you need a computer.

  8. #8
    iMc
    iMc is offline
    Senior Member iMc's Avatar
    Join Date
    Aug 2003
    Location
    Northants
    Posts
    3,616
    Thanks
    2
    Thanked
    0 times in 0 posts
    My comp fails the 3rd test at everything stock. varys how long it take to fail though, anywhere from 1 second to 25mins. Had it run for 6hours before w/o failing but could never do it again. Strange
    HEXUS|iMc

  9. #9
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    Quote Originally Posted by yamangman
    Sorry, what part of the system have you overclocked?

    Basically if it fails, keep increasing the voltage of whatever you have overclocked until it is stable (within reasonable limits ). My Barton 2500+ isn't happy at 3200+ speeds unless it's at 1.85-1.90 vCore.
    I haven't actually overclocked anything. I did make a thread about this a while ago called "Normal 2600 Speed?", but now I'm using this one because the title's better.

    Anyway, I wanted to try prime95 to check my sytem was stable under default settings at it's normal speed (1.91 GHz). I did this under setting 3 and it failed. So I upped the voltage to 1.60 (it was set to H/W). This changed nothing, but I am now able to run the 1st test. I can try upping the voltage again, but can it harm the system if I go too far? I think that maximum I can do is 1.80V.
    Last edited by ajbrun; 10-11-2004 at 12:19 AM.

  10. #10
    Large Member
    Join Date
    Apr 2004
    Posts
    3,720
    Thanks
    47
    Thanked
    99 times in 64 posts
    Well perhaps as a reference, the default speed of my Barton 2500+ is 1.84GHz, but the default vCore is 1.650 volts. Is yours a thoroughbred?
    To err is human. To really foul things up ... you need a computer.

  11. #11
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    Mines a barton. What does that mean?

  12. #12
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    Also, my temperature is 59C. Is that OK?

  13. #13
    HEXUS.social member Agent's Avatar
    Join Date
    Jul 2003
    Location
    Internet
    Posts
    19,185
    Thanks
    738
    Thanked
    1,609 times in 1,048 posts
    Its on the warm side, but well within AMD's guidelines for opperation.
    You can only eliminate another thing if you can bring the temp down though
    Quote Originally Posted by Saracen View Post
    And by trying to force me to like small pants, they've alienated me.

  14. #14
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    Quote Originally Posted by Agent
    Its on the warm side, but well within AMD's guidelines for opperation.
    You can only eliminate another thing if you can bring the temp down though
    What did you mean in your last sentence there? I'm sorry - I don't understand that at all. I should also point out that the temperature was taken when it was doing test 2 where it produces maximum heat and power consumption. It's now at 44C.
    ________________________________________________________
    I've just run test 2, and that failed after 2 minutes. So, it's like this:

    Test 1: Successful
    Test 2: FAILED
    Test 3: FAILED

    Any idea on how to get 2 and 3 working?

  15. #15
    Member
    Join Date
    Mar 2004
    Location
    Peterborough
    Posts
    67
    Thanks
    0
    Thanked
    0 times in 0 posts
    Ok

    I've not read through the thread as I am tired and eyes hurt.
    Can you post your full spec and your memory timings for me.
    If you have CPU-Z, a screenshot of the main screen and memory screen will be good, then we can go from there.

  16. #16
    Senior Member ajbrun's Avatar
    Join Date
    Apr 2004
    Location
    York, England
    Posts
    4,840
    Thanks
    4
    Thanked
    25 times in 13 posts
    I can tell you haven't read through the thread (no offence).

    1. My spec is my sig
    2. I said earlier I don't know anything about memory timings, so I have no idea there - sorry
    3. I don't have CPU-Z, but am willing to download it. I might download it, I will if you link me

Page 1 of 2 12 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Pakistan carries out new missile test
    By Bunjiweb in forum General Discussion
    Replies: 21
    Last Post: 20-08-2006, 09:38 AM
  2. Hazard perception test
    By dkmech in forum Automotive
    Replies: 12
    Last Post: 20-08-2004, 04:29 PM
  3. Please test your RAID 0 with this....
    By Wam7 in forum PC Hardware and Components
    Replies: 10
    Last Post: 08-04-2004, 09:54 AM
  4. Glitch in Prime 95
    By Spritzup in forum PC Hardware and Components
    Replies: 6
    Last Post: 16-02-2004, 02:43 AM
  5. Prime 95 errors
    By iMc in forum PC Hardware and Components
    Replies: 4
    Last Post: 16-02-2004, 02:41 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •