Page 4 of 8 FirstFirst 12345678 LastLast
Results 46 to 60 of 111

Thread: opteron 165

  1. #46
    Joined
    Feb 2003
    Posts
    8,739

    Re: opteron 165

    core temp may give a tcase temp.
    yeah that's why i said may. i have a 170 & 175 that don't give a tcase # either.

    it was worth a try.

    ud

  2. #47
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    hey i was wondering, in order to test for stability, do i have to load the cpu to 100%... i mean, i know yes if i want to know that the system wont crash because of heat and heavy load. but what if i just want to know if there are no errors? wont prime 95 tell me that even though the load is not 100%
    - i want to know because right now i cant fully test 100% load on the 2500 OC
    but if i just want to know that there are no errors or other problems except heat, then cant i run only one instance of prime 95? if the OC is not stable, wont i get an error even though im not loading the cores to 100% ?

    it was worth a try.

    thanks anyways.. you guys have been very helpful... i couldnt even reach 2.0GHz on this without all the precious advice...
    Last edited by Bauglir; 09-25-2007 at 02:07 AM.
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  3. #48
    Joined
    Jan 2004
    Posts
    3,463

    Re: opteron 165

    A single Prime95 thread will only test a single core, and will not give the memory controler as much of a workout as two threads. It can be done, and would tell you if there is something fundamentally wrong with the core it ran on, but is not really that useful.

    An interesting thing to note about the AMD stock cooler is how retarded the temp control is. I've used the heatsink before, and my CPU would be cooler in 90F ambient temps, than in 70F ambient temps, because at 90F the fan is running at 5800 rpm or thereabouts and only 2800 at 70F. Cold air getting to the CPU = slower fan speed = hotter CPU.

    If you can put up with the noise you can simply bridge the pins of the thermal sensor (ithe little green bulb sticking fromt he base of the fans motor; use a small blob of solder on the bare wire extending from it into the motor) and it will run full blast all the time and cool much better.

  4. #49
    Joined
    Feb 2003
    Posts
    8,739

    Re: opteron 165

    Quote Originally Posted by Bauglir View Post
    hey i was wondering, in order to test for stability, do i have to load the cpu to 100%... i mean, i know yes if i want to know that the system wont crash because of heat and heavy load. but what if i just want to know if there are no errors? wont prime 95 tell me that even though the load is not 100%
    - i want to know because right now i cant fully test 100% load on the 2500 OC
    but if i just want to know that there are no errors or other problems except heat, then cant i run only one instance of prime 95? if the OC is not stable, wont i get an error even though im not loading the cores to 100% ?
    a unique question. i don't think anyone who was on a roll with their oc but waiting on a fan that's in the mail, has asked that before.

    so when is that fan scheduled to arrive?


    ud

  5. #50
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    hehe... i got the itch now.... it's on its way
    .....
    "scratches head"
    i wish i had it NOW... LOL
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  6. #51
    Joined
    Feb 2003
    Posts
    8,739

    Re: opteron 165

    cpu temp readings are quite often suspect.

    with your "current settings", the ones that appear to be causing the system crash, would be a good place to start with the new fan. since it will help lower the cpu temp and if the system no longer crashes, then you have a pretty good idea what the max temp of the cpu is. a64s will crap out before damaging themselves because of heat.


    ud

  7. #52
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    yeah, thats cool....
    i just need to be patient.... untill the fan arrives....
    hard times..lol
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  8. #53
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    *BUMP *BUMP BUMP
    my fan is here
    il install it tomorrow and start testing
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  9. #54
    Joined
    Feb 2003
    Posts
    8,739

    Re: opteron 165

    i just need to be patient.... untill the fan arrives....
    i didn't want to say this till it was at your door but i had a package that was out for delivery for 4 days.

    have fun tomorrow.

    Ud

  10. #55
    Joined
    Feb 2006
    Location
    Fort Myers, FL
    Age
    41
    Posts
    471

    Re: opteron 165

    Quote Originally Posted by oralpain View Post
    A single Prime95 thread will only test a single core, and will not give the memory controler as much of a workout as two threads. It can be done, and would tell you if there is something fundamentally wrong with the core it ran on, but is not really that useful.

    An interesting thing to note about the AMD stock cooler is how retarded the temp control is. I've used the heatsink before, and my CPU would be cooler in 90F ambient temps, than in 70F ambient temps, because at 90F the fan is running at 5800 rpm or thereabouts and only 2800 at 70F. Cold air getting to the CPU = slower fan speed = hotter CPU.

    If you can put up with the noise you can simply bridge the pins of the thermal sensor (ithe little green bulb sticking fromt he base of the fans motor; use a small blob of solder on the bare wire extending from it into the motor) and it will run full blast all the time and cool much better.
    Then why not stick a Tornado on it? Who likes a noisy computer?

  11. #56
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    ok, my fan is installed and i tested a full 5 hrs of prime 95 with no crash....
    i noticed something wierd though...
    i was monitoring the temps with both core temp and speed fan, and idle temps have dropped to 40C, 15C lower than with the panflo fan. i thought that that was cool, so i went ahead with a dual prime 95 test.
    temperatures quickly rised to 60C under load. the thing is that in core temp i can see a 10C difference betwen the cores under load. in idle there is a 2C difference. so while one core was on low sixties the other wason the low 70s.
    i would be more or less okay with that, but i also noticed that core 2 would oscilate from a regular 62C to a scorching 74C and back down again. completing the cycle in about 15 minutes.
    has anyone seen such erratic behaviour before...
    i have one possible cause.: i tried to install the temp. sensor that came with the fan in between the xp90 and the cpu,... didnt work, i damaged the sensor because of the pressure, and i might have disturbed the thermal compund nderneath... would such a situation give rise to the difference in core temps?
    the temperatures correspond in core temp and speed fan so its not a software issue....


    by the way: thanks uddarts i love the fan!!!
    Last edited by Bauglir; 09-29-2007 at 02:06 AM. Reason: spelled name wrong
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  12. #57
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    just so that we get ourselves back in contest, this is the current config
    from cpuz:
    CPU: 2592 Mhz
    Mult : 9X
    Bus: 288.10
    HTT: 834.s MHz
    Vcore 1.45V

    RAM:
    144MHZ
    2.5-2-2-5-7
    avg load temps 65-68, with cycling of core 1 as mentioned above
    idle temp: 40C

    Preliminary Stability Testing:
    prime95 both cores, 5hrs 5 min, no errors
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  13. #58
    Joined
    Nov 2005
    Location
    northern USA, aka CANADA
    Age
    38
    Posts
    99

    Re: opteron 165

    i was noticing too that when my cpu was at 2.5, with the older fan, the idle temps wer higher but the load temps were lower...
    it must have something to do with the termal paste....

    does prime 95 have a log file, so that when the computer crashes we can see where it was, how ong it had run etc?
    The green Earth say you?
    That is a mighty matter of legend, though you tread it under the light of day!




    Ozma. He is like a big rainbow ball of hate...

  14. #59
    Joined
    Feb 2003
    Posts
    8,739

    Re: opteron 165

    i hope you have given up on the sensor. they're not made for the a64s. they worked fine with the xp cpu.


    it's not unusual for the core temps to be different. some more, some less.


    you shouldn't have a problem with prime95 crashes anymore with the fast fan.

    it probably would be good to redo the arctic silver.

    no need to go above 1.45 on the vcore.

    both the higher vcore and htt will increase the temps.

    was prime halting with the higher (288)htt? did you try 1.40v or 1.425v before going to the 1.45?

    ud
    Last edited by uddarts; 09-29-2007 at 06:52 PM.

  15. #60
    Joined
    Jan 2002
    Location
    Joliet, Illinois
    Age
    34
    Posts
    15,846

    Re: opteron 165

    i hope you don't mean the temp sensor is between the heatsink and the actual cpu. that's not good. the only thing you want between the two is your thermal paste, the probe will only make things worse.

    i wouldn't bother with the probe. if the fan you got has a 3 pin header and your mobo has support for sensing fan speeds you can plug the fan into the cpu fan header on the board and use speedfan to adjust the speed of the fan based on your core temp. in that case it doesn't matter if the temp is 100% accurate, you're just using that temp to control your fan speed. you'll just have to custom set it to your particular system. so basically if the max temp you get is 60c and idle is around 40c, set the desired temp to say 50c, this way speedfan will try to keep the temp at or below that temp and will go full force when it hits that temp. also set the minimum fan speed to a speed just below where you can hear it, this way when the machine is idle or you're just srufing around you have peace and quiet.

    but i can't tell you more how important it is not to put a temp probe between your heatink and cpu.
    Corsair 800D / Corsair TX750 PSU
    ASUS Sabertooth P67 / Win7 Pro
    Intel i7 2600k @ 4.4ghz (44 x 100) + Corsair H100
    Corsair Vengeance DDR3 1600 x 16GB
    2 x XFX R9 280x + 3 x Asus VS24AH-P 24" IPS 1920x1200
    Samsung 840 Evo 250GB OS + 2 x 1TB WD Caviar Black




    The member formerly known as SuBX3r0 HEAT

Posting Permissions

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