Results 1 to 14 of 14

Thread: More vcore, less heat. Why?

  1. #1
    Registered User
    Join Date
    Jul 2007
    Posts
    12

    More vcore, less heat. Why?

    I don't know if this is the right place to post this but here's my story. I recently purchased PDC E6500 to resurrect my old system. Spec is as follows,

    PentiumDualCore E6500
    Gigabyte GA-P35-DS3L
    PointOfView 8800GT 512MB
    4GB Kingston DDR2 800MHz/PC2-6400
    Realtek ALC888 Onboard
    Jeantech 450W
    Thermaltake Shark Silver

    So yesterday I put these stuffs together and installed windows7, everything went ok. I decided to start overclocking E6500. Upto 3.66G it ran fine, IBT max stress 20 pass no problem. But once I brought up fsb to 350 and started stressing, I noticed something very weird. Although I increased vcore for 3.85G(1.28v for 3.66, 1.325v for 3.85), for some reason max load temp went down by 5C per core or so. And test seemed to take much longer to complete(1 test took like 400s to complete as opposed to only 210s at 3.66G). I got no errors and test completed ok but apparently something isn't right, is it? Could someone here possively guess why this happened and suggest me how to fix it please? Here are some stuffs that came to my mind but I don't know if it is actually the source of the problem.

    1. PSU is too weak for the system and not supplying enough juice.
    2. CPU doesn't like 2.00 strap.
    3. CPU hit FSB wall.

    Any input would be greatly appreciated, thanks.(sorry for my weird English)

  2. #2
    Registered User
    Join Date
    May 2005
    Location
    shanghai, pr china
    Posts
    61
    IBT=Intel Linpack has a random performance(GFlops) with up to ~10% variance. Less performance(GFlops)=>much less heat.
    Occasionally it is completely screwed, resulting in e.g. 100% slower and >15C less core temp.

    Before you run IBT for many pass to test stability, you had better run it 1 pass for several times, to get an idea of a "normal" performance(GFlops) number.
    i7 920/950/w3565 @ ek hf
    R3E @ bios 0602/0704
    bbse gs 2133ps/hyper pdp2kc8 stt2kc8 stt2kc7 stt2.2kc8
    zotec gtx480 @ stock @ pcie x8
    x-fi elite pro @ pci
    areca arc-1300-4e eSATAx4 controller @ pcie x1(x8)
    hd*N @ ahci
    corsair hx850w
    cosmos rc-1000 case
    windows xp pro 32-bit sp3/windows 7 ult 64-bit

  3. #3
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    Ok thanks for your input shadewither, that seems to be the cause of the problem then. But how should I optimise this "random performance" thing? Could you share some tips if possible? Increasing vcore further doesn't seem to help the situation. I've never ran into this kind of problem before and I'm not quite sure what you are saying to be honest.

  4. #4
    Registered User
    Join Date
    Nov 2008
    Posts
    3
    Run some custom prime95 blend for a minute or two, using alot of ram, like 3300mb. It will clear the cache and will allow IBT to load the memory properly Moar gflops.

  5. #5
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    OK thanks I'll try that. But doesn't this cache thing cleared when you restart your system? IBT runs slow at 3.85G even after restarting. Is this normal?

  6. #6
    Registered User
    Join Date
    Nov 2008
    Posts
    3
    It can be, but when using alot of ram (ur using max ram as per OP), u will need to clear the cache even further. I've found that 3072mb in IBT is the perfect point, anymore and i get less gflops no matter how much ram i try to clear. It will probably be different for u, so just experiment
    Last edited by Oztepher; 07-16-2010 at 12:48 AM.

  7. #7
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    Ok thanks for your help, I'll experiment with ram settings and see if that makes any difference then BTW I thought I was the slowest poster here but looks like you beat me to it, well done. lol

  8. #8
    Xtreme Member
    Join Date
    Nov 2008
    Location
    London
    Posts
    300
    Yeah, this is my normal account i have 2 for some reason, i had difficulties with creating an account at first and created 2 by accident... didn't realise i logged into the other one until it saw it had 2 posts and no problem mate, keep us updated if it solves ur problem... or not.
    -
    Core i7 860 @ 3.80GHz, 1.28v | GA-P55A-UD4 | G.Skill Ripjaw 4GB DDR3 @ 1900MHz 7-9-8-24 1N, 1.57v | HIS HD 6950 2GB, 1536sp @ 900/1400, 1.10v | Samsung F3 500GB | Thermaltake 750W | Windows 7 64bit | Air

    Crunching away...

  9. #9
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    Sorry for double post, but I just googled with keywords "IBT gflops slower" and found this thread. Looks like one guy here having almost identical problem to mine with his E6300. Maybe this is E6xxx specific problem? Very interesting.
    http://forums.vr-zone.com/hardware-a...finally-8.html

  10. #10
    Xtreme Cruncher
    Join Date
    Jun 2006
    Posts
    6,215
    Or maybe the chip is just throttling itself to prevent perma damage?

  11. #11
    Xtreme Member
    Join Date
    Nov 2008
    Location
    London
    Posts
    300
    It could be something like that, what temperatures were u actually getting at load? I've re-read the OP but i can only find that u said u get 5c less than 3.66
    Last edited by Oztopher; 07-16-2010 at 01:33 AM.
    -
    Core i7 860 @ 3.80GHz, 1.28v | GA-P55A-UD4 | G.Skill Ripjaw 4GB DDR3 @ 1900MHz 7-9-8-24 1N, 1.57v | HIS HD 6950 2GB, 1536sp @ 900/1400, 1.10v | Samsung F3 500GB | Thermaltake 750W | Windows 7 64bit | Air

    Crunching away...

  12. #12
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    Quote Originally Posted by Oztopher View Post
    Yeah, this is my normal account i have 2 for some reason, i had difficulties with creating an account at first and created 2 by accident... didn't realise i logged into the other one until it saw it had 2 posts and no problem mate, keep us updated if it solves ur problem... or not.
    Oh I see, and yes I will Meanwhile I will be recieving a replacement PSU for my first system sometime next week(coincidence) so I'll try to install that first on this system and see if that makes any difference. I still cannot rule out the possiblity of weak power supply not delivering enough power to be honest, especeially after reading above thread. It's 850W one so it will be more than sufficient for this setup so it's a good opportunity to try I guess. And it would surely be more reliable than Jeantec one which I bought from PC World about 3 years ago:P

    Quote Originally Posted by informal View Post
    Or maybe the chip is just throttling itself to prevent perma damage?
    Thanks, but the temp wasn't that alamingly high and voltage I used was within intel spec so I don't think that would be the case.

    Quote Originally Posted by Oztopher View Post
    It could be something like that, what temperatures were u actually getting at load? I've re-read the OP but i can only find that u said u get 5c less than 3.66
    Temp readings while running IBT was something like 58C both core at 3.66G, and 52-53C at 3.85G I think. But the strange thing is when I first ran at 3.85G for 5 pass I saw temp climed up to 65C, but probably this is due to GFlops thing you guys have mentioned above.

  13. #13
    Xtreme Member
    Join Date
    Nov 2008
    Location
    London
    Posts
    300
    Take alook at this link
    http://www.overclock.net/intel-cpus/...test-linx.html

    Correct me if i'm wrong (which is not rare), but if it was a PSU problem wouldn't it just crash out or something? rather than throttle down.
    -
    Core i7 860 @ 3.80GHz, 1.28v | GA-P55A-UD4 | G.Skill Ripjaw 4GB DDR3 @ 1900MHz 7-9-8-24 1N, 1.57v | HIS HD 6950 2GB, 1536sp @ 900/1400, 1.10v | Samsung F3 500GB | Thermaltake 750W | Windows 7 64bit | Air

    Crunching away...

  14. #14
    Registered User
    Join Date
    Jul 2007
    Posts
    12
    Thanks for the link I'll sure have a look at it, and yeah you are right, I was actually thinking about the same thing, if PSU is not supplying enough power PC would have shut down while I was running IBT or playing GTA4 and that never happened. Hmm...I don't know, but I still think it's worth trying another PSU I mean this Jeantech one is so old, it probably came out just after graphic cards with PCI-E power connector first hit the market and probably there is a possibility of some specification not meeting todays standard. I'm just not comfortable with it and on top of that I bought it off PC World if you know what I mean lol. Anyway I'll try to eliminate the problem and I'll report back once things have settled hopefully. Anyway thanks all again for your valuable inputs

Bookmarks

Bookmarks

Posting Permissions

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