Results 1 to 25 of 1734

Thread: [First Look] DFI 680i LT NF-680i LT SLI (56K Warning~~~)

Threaded View

  1. #10
    Xtreme Addict
    Join Date
    Jul 2006
    Location
    Vancouver, BC
    Posts
    2,061
    Quote Originally Posted by SchiuZ View Post
    Thank you again. This CPU was capable of 3,6Ghz 1,58v (real voltage, not selected in BIOS but readed with a tester) 400x9Mhz. This frequency was not Prime 100% stable, but at 3,6Ghz it could run 3DM06 and SuperPi with no problems (also 4x32M SuperPi).

    The CPU was perfecrly stable at 3,4Ghz (425x8) vCore 1,45v. Liquid cooled, 8HRs and more QuadPrime and double Orthos blend stable (and not only, I also tried OCCT and other stability test for hours and always setting high priority).

    So the CPU is not "exceptional" but it is a good CPU, and I would like to make it run @ 400x8Mhz or something like that. The main problem that I have is not possibile to have a good BIOS, because the 2CTST BIOS is not good for 4Core CPUs, I also spoke with some friends that have direct contacts with DFI, and in theri opinion the correct BIOS is the 4CTST, not the 2CTST (and it is what I also confirm trying them both), but the Q6600 has locked molti with the BIOS, and the reason is that the 4CTST521 BIOS doesn't include the support for changing the multiplier in the non-extreme CPUs. The Extreme CPUs have a differet hardware function to change the multiplier, not depending on the NB freqeuncy, while the non-extreme can lower their multiplier changing (and incrasing) the NB frequency. The forumla is NB Core Clock = (Default Multiplier/Current Multiplier) x FSB and also Praz (DFI Club Moderator) told me this think...
    Ok, several things to consider here...

    First, we know your CPU is stable at 3.4GHz and a reasonable 1.45VCore.

    Keep in mind that the DFI undervolts the Vcore by about 0.07V and then droops an addition 0.05V under load. So to get an equivalent of 1.48V under load with the DFI you are going to want to max the CPU Vid at 1.6V... even to hit 3.4GHz.

    3.4GHz is 377x9 or 425x8... Forget about the 425x8... hitting 400+FSB with a quad is extremely difficult on any board... you were lucky to achieve that on the EVGA... not many can do that... it's like winning the lotto. You should just go back to the eVGA with results like that.

    With the DFI it may not be possible or would take an endless amount of tweaking GTL. Hence in my opinion you should aim for 377x9 and by the sounds of it you could use either BIOS to do this... certainly the locked multi version. The reason you may not be able to hit 400x8 is possibly due to the particular NB silicon you have on the DFI... perhaps it has a wall or hole there. Also keep in mind that the DFI undervolts the NB like crazy so you really need to crank up the voltage. You should be using something like 1.6x Volts on the NB I think. Also, my NB has holes... try going up in 5MHz increments from 360 and see if any of them work... I found my board wouldn't post at 360 but was perfectly stable at 380 (for example).

    Ultimately, I don't think your issue is BIOS related at all (although I share your frustration with the various versions)... perhaps you just have unrealistic expectations for the silicon you have in hand (trying for 400+ FSB)?
    Last edited by virtualrain; 08-30-2007 at 11:06 AM.

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
  •