Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Page 3 of 3 FirstFirst 123
Results 21 to 23 of 23

Thread: GA-X48-DQ6;Q6600 - OC/Memory Problem

  1. #21
    Join Date
    Nov 2009

    Default Re: GA-X48-DQ6;Q6600 - OC/Memory Problem

    Interesting...what's the highest max FSB you've achieved with your board? I have to say you're amongst a few I've found who have the same board as I do, it's not very common. My case is f'ed up and right now I have the mobo open on the tray, there is a 120 mm fan blowing onto the NB though, and at finger touch it's warm, but not hot. I'll flash the latest BIOS then try messing with the GTLREFs, which is which though? I assume one is the CPU, the other is the MCH?

    Also, what do you say about this (quote from AnandTech):
    "FSB Overvoltage Control
    This voltage setting is more commonly known as VTT. Default is 1.10V, with a maximum of 1.41V available. This voltage is critical for quad-core overclocking. 400FSB will require 1.41V right off the bat if stability is desired. (Ed: When isn't that the goal?) Unfortunately, we have no direct control over GTL (Gunning Transceiver Logic) reference values, which are locked at 67% of VTT for CPU die 1 and 63% of VTT for CPU die 2. On a top-end board like this, we had expected to find GTL adjustments, which can be crucial for finding quad-core CPU stability at high FSB speeds."

    This is regarding the X48-DQ6T but I doubt the voltages are any different. What they're basically saying is the FSB voltage should be set at +0.35 or what the max is for any kind of serious quad-core overclocking.
    This may as well impact my ability to reach a high FSB. I don't wanna try it before I get more opinions though. They didn't seem to find the GTLREF controls, must've not known about the CTRL-F1 method then.
    Last edited by WrxSTI; 11-21-2009 at 01:20 AM.
    Intel(R) Core(TM)2 Quad CPU Q6600 G0 @2.8 GHz / Prolimatech Megahalems Cooler (2x Scythe Slipstream 1200)
    MUSHKIN XP Series 991599 4 GB (2GB + 2GB) PC-8500
    SAMSUNG HD103UJ Spinpoint F1 1000 GB / Western Digital WD1600JS-55MHB0 Caviar 160GB
    Sapphire ATi Radeon HD 4870 X2 2 GB GDDR5 (750/900MHz)@PCI-E 2.0 x16, COOLER: Arctic Cooling Accelero XTREME 4870X2
    Windows XP Professional
    GIGABYTE GA-X48-DQ6, iX48(Bearlake) + iCH9R, F8D bios
    Corsair TX750W PSU

  2. #22
    Join Date
    Oct 2008
    Nimbin, Australia

    Default Re: GA-X48-DQ6;Q6600 - OC/Memory Problem

    My max so far is 466fsb, but not 24/7 stable. I have not gone for high fsb/low multiplier, but use full multiplier.

    Think default vtt is 1.1 for 45nm and 1.2 for 65nm. I have not used more than +.2 on my core duo. Have not tried a quad.

    No control over gtl's? Push Control and F1 simultaneously, as u say. Don't know what Anandtech were thinking there. Only the two cpu gtlref, no gtlref for MCH. (No PLL either.)

    Data busses are much more sensitive to adjustments as these lines are consistently more heavily loaded than the address busses. Enterprising enthusiast who wish to experiment would do much better to modify the data bus GTL reference voltage level(s) before any others. Adjusting A/GTL+ Levels for Increased FSB Signaling Margins and Overclocking - The Tech Repository Forums

    GTL Reference Data Strobe Input Buffer (GTLREF0)
    GTL Reference Address Strobe Input Buffer (GTLREF1)

    GTLREF values do not represent individual cores, but pairs of cores. One for data strobe for two cores and one for address strobe for two cores. If your mobo has four GTLREFs you can set two cores differently to the other two on a quad. That's what it looks like to me. (There's a pair of cores on each die, so 2 refs for each die.)

    Vtt is "FSB Overvoltage Control"

    67% of nominal Vtt of 1.1v is a nominal .737v. All these are nominal voltages, not measured.

    My Vtt was set at +.1 (1.2v) from the start giving a subsequent nominal 67% GTLREF of .804v This became obviously unstable by 445 fsb-bclk.
    At 445 fsb-bclk -3% on the data GTLREF0 gave .768v (64%)
    At 454 Vtt was changed from 1.2v to 1.25v, but left at -3%, 1.2v =.768v, 1.25v = .8v (Both 64%)
    At 466 with Vtt of 1.3 -6% on the data GTLREF0 gave .793v (61%)

    At 466 with Vtt of 1.25v -3% on the data GTLREF0 gave .8v (64%), -6% gives .7625v (61%), -9% would be .725v (58%)

    The above being successful boot setting voltages GTLs appear totally Vtt dependent, and want to stay between .737v and about .8v for my particular hardware.
    My GTLREF0 (Data) seems to be the important one.

    If GTLs are bringing "fsb holes" into sync, as I've seen suggested, then I've got an fsb trench starting around 440, and getting deeper as it goes.

    I see many use -3%, -3%, but also note the native bios settings of AUTO/NORMAL which imply that only REF0 would change..

    I see vtt used two entirely different ways.
    Keep it between .05 and .15 below vcore, or make it a greater number than a low vcore. Seen both used, and no idea what is absolutely right.

    Good info:
    Gigabyte GA X38 DQ6 Blog - Clunk's Forums.

    Gigabyte X48T DQ6 Living Review - Clunk's Forums.

    X48T - Anandtech's 1.45v reading of MCH voltage is disputed at Clunk.

    Nimrod has a quad on x48-ds5, and e8500 on EP45.

    Your comment on 100 pcie setting was interesting. Testing that out.

    My oc thread;
    Last edited by Salty; 11-22-2009 at 05:46 PM.

    i7 950 - DO stepping - Batch #3014A844
    Cooling = EK HF Full Nickel
    - XSPC RX360 - Laing DDC 18W - XSPC Bay Reservoir
    Motherboard = GA-X58A-UD3R Revision 2
    , BIOS Version = Award FC
    Ram = 3x2gig Corsair TR3X6G1600C8D G
    Video = 2 X Palit GTX460 Sonic Platinum 1GB OC sli
    Monitor = 22" LG Flatron W2242T 1680 x 1050 Case = Coolermaster CM690
    PSU = Corsair HX850
    OS = Windows 7 Professional x64

  3. #23
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    PCB Island

    Default Re: GA-X48-DQ6;Q6600 - OC/Memory Problem

    Yes to see GTLs you may need to use the hotkey. On the MAIN page of the BIOS press control + F1 and the screen will flash, then go into MIT and you will see the hidden settings.

    You may need +3 or more FSB for 450+ on a quad.

    On your comments about PCIE, have you tried manually setting 103+ to see if that had any effect since auto seemed to help? Just wondering

    I also replied to you at Mushkin forums, I really don't think memory is the issue here. I assume you think it might be right? Is that because you were able to go higher with looser timings or tRD? If so, this likely is indicative of your NB Failing to run stable at the voltage or timing (Strap) settings you were trying before, thus looser ones worked easier for you. What Memory multi are you using by the way?

Page 3 of 3 FirstFirst 123

Thread Information

Users Browsing this Thread

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

Posting Permissions

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