Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Page 20 of 25 FirstFirst ... 101819202122 ... LastLast
Results 191 to 200 of 245

Thread: X48 DQ6 E0 E8500 at 4085Mhz




  1. #191
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    You can't escape your destiny. Time for GTLs.

  2. #192
    Join Date
    Oct 2008
    Location
    Nimbin, Australia
    Posts
    339

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    After a measly 45Mhz over the last bus? That's cruelty to dumb overclockers. I'm just a literate noob really, not fully comprehending.

    There are GTLref tables for the DFI board, but not the Gigabyte, and I still don't understand parts of it. I would be messing with things I didn't fully understand. Will have to sleep on it.

    Its the options that drive u crazy.





    CPU
    =
    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. #193
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Quote Originally Posted by Salty View Post
    There are GTLref tables for the DFI board, but not the Gigabyte
    And those tables work for all processors? I'd be surprised if they did or why wouldn't they be implemented as auto BIOS settings?

    From what I understand they correct one of two problems, the ring back, or the ground bounce.
    I would try maybe lower vtt to keep the "A" of AGTL+ in effect and higher vtt to see if the ground bounce is the problem.

  4. #194
    Join Date
    Oct 2008
    Location
    Nimbin, Australia
    Posts
    339

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Prime is running small ffts.
    4220 started failing large ffts, just as 4370 did, then 4300, then 4250.

    Have changed GTLref1 from auto to normal, and running Prime. Considering it's 27C ambient using LinX probably not a good idea.

    Into the 14k without stopping. Will progressively test the limited range of settings. Instead of entering actual values the x48-dq6 has -3%, normal, +3%, +6% - so four settings.

    I am wondering if my ram has degraded. Will run memtest when time and ambient permit.

    Is 4.22 a "high overclock"? I was thinking 4.4 up would be high.

    EDIT: Either using normal for GTLref or temperature drop seems to have given stability... what cold gives, heat takes?
    12:34am 23C ambient YES (LinX passes)
    Last edited by Salty; 10-03-2009 at 10:43 PM.





    CPU
    =
    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


  5. #195
    Join Date
    Apr 2009
    Location
    J'Habite En Angleterre
    Posts
    2,431

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Quote Originally Posted by Chike View Post
    Wouldn't recommend even to try the rubber washers. The backbplate was not made for nothing, they could just have used muts instead if they thought it isn't needed, it is made to take the presure instead the board, with the rubber washer all the preasure is on the board.
    Missed this a while ago. Sorry Chike but, respectfully, that's rubbish. All you're doing when adding rubber washers is extending the thickness of the back plate in certain areas and moving the screw threads further away. Most of the strain will still be taken up by the back plate. Basic physics. As pressure is applied by the screws, opposite pressure is being applied by the back plate, regardless of any washers fitted.

    The statement "with rubber washers, all the pressure is on the board" is completely false.

    If anything, the first method with the large washer may potentially flex the board as therer is much more concentrated pressure in the centre. When the washer is too thick it may over whelm the capacity of the back plate to keep the board straight.

    The rubber washer method is the safest, and applies the most uniform pressure. There is ZERO flex applied to the board.
    Coolermaster CM 690 II advance Case
    Corsair HX750 (CWT, 91%(80+ Gold rated @230V) single 62A 12V rail
    P55A-UD4 v2.0 @ F14
    Core i5 760 @ 20 x 201, 4.02GHz
    TRUE Black with a single Noctua NF-P12 pumping out 55 CFM @ 19db .
    2 x 2GB Mushkin Ridgeback (996902), @ 7-10-8-27, 2010-DDR, 1.66v
    2 x Gigabyte GTX 460 1024MB in SLI (Pre OC'd to 715MHz core and 1800MHz VRAM) @ 850 Core / 4100 Mem.
    Intel X25-M Boot Drive (OS and Programs) 200MB/s Read & 90MB/s Write
    Corsair X32 200MB/s Read & 100MB/s Write
    WD Caviar Blue 640GB C (Steam, Games, Storage, Temp Files & Folders, etc)
    Samsung F3 500GB Backup/Images
    Noctua 1300RPM 19dB case fan (rear extraction)
    3 x 140 MM Coolermaster LED fans (one front intake, one top extraction, one side intake)
    Dell Ultra Sharp 2209WAf E-IPS @ 1680x1050

  6. #196
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Quote Originally Posted by Psycho101 View Post
    Missed this a while ago. Sorry Chike but, respectfully, that's rubbish. All you're doing when adding rubber washers is extending the thickness of the back plate in certain areas and moving the screw threads further away. Most of the strain will still be taken up by the back plate. Basic physics. As pressure is applied by the screws, opposite pressure is being applied by the back plate, regardless of any washers fitted.

    The statement "with rubber washers, all the pressure is on the board" is completely false.

    If anything, the first method with the large washer may potentially flex the board as therer is much more concentrated pressure in the centre. When the washer is too thick it may over whelm the capacity of the back plate to keep the board straight.
    What is going to prevent the board from flexing if there's a gap between the backplate and the board? How is the backplate going to take any pressure?
    With the top washer the board is not going to flex unless the backplate does, there may be pressure on the board, but it would not flex.

    Quote Originally Posted by Psycho101
    The rubber washer method is the safest, and applies the most uniform pressure. There is ZERO flex applied to the board.
    Since a completely false statements are already discussed.

  7. #197
    Join Date
    Apr 2009
    Location
    J'Habite En Angleterre
    Posts
    2,431

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    the backplate doesn't need to touch the board completely for it to take the strain and stop the board flexing.

    A board will flex if there is no back plate. With the back plate on and the rubber washers fitted, the force is the same. It's applied to the board at the points where the screws are, ie at the four corners. There is virtually no pressure at the centre. If there were then the TRUE wouldn't rotate the way it does. Bowing would be caused by pressure at the four mounting points curving the board around the TRUE bottom plate. With the back plate in place fitted with the washers, as without, the force is transfered to the plate. The force transfer causes the mounting of the TRUE to now try and bend the back plate. It can't do that because it's much stiffer than the board.

    Like I said, the board does not flex with the washers added. I suggest you try it and see for yourself rather than just guess. Your reasoning is flawed.

    Edit: The reason that the top washer mod causes more flex is because the back plate itself is potentially more flexible if pressure is applied dead centre at a single point than if it's spread out over all four corners.
    Coolermaster CM 690 II advance Case
    Corsair HX750 (CWT, 91%(80+ Gold rated @230V) single 62A 12V rail
    P55A-UD4 v2.0 @ F14
    Core i5 760 @ 20 x 201, 4.02GHz
    TRUE Black with a single Noctua NF-P12 pumping out 55 CFM @ 19db .
    2 x 2GB Mushkin Ridgeback (996902), @ 7-10-8-27, 2010-DDR, 1.66v
    2 x Gigabyte GTX 460 1024MB in SLI (Pre OC'd to 715MHz core and 1800MHz VRAM) @ 850 Core / 4100 Mem.
    Intel X25-M Boot Drive (OS and Programs) 200MB/s Read & 90MB/s Write
    Corsair X32 200MB/s Read & 100MB/s Write
    WD Caviar Blue 640GB C (Steam, Games, Storage, Temp Files & Folders, etc)
    Samsung F3 500GB Backup/Images
    Noctua 1300RPM 19dB case fan (rear extraction)
    3 x 140 MM Coolermaster LED fans (one front intake, one top extraction, one side intake)
    Dell Ultra Sharp 2209WAf E-IPS @ 1680x1050

  8. #198
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Quote Originally Posted by Psycho101 View Post
    There is virtually no pressure at the centre. If there were then the TRUE wouldn't rotate the way it does.
    The TRUE rotate because it is only held down by the presure, and you will need a lot more presure to prevent it from rotating with grease applied between the CPU and it's base.

    Quote Originally Posted by Psycho101
    Edit: The reason that the top washer mod causes more flex is because the back plate itself is potentially more flexible if pressure is applied dead centre at a single point than if it's spread out over all four corners.
    I think you have found the ultimate solution to prevent it from flexing with the top washer. Just add some very small washers between the backplate and the board, once there is no contact between the board and the backplate it would no longer flex.... I think.... going with your reasoning.

  9. #199
    Join Date
    Oct 2008
    Location
    Nimbin, Australia
    Posts
    339

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    GTLREF and FSB VTT Voltage Tweaking

    Warning: MSI board discussion.





    CPU
    =
    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


  10. #200
    Join Date
    Oct 2008
    Location
    Nimbin, Australia
    Posts
    339

    Default Re: X48 DQ6 E0 E8500 at 4085Mhz

    Don't know that I've achieved much here beyond beginning to play with GTRLREF.
    445fsb seems to like both Normal and -3% better than auto. Does not like +3%. Fails small fft.
    Dropping vtt same for "Normal" GTRLREF1 failed LinX.
    Did not pass LinX with lower mch or cpu voltage.

    Code:
    Robust Graphics Booster________________ [Auto]
    CPU Clock Ratio _______________________ [9] 
    Fine Clock Ratio ______________________ [0.5]
    CPU Host Clock Control_________________ [Enabled]
    CPU Host Frequency (MHz)_______________ [445] 
    PCI Express Frequency (Mhz)____________ [100]
    C.I.A. 2_______________________________ [Disabled]
    System Memory Multiplier (SPD)_________ [2.40B]
    DRAM Timing Selectable_______ SPD _____ [Manual]
    
    Performance Enhance =                   [Standard]
    
    CAS Latency Time_______________________ 5
    Dram RAS# to CAS# Delay________________ 5
    Dram RAS# Precharge Delay______________ 5
    Precharge Delay (tRAS)_________________ 15
    
    ACT to ACT Delay (tRRD)________________ 3
    Rank Write to READ Delay_______________ 3
    Write to Precharge Delay_______________ 6
    Refresh to ACT Delay___________________ 54 
    Read to Precharge Delay________________ 3
    Static tRead Value_____________________ 7
    Static tRead Phase Adjust______________ [Auto]
    
    CPU/PCIEX Clock Driving Control         [800mV]
    CPU Clock Skew Control                  [normal]
    GMCH Clock Skew Control                 [Normal]
    
    System Voltage Control_________________ [Manual]
    DDR2 OverVoltage Control_______________ [+0.300V] 
    PCI-E OverVoltage Control______________ [+0.1V] 
    FSB OverVoltage Control________________ [+0.1V] 
    (G)MCH OverVoltage Control ____________ [+0.225V]
    CPU GTRLREF1 Voltage Ratio ____________ [-3%] 
    CPU GTRLREF2 Voltage Ratio ____________ [Normal]
    CPU Voltage Control____________________ [1.35625]
    
    Limit CPUID Max. to 3.................: [Disabled]
    No-Execute Memory Protect.............: [Enabled]
    CPU Enhanced Halt (C1E)...............: [Disabled]
    C2/C2E State Support..................: [Disabled]
    x C4/C4E State Support................: [Disabled]
    CPU Thermal Monitor 2(TM2) ...........: [Enabled]
    CPU EIST Function.....................: [Disabled]
    Virtualization Technology.............: [Disabled] Enabled if you use Vmware/Virtual PC
    
    Integrated Peripherals
    Legacy USB Storage Detect______________ [Disabled] *Note* Must be enabled to flash from USB
    Couldn't resist seeing if it would revive the 4.31GHz profile: It did.



    CPU GTL Reference (0/2 Lane0/Lane2)
    GTL Reference Data Strobe Input Buffer Middle/End Landing Pins 0 & 2. (GTLREF0/GTLREF2)

    CPU GTL Reference (1/3 Lane1/Lane3)
    GTL Reference Address Strobe Input Buffer Middle/End Landing Pins 1 & 3 ( GTLREF1/GTLREF3 )
    On GA-X48T-DQ6 running 45nm Wolfdale I have found that the best settings are 64%(-3%) & 60%(-3%)
    Previous experience on the Abit with 65nm Quads I found 67% /66% worked best which here means Auto(67%) and Normal(63%) are the closest.
    Okay, just have to test that out. EDIT* Tried it. Out of memory error in LinX.

    GTL (Gunning Transceiver Logic) reference values are set at 67% of VTT for CPU die 1 and 63% of VTT for CPU die 2.
    Summing up, a lower fsb voltage (vtt) by using GTLRef1 of -3%, effective at 445 and 454fsb, so far.
    Last edited by Salty; 10-04-2009 at 05:47 PM.





    CPU
    =
    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


Page 20 of 25 FirstFirst ... 101819202122 ... LastLast

Thread Information

Users Browsing this Thread

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

Tags for this Thread

Posting Permissions

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