+ Reply to Thread
Page 6 of 9 FirstFirst ... 4 5 6 7 8 ... LastLast
Results 51 to 60 of 88

Thread: Overclocking Using Offset Mode for CPU Core Voltage

  1. #51
    ROG Guru: White Belt JJFIVEOH +10
    Join Date
    Mar 2011
    Location
    Boca Raton, FL
    Posts
    92

    Correct. The trick is figuring out the best offset and range from min to max vcore. Based on this thread it's pretty much a case by case basis depending on OC settings, hardware and usage.

  2. #52
    ROG Member AzO +10
    Join Date
    Nov 2011
    Location
    SF
    Posts
    7

    Ok, I enabled down clocking by enabling C1E (default was 'Auto' and would not down clock in that setting) My OC (4.6GHz) seems stable so far after running prime95, 20 loops of IntelBurn Test, 20 loops of LinX.

    However I feel that my voltage is a bit high on load (~1.375). Is that a normal load voltage for a 4.6GHz OC?

  3. #53
    Not to be taken seriously Raja@ASUS +20
    Join Date
    Apr 2011
    Location
    ASUS North America
    Posts
    2,845

    Use a negative offset and reduce the load vcore if your CPU can handle a lower voltage. 1.375V is about average for most CPUs at that frequency.

  4. #54
    ROG Enthusiast ugotd8 +10
    Join Date
    Nov 2011
    Location
    USA
    Posts
    35

    VID occasionally changes after reset/power cycle

    Hi Raja / all, quick question...

    Why does my VID occasionally read lower than other times after a boot ?

    For example, I've got my offset mode all setup, at +.060 for 4.9Ghz with HT on my 2600K.

    Most of the time, after a boot, realtemp shows my VID as 1.3661 and cpuZ shows my vcore at 1.424. That will pass prime95 no problem (I've ran it up to 9 hours once).

    Occasionally, after a boot (power-on) my VID will read 1.356 and vcore will be obviously too low at 1.408-1.416. This will not work, and I get BSODs when this happens.

    My understanding was that Intel had pre-programmed these VIDs into each 2600K and at they did not change. If it is possible for the VID to change, doesn't that make overclocking with the offset mode method unusable ?

    I dont know if this is a BIOS issue (I'm on 2050) or an Intel issue. Most likely it's my not understanding how the Dynamic VID feature of the SB works.

    If you could shed some light on this I'd appreciate it.

    A couple of other particulars that may or may not matter...

    Internal CPU PLL = AUTO
    CPU PLL = 1.693
    VCCSA = 0.90
    VCCIO = 1.05

    C3 & C6 = disabled

    Speedstep is working well, my VCC is 1.072 @ 1.60Ghz as I write this. :-)

    Thanks in advance.
    I7-3930K @ 4.9Ghz (1.384) / ASUS R4E / Samsung 30nm @ 1866 / Crucial M4 128GB / Caviar Black 1TB / HD6850 / Corsair AX850 / MCP35X2 -> RX360 - > SR1.360 -> Summit EF / DD Double-Wide Horizontal

  5. #55
    Not to be taken seriously Raja@ASUS +20
    Join Date
    Apr 2011
    Location
    ASUS North America
    Posts
    2,845

    In overclocked configurations the VID may drift sometimes on power cycle, the VID requests are based on power consumption, not much can be done. I would however set CPU PLL back to 1.80V and not 1.69V. Unlike older architectures CPU PLL needs to be kept at least around stock otherwise you may need more Vcore to stabilize an overclock.

    -Raja
    ASUS Technical Marketing

    If you are located outside North America and require technical assistance or support, please contact ASUS Support for your region.

    For ASUS North America Service: http://www.service.asus.com/#!contact-us/cw8z

  6. #56
    ROG Enthusiast ugotd8 +10
    Join Date
    Nov 2011
    Location
    USA
    Posts
    35

    Quote Originally Posted by Raja@ASUS View Post
    In overclocked configurations the VID may drift sometimes on power cycle, the VID requests are based on power consumption, not much can be done. I would however set CPU PLL back to 1.80V and not 1.69V. Unlike older architectures CPU PLL needs to be kept at least around stock otherwise you may need more Vcore to stabilize an overclock.

    -Raja
    Excellent info, thank you Sir.

    I will try resetting CMOS and re-building my OC keeping CPU PLL at 1.8.

    Just for the sake of discussion, there are folks across the different forums claiming that lowering CPU PLL can help stabilize an OC. One such post goes so far as to say that adding vcore had no effect on solving BSOD 124's. Only lowering CPU PLL made this person's issue with 124s go away. Apparently the consensus there was that in order to pass multiple hours of P95 with a 2600K, one needed somewhere between 1.6 and 1.7 on CPU PLL.
    I7-3930K @ 4.9Ghz (1.384) / ASUS R4E / Samsung 30nm @ 1866 / Crucial M4 128GB / Caviar Black 1TB / HD6850 / Corsair AX850 / MCP35X2 -> RX360 - > SR1.360 -> Summit EF / DD Double-Wide Horizontal

  7. #57
    Not to be taken seriously Raja@ASUS +20
    Join Date
    Apr 2011
    Location
    ASUS North America
    Posts
    2,845

    Quote Originally Posted by ugotd8 View Post
    Excellent info, thank you Sir.

    I will try resetting CMOS and re-building my OC keeping CPU PLL at 1.8.

    Just for the sake of discussion, there are folks across the different forums claiming that lowering CPU PLL can help stabilize an OC. One such post goes so far as to say that adding vcore had no effect on solving BSOD 124's. Only lowering CPU PLL made this person's issue with 124s go away. Apparently the consensus there was that in order to pass multiple hours of P95 with a 2600K, one needed somewhere between 1.6 and 1.7 on CPU PLL.
    My experience is different and I take most of my reasoning for the fact that we have to enable PLL OV to maximise OC headroom. It appears the 1.8VPLL rail is internally regulated, so the output is likely proportional to the input voltage or you are playing with the internal regulator's drop out margin. The tip is to play with VCCIO before fiddling with PLL as we know that VPLL over voltage is necessary for maximum processor overclocks.

    If you want to lower Vcore, try tuning BCLK skew in UEFI (can save a couple of steps). 124 errors are generally VCCIO related in my experience, clock interupt issues are usually Vcore.

    -Raja
    ASUS Technical Marketing

    If you are located outside North America and require technical assistance or support, please contact ASUS Support for your region.

    For ASUS North America Service: http://www.service.asus.com/#!contact-us/cw8z

  8. #58
    ROG Enthusiast broncogr +10
    Join Date
    Nov 2011
    Location
    Greece
    Posts
    42

    I think I have found a bug. I was using offset with a -0.005 to test my 4.8 overclocked 2700k. I foung some small instability after a couple of hours of prime95 I got a 124 BSOD. That overclock was getting me an idle vcore of 0.992 and load 1.344. So I left the offset to auto believing it will go to a bit higher load and idle vcore.
    It went to 1.50 vcore as soon as windows was loading !!!! and 1.346 in bios!!!! and dont know what on idle.
    SoI did a +0.005 and got fully stable at 1.352 -1.360 load 1.146 in bios and 1.008-1.010 idle.
    Still I would like to use +or- 0.000 which would be like leaving it on Auto.
    But on AUTO offset both + and - vcore goes just crazy!!!
    Probably a bug with 0902
    Any ideas?

    PS: All this on an Maximus Extreme IV Z motherboard

  9. #59
    ROG Enthusiast Fr0stX +10
    Join Date
    Apr 2011
    Location
    Montreal, Canada
    Posts
    44

    Quote Originally Posted by broncogr View Post
    I think I have found a bug. I was using offset with a -0.005 to test my 4.8 overclocked 2700k. I foung some small instability after a couple of hours of prime95 I got a 124 BSOD. That overclock was getting me an idle vcore of 0.992 and load 1.344. So I left the offset to auto believing it will go to a bit higher load and idle vcore.
    It went to 1.50 vcore as soon as windows was loading !!!! and 1.346 in bios!!!! and dont know what on idle.
    SoI did a +0.005 and got fully stable at 1.352 -1.360 load 1.146 in bios and 1.008-1.010 idle.
    Still I would like to use +or- 0.000 which would be like leaving it on Auto.
    But on AUTO offset both + and - vcore goes just crazy!!!
    Probably a bug with 0902
    Any ideas?

    PS: All this on an Maximus Extreme IV Z motherboard

    From asus, this is not a bug. Read the whole thread and you will see that I reported this behavior and was answered that this is normal. I just gave up on using this offset mode for overclock over 4.5Ghz. I guess if you have a golden CPU it works better, but I don't have a golden one..
    I7-2600K @3.40Ghz 1.088v
    Asus Maximus IV Extreme
    16Gb Gskill RipjawsX 9-9-9-24 @1600Mhz 1.35v
    EVGA GTX580 @775Mhz 1.000v
    Kingston SSDNow V Series Gen II 128GB
    Seasonic X-850 gold
    Corsair H50 w/2X Scythe Kaze Jyuni 120mm PWM fan
    AC Accelero Extreme Plus + VR004 GPU cooler
    Fractal Design Define XL Case w/4X Scythe Kaze Jyuni 120mm fan

  10. #60
    Banned jhuk +10
    Join Date
    Sep 2011
    Location
    Scotland
    Posts
    187

    I found it useless as it adds voltage to the voltage you want for giving OC!

    Example you want 1.3v for XXXX MHZ and also want Idle voltage to work for 1600mhz (2600K) so you must change the Power VRM to any setting apart from default Extreme (kinda strange imo)!

    You then use an offset (in my case minus) to get lower Idle voltage for 1600mhz and it works as it ends up hovering around 1.0v but it raises the normal VCore (dangerously high depending on your initial voltage) for some reason and taking of Auto and trying it that way meant I had to add some Turbo voltage but it also raised Idle voltage not just Turbo voltage!

    Bit Irish to take something off with one setting and add more back on with another setting!

    I am going to try Positive offset another day!

+ Reply to Thread

Similar Threads

  1. need help stabalizing voltage to cpu
    By FalloutGod13 in forum Rampage Series Motherboards
    Replies: 6
    Last Post: 02-13-2011, 12:39 AM
  2. ASUS Rampage II Extreme Extreme Core i7 LN2 OverClocking
    By V2-V3 in forum Rampage Series Motherboards
    Replies: 4
    Last Post: 01-26-2011, 07:39 PM
  3. GTX 580 Overclocks and voltage spreadsheet
    By cblaxx19 in forum General Hardware, Overclocking & Tweaking
    Replies: 0
    Last Post: 01-03-2011, 04:07 AM
  4. Replies: 2
    Last Post: 12-18-2010, 07:19 AM
  5. Intel to Release Core i7 990X Extreme Edition Six-Core Processor in Q4 2010
    By ottoyu34 in forum General Hardware, Overclocking & Tweaking
    Replies: 5
    Last Post: 06-13-2010, 02:29 PM

Posting Permissions

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