Results 1 to 25 of 5952

Thread: ***Asus Rampage III Extreme Owners Thread***

Hybrid View

  1. #1
    Xtreme Member
    Join Date
    Feb 2006
    Location
    Richmond, VA
    Posts
    471
    Quote Originally Posted by Viss View Post
    QPI / VTT LoadLine Calibration
    Can't find that in the manual. Does it come enabled by default (pins 1-2) or do you need to move to pins 2-3 to enable?

  2. #2
    Xtreme Enthusiast
    Join Date
    Oct 2007
    Location
    London, UK
    Posts
    575
    Quote Originally Posted by Kilyin View Post
    Can't find that in the manual. Does it come enabled by default (pins 1-2) or do you need to move to pins 2-3 to enable?
    VTT/QPI Load-Line Calibration is set to OFF by default. I would recommend people to manually enable it, as from using a multimeter, the VDrop on the unmodified VTT is around 0.05V, and the VDroop is another 0.05V again. So under load the real VTT value is actually around 0.1V less then what you set in BIOS. With the VTT/QUI LLC enabled, the VDrop is +.004V, and VDroop is -.005V compared to what is set in BIOS.


    Quote Originally Posted by creidiki View Post
    We are a band of fearless modern-day alchemists who, for fun, run solutions through sophisticated, if overpriced, separator setups, and then complain when we succeed in separating said solution.

  3. #3
    Xtreme Cruncher
    Join Date
    Feb 2010
    Location
    Poland
    Posts
    56
    Quote Originally Posted by eternal_fantasy View Post
    VTT/QPI Load-Line Calibration is set to OFF by default. I would recommend people to manually enable it, as from using a multimeter, the VDrop on the unmodified VTT is around 0.05V, and the VDroop is another 0.05V again. So under load the real VTT value is actually around 0.1V less then what you set in BIOS. With the VTT/QUI LLC enabled, the VDrop is +.004V, and VDroop is -.005V compared to what is set in BIOS.
    It's not as bad, I measured my droop on QPI with multimeter and 1.300 bios ends up with 1.260v after droop and while linxing, while 1.3500 ends up with 1.300v. I guess I won't be swapping that jumped anytime soon just for average water cooled OC.
    Quote Originally Posted by eternal_fantasy View Post
    Ah great.
    Well... It could have been worse!

  4. #4
    Xtreme Enthusiast
    Join Date
    Nov 2008
    Posts
    877
    Got myself one of these beast too



    Didn't do much testing as I got it yesterday, but crashed my car today, so not in the mood.

    First impression is that it's much more stable and cooler than my DFI UT X58.

    Is there a BIOS template, or should I make one?

    BTW, would anyone share that new beta bios?
    Maximus 5 Gene | i7-3770K @ 5GHz | ADATA 2x2GB @ 2.6GHz 9-12-10-28-1T | HD7970 @ 1200/6400
    Rampage 4 Extreme | i7-3930K @ 5GHz ||| X58-A OC Orange | i7-980X @ 4.6GHz

  5. #5
    Xtreme Enthusiast
    Join Date
    Oct 2007
    Location
    London, UK
    Posts
    575
    Quote Originally Posted by usz View Post
    It's not as bad, I measured my droop on QPI with multimeter and 1.300 bios ends up with 1.260v after droop and while linxing, while 1.3500 ends up with 1.300v. I guess I won't be swapping that jumped anytime soon just for average water cooled OC.
    Anyone that clocks high Uncore with high RAM speeds knows that VTT voltage is imperative to those settings.

    The Intel Spec advices maximum VTT of 1.4V for 32nm 980X, and 1.35V for the rest of the i7 900 series. And to run Memory at speeds over 1000Mhz often requires over 1.4V for 40nm where the UCLK multi is at least 2X the DRAM multi, and with the 32nm 980X with it's ability to run with 1.5X UCLK multi, requiring over 1.3V VTT for stable operation. So a deviance of 0.05~0.10V from what is set is VERY significant when the margin between what is regarded as "Safe" and "Stable" voltages are so small.

    Now, Asus provides a jumper switch to turn on load-line calibration, which brings the R3E's actual VTT voltage in line with what is actually set, with minimal droop, and in line with what other motherboards has already incorporated as default configuration such as the EVGA Classified.

    From your tone it seems that you are hesitant to touch the jumper as it's a physical switch on the motherboard as opposed to a setting inside the BIOS. Believe me, you can do as much if not more damage changing settings in the BIOS, assuming it does any damage at all in the first place. Before all these fancy BIOS menus and settings, many of the overclocking features were unlocked or even set using on board jumpers, and it actually turns out to be more reliable as it is a physical connection, and not dependent on the quality of the BIOS.

    What you do with the VTT-LL jumper is of course up to you, but don't undermine it's significance as a viable and very accessible setting, just implemented in a different way, which plays a very large role at increasing set voltage accuracy and stabilising memory/Uncore overclocks.
    Last edited by eternal_fantasy; 05-14-2010 at 06:14 AM. Reason: Corrected voltages


    Quote Originally Posted by creidiki View Post
    We are a band of fearless modern-day alchemists who, for fun, run solutions through sophisticated, if overpriced, separator setups, and then complain when we succeed in separating said solution.

  6. #6
    Xtreme Enthusiast
    Join Date
    Nov 2008
    Posts
    877
    Quote Originally Posted by eternal_fantasy View Post
    The Intel Spec advices maximum VTT of 1.4V for 32nm 980X, and 1.875V for the rest of the i7 900 series.
    Please change 1.875V to 1.35V as stated in Intel Spec, so someone don't get confused.
    Maximus 5 Gene | i7-3770K @ 5GHz | ADATA 2x2GB @ 2.6GHz 9-12-10-28-1T | HD7970 @ 1200/6400
    Rampage 4 Extreme | i7-3930K @ 5GHz ||| X58-A OC Orange | i7-980X @ 4.6GHz

  7. #7
    Xtreme Enthusiast
    Join Date
    Oct 2007
    Location
    London, UK
    Posts
    575
    Quote Originally Posted by donmarkoni View Post
    Please change 1.875V to 1.35V as stated in Intel Spec, so someone don't get confused.
    Yeah was a pretty obvious mistake. 1.875 was the maximum VDIMM voltage for the 40nm i7s. Corrected.


    Quote Originally Posted by creidiki View Post
    We are a band of fearless modern-day alchemists who, for fun, run solutions through sophisticated, if overpriced, separator setups, and then complain when we succeed in separating said solution.

Tags for this Thread

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
  •