Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Page 1 of 2 12 LastLast
Results 1 to 10 of 11

Thread: Load-Line Calibration...




  1. #1
    Join Date
    Apr 2009
    Posts
    52

    Default Load-Line Calibration...

    I have an ex58-ud3r...
    Is Load-Line Calibration ok to use?
    Can someone give me a good explanation of what it does, and any potential dangers?
    I'm trying to find some good info, but some people say it's safe...
    others say it's not safe... what's the truth?

  2. #2
    Join Date
    Sep 2008
    Posts
    32

    Default Re: Load-Line Calibration...

    Quote Originally Posted by someguyonapc View Post
    I have an ex58-ud3r...
    Is Load-Line Calibration ok to use?
    Can someone give me a good explanation of what it does, and any potential dangers?
    I'm trying to find some good info, but some people say it's safe...
    others say it's not safe... what's the truth?
    I have load line calibration "enabled" and my OC works better.
    I think is not dangerous if your cpu voltage is not excessive (in my case 1.3v for E8400) and you control the heat.
    You will avoid Vdrop/Vdroop and this can be better for overclock.
    Last edited by Alicantino32; 05-13-2009 at 01:00 PM.

  3. #3
    Join Date
    Apr 2009
    Location
    J'Habite En Angleterre
    Posts
    2,431

    Default Re: Load-Line Calibration...

    LLC controls the amount that your core voltage will "droop" when loading the CPU. It often does little for "vDrop" which is the difference between voltages set in BIOS, and ones actually supplied to the CPU in Windows.

    LLC is considered out of spec on Intel platforms at least. The processor is designed with vDroop in mind. You won't particularly do any harm enabling it, however why bother doing so if you can get stable up to ~1.4v without it?

    With LLC off, the pattern of power fed to the CPU will never exceed the set vCore, however LLC can cause patterns of voltage that often exceed said set vCore.

    A rule I stick to is if I can get a Core 2 processor (dual or quad) stable at 1.4v or less with no LLC, I will. If I can't I drop back to 1.3500v with LLC enabled and test up/down accordingly.

    Not saying this is 100% correct, I just wouldn't want to use LLC to be able to use low looking vCores, that in actual fact go out of spec and aren't so perfect looking in reality.
    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

  4. #4
    Join Date
    Apr 2009
    Posts
    52

    Default Re: Load-Line Calibration...

    Is there a way to measure the voltage spike?
    I have my Vcore set to 1.3375 in BIOS... cpuz shows 1.328 idle; 1.312 to 1.296 under load with LLC enabled. How high would it be spiking?

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

    Default Re: Load-Line Calibration...

    To measure the CPU clock signal and supply voltage in any meaningful way, you'll probably need an oscilloscope, and the knowledge of where to connect it.

    Here is a link to an Anandtech article. They do make LLC out to be quite evil, however it isn't. Still would be preferable to not use it unless you have to go over recommended voltages for your CPU with it off.

    AnandTech: Overclocking Intel's New 45nm QX9650: The Rules Have Changed
    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. #6
    Join Date
    Apr 2009
    Posts
    52

    Default Re: Load-Line Calibration...

    I have my pc connected through a good APC surge protector which is supposed to help reduce spikes and noise by sending excess voltage to the ground.
    I don't know how good these things are, but it sounds good.

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

    Default Re: Load-Line Calibration...

    That's good practice, many don't bother with surge protectors. I have one too, and run all my PC equipment through it.

    The voltage spikes referred to here though, go on internally, and are controlled by the motherboard. They are created by the power circutry itself.

    As you can see in the link, when LLC is disabled, these voltage spikes are perfectly OK, and intel actually state that VRMs for their CPU's must behave like this. When LLC is off, the peak voltage is never more than the vcore you set. However when you enable LLC, the voltage changes (which are normal) are still there, but the graph axis has been shifted by artificially eliminating vDroop. This then means that the perfectly normal voltage peaks are now higher than the maximum vCore you set.

    In all likelihood, you won't have a problem setting LLC, however like I say, it is out of spec, and causes you to loose fine control over vCore rather than gain it. If you can run up to 1.3625v stable with it off, I would recommend you do so. Otherwise enable it, and use the least amount of voltage to be stable (as always).

    *EDIT* As you are using a Core i7 CPU, max recommended vCore is 1.375v.
    Last edited by Psycho101; 05-14-2009 at 06:01 AM.
    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. #8
    Join Date
    Apr 2009
    Posts
    52

    Default Re: Load-Line Calibration...

    Right now I've got Vcore 1.34375 in BIOS... LLC enabled.
    To get the same OC with LLC disabled I need 1.37500 Vcore in BIOS.
    Seems that my CPU needs no less than 1.296V under load to remain stable.
    How far could 1.34375 Vcore be spiking?
    Can anyone offer a reasonable guess?

  9. #9
    Join Date
    Apr 2009
    Location
    J'Habite En Angleterre
    Posts
    2,431

    Default Re: Load-Line Calibration...

    The approximate answer is in the Anandtech article.

    Look at the last graph. Move the axis of the graph so that it matches yopur core voltage settings, and you have your answer. You will have to either work out the axis shifting in your head, or plot the graph on some graph paper, using the apropriate axis orientation for your current settings.

    If you only need 1.37v with LLC off, I'd be tempted to leave it off. Do you see any difference in load temperatures between the two stable settings? (check LLC on, then LLC off after powering down for 5 min to let everything cool). 20 minutes of Prime Small FFT or 5 iterations of IBT for each measurement of temperature should be fine.

    How much of an over volt isn't really that important. Sufficed to say at those voltages it's not going to be harmful, but if not needed, don't use LLC. The amount vCore will go out of specified range increases as the vCore set increases.

    If it were my PC I'd run 1.37v LLC off.
    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

  10. #10
    Join Date
    Apr 2009
    Posts
    52

    Default Re: Load-Line Calibration...

    Enabled - Vcore in cpuz shows 1.328 idle; 1.312 to 1.296 load.
    Disabled - Vcore in cpuz shows 1.344 idle; 1.312 to 1.296 load.

    From that I would say that it really doesn't matter much.
    Idle voltage is slightly less, but nothing drastic.
    I'll play around a bit more, but I don't see a reason to not use it.
    Oh, and I was under the impression that 1.365V was Intel max Vcore for the i7.
    1.375V would make things a tiny bit easier.

Page 1 of 2 12 LastLast

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
  •