Page 1 of 2 12 LastLast
Results 1 to 25 of 28

Thread: G1.Sniper 2 Power Check

  1. #1
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182

    G1.Sniper 2 Power Check



    Welcome peoples.

    I'm going to take a quick look at the power output of the G1.Sniper 2 Motherboard.
    I won't talk too much people just like looking at pictures anyway . So here we go.

    1st off checking what Voltages the board sets at default.

    Bios vs Multimeter (real)





    Pretty damn accurate, memory is spot on.

    Next up we have Gigabyte's famous array of LLC (Load Line Calibration) levels.

    What is LCC? LLC counters the voltage lost when the CPU go's under load.
    So what LLC does it tries to equalise the difference between IDLE and LOAD voltages.

    Gigabyte has 10! Levels, this is how they perform.


    BIOS SETTINGS

    Vcore 1.3
    Memory 1.65
    VTT / IO 1.10




    Very acceptable no LCC result!
































    After checking The IDLE and LOAD voltages we will now be looking at the cleanliness of the voltage feed.

    Very basic i measured the Ripple with an Oscilloscope.

    BIOS SETTINGS

    Vcore 1.3
    Memory 1.65
    VTT / IO 1.05





    Tests!



























    This board has surprisingly clean volts. Compared to the Z68X-UD4 this board is in a higher league.
    Also with UD4 the volts on Superpi32M had much more noise. So this is a definite improvement.

    I also did a quick temp check. But since Sandybridge came out PWM area's don't really get warm at all.







    aaand thats it! Told you it was quick

    Overall performance is very satisfying, as stated above it's visually and physically much stronger than the UD4,
    and from other reviews seems to be on par with UD7 with overclocking capability. And of course with it's better sound,
    network and PCI-E Gen3 features it is certainly Gigabyte's Highest end Z68 board and deserves that title.
    Last edited by Vivi; 09-04-2011 at 12:57 PM.

  2. #2
    Aussie God
    Join Date
    Feb 2005
    Location
    Copenhagen, Denmark
    Posts
    4,702
    Pretty sweet Vivi!

    I admire that you dont suck nuts on the vendor, and bring forward something interesting to read, you didnt do a massive speach of none sense for the users in here; such as NIC, sound and other features of no interest for overclocking

    I especially find the LLC part interesting, simple and interesting - I think 10 levels are silly... Off, slight, moderate, high is fine by me...
    Competition ranking;
    2005; Netbyte, Karise/Denmark #1 @ PiFast
    2008; AOCM II, Minfeld/Germany #2 @ 01SE/AM3/8M (w. Oliver)
    2009; AMD-OC, Viborg/Denmark #2 @ max freq Gigabyte TweaKING, Paris/France #4 @ 32M/01SE (w. Vanovich)
    2010: Gigabyte P55, Hamburg/Germany #6 @ wprime 1024/SPI 1M (w. THC) AOCM III, Minfeld/Germany #6 @ 01SE/AM3/1M/8M (w. NeoForce)

    Spectating;
    2010; GOOC 2010 Many thanks to Gigabyte!


  3. #3
    HWGurus
    Join Date
    Mar 2006
    Posts
    812
    Nice work Vivi

    What I would like to see is cpu voltage ripple at >5GHz with some higher voltage

  4. #4
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182
    Quote Originally Posted by M.Beier View Post
    Pretty sweet Vivi!

    I admire that you dont suck nuts on the vendor, and bring forward something interesting to read, you didnt do a massive speach of none sense for the users in here; such as NIC, sound and other features of no interest for overclocking

    I especially find the LLC part interesting, simple and interesting - I think 10 levels are silly... Off, slight, moderate, high is fine by me...
    I agree, anything about LLC level 5 is probly not recommended! so 3 levels should be good. 10 is a little confusing if you don't have a multimeter and test them all

    Quote Originally Posted by quake6 View Post
    Nice work Vivi

    What I would like to see is cpu voltage ripple at >5GHz with some higher voltage
    I will test that for you sir!

  5. #5
    Aussie God
    Join Date
    Feb 2005
    Location
    Copenhagen, Denmark
    Posts
    4,702
    Quote Originally Posted by Vivi View Post
    I agree, anything about LLC level 5 is probly not recommended! so 3 levels should be good. 10 is a little confusing if you don't have a multimeter and test them all
    Piece of advice though; NEVER tell Gigabyte marketing, you will be listed somewhere between graylist and as a "hater".
    Competition ranking;
    2005; Netbyte, Karise/Denmark #1 @ PiFast
    2008; AOCM II, Minfeld/Germany #2 @ 01SE/AM3/8M (w. Oliver)
    2009; AMD-OC, Viborg/Denmark #2 @ max freq Gigabyte TweaKING, Paris/France #4 @ 32M/01SE (w. Vanovich)
    2010: Gigabyte P55, Hamburg/Germany #6 @ wprime 1024/SPI 1M (w. THC) AOCM III, Minfeld/Germany #6 @ 01SE/AM3/1M/8M (w. NeoForce)

    Spectating;
    2010; GOOC 2010 Many thanks to Gigabyte!


  6. #6
    XIP
    Join Date
    Nov 2004
    Location
    NYC
    Posts
    5,535
    Nice info there Vivi

    I also want to know if this board is ok for extended benching @ 1.64+Vcore like its UD7 brother

  7. #7
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182
    Quote Originally Posted by Dumo View Post
    Nice info there Vivi

    I also want to know if this board is ok for extended benching @ 1.64+Vcore like its UD7 brother
    i Will answer that question soon dumo,

    overclocking it with LN2 this weekend, for GTX 580 comp

  8. #8
    I am Xtreme
    Join Date
    Dec 2004
    Location
    France
    Posts
    3,563
    nice job vivi, i found that the VRM was pretty good as well. The LLC levels work well as well, i was pretty impressed by this board.

  9. #9
    I am Xtreme
    Join Date
    Sep 2006
    Posts
    10,379
    Nice testing there....

    So in fact we need LLC version 4.5 to remain +/- spot on with the bios setting... Good to see the GB's positively evolve...

    I would add ths info into your existing thread mate, it will become a stickie for sure...
    Last edited by Leeghoofd; 09-05-2011 at 11:25 PM.
    Question : Why do some overclockers switch into d*ckmode when money is involved

    Remark : They call me Pro Asus Saaya yupp, I agree

  10. #10
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182
    Thanks leeghoofd good idea, i will go add the necessary's in the other thread . Like the voltage output and LLC.

  11. #11
    Xtreme Member
    Join Date
    Nov 2006
    Location
    Bar,Montenegro
    Posts
    164
    Nice, simple and clear....Can't wait to see how it behaves on higher clocks and voltage also.......



  12. #12
    Xtreme Addict
    Join Date
    Apr 2005
    Location
    Calgary, AB
    Posts
    2,252
    I find these results interesting, I got a bit different results for Z68XP-UD5 & UD7. I found that LLC 7 gives closest to what you set in BIOS for 1.3v, 1.4v, and 1.5v. I also found that the more voltage you set in BIOS, the more LLC will overvolt so the results at 1.3v are quite different than 1.6v. I can share my charts and tables once my review goes live.
    MB Reviewer for HWC
    Team OCX Bench Team

  13. #13
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182
    Yea i've heard that aswel,

    so we would need to test Air cooling optimum LLC at your stable volts,

    and find optimum LLC for extreme cooling at 1.6v.

    Thanks i'll be interested in your reviews.

  14. #14
    Xtreme Addict
    Join Date
    Apr 2005
    Location
    Calgary, AB
    Posts
    2,252
    Oh yah I also found that the more volts you use, idle vcore starts to increase with LLC level. Up to 1.4 idle doesnt move but above that it increases with LLC which in turn gives higher load. Have you ran any 1.5v+ LLC tests? You will see what I mean. I would be curious to see how an 8 phase board handles hi volt LLC compared to 20 or 24.
    MB Reviewer for HWC
    Team OCX Bench Team

  15. #15
    I am Xtreme
    Join Date
    Dec 2004
    Location
    France
    Posts
    3,563
    I think it will do just fine. if you look at the M4G-Z or M4E, or asrock extreme 7 they are ll 8 phase CPU power boards. 8 phase IMO is more than enough. They all output around the same amount of power. The Chokes in the UD7 and Sniper 2 are rated differently. the sniper 2 can output a bit more current. in theory ripple should be lower on the UD7 or UD5 because of increased ripple frequency with increased phase count, which lowers ripple current. but its in theory, but the Sniper 2 has more output capacitors, which helps tremendously. if you compare the amount of output capacitors you will see that the higher the # of phases the lower the cap count on GB LGA1155 boards, this is because they are banking on the increased phases to help keep ripple at bay. i think the reason the Sniper 2 has great voltage regulation, as i tested with my scope and found the same thing, is because its an almost perfect balance of phase count to output capacitance. Also switching frequency can be higher on the 8 phase than 24, 500khz vs 250khz(estimates), but we see that they do operate at the same switching frequency. Now if GB has 20x470uF output caps with a 20 or 24 phase board, that will be freaking interesting. i think the X79-UD7 has 19 of those POSCaps with 20 phase for CPU. but if we look at teh X79-UD5 is has many less phases, but same number of capacitors on the output, which to me indicates that GB didn't rely solely on phase interleaving for ripple control, they did their best on the X79-UD7s VRM, but only testing will tell. for reference the X58A-OC had 12 phases with 20x 470uF, so bank on the X79-UD7 to have a much better performing VRM. See there are different approaches to maintain fast transient response and ripple control, the most common is to have increased switching frequency, but GB has kept SF low on their LGA1155 boards, and used the fact that their PWm is fast as well as some other tech to help keep up. its like a paradox.
    Last edited by sin0822; 09-23-2011 at 03:01 PM.

  16. #16
    ODOC
    Join Date
    Aug 2006
    Location
    Copenhagen - Denmark
    Posts
    2,193
    Hey man thats pretty nice work

    keep power checking it

  17. #17
    Hardware Nightmare
    Join Date
    May 2004
    Location
    Taipei :)
    Posts
    855
    Oh, good try, but...
    Nice pictures, but that say nothing tbh. No any idea about scale of that scope photos, zero information about testing method and connections, subzero about used test tools.

    What's up with all this recent "scifi"-looking GBT reviews, guys? It's really not worth time wasted on this, much more useful data could be done by testing storage/BIOS/software tools, some OC guides, than this.

    Vivi, get me right, it's nothing against you, just community will benefit more for more reallife tests and reviews.
    Taking GND reference from another galaxy

    Electronics engineering @ extreme overclocker
    LN2: Cel347@8199.5MHz,920@5300,E8600@6610,QX9650@5700,X 3050@4311,X3470@5060

  18. #18
    I am Xtreme
    Join Date
    Dec 2004
    Location
    France
    Posts
    3,563
    Tin get over it, stop being so nice. Maybe 20 ppl in this community actually know how to read a scope, and the rest don't care much for it. At least presentations like this make them go look up how to read a scope shot. I think anything technical is good. Why don't you go test I/O speeds? i see no issue if he takes his readings, and compares against his other readings.

    he did a OC guide for the Sniper 2.

    Tin, get me right it's, nothing against you.
    Last edited by sin0822; 09-27-2011 at 06:40 AM.

  19. #19
    Hardware Nightmare
    Join Date
    May 2004
    Location
    Taipei :)
    Posts
    855
    Why to bother then with scope pics if only 20ppl in community know what is that? For "wow"-factor?

    I think much more enthusiasts actually understand simple time-level XY chart, than you think. Technical is good, that's right, but there is nothing technical in scaleless numberless photos of something Who knows, maybe that's 50Hz AC mains waveform, and you responding "wow, cool, so nice measurements data, VRM is really good!" Any comparsion should be meaningful, not just comparsion for comparsion.

    P.S. I'm technical ass, yea, working with PC hardware on daily full-time job basis get some requirement for data posted on web by other well-known power users. If you think that posting wrong data is ok, go first to local mods and ask them to ban me :P
    Last edited by TiN_; 09-27-2011 at 06:13 AM.
    Taking GND reference from another galaxy

    Electronics engineering @ extreme overclocker
    LN2: Cel347@8199.5MHz,920@5300,E8600@6610,QX9650@5700,X 3050@4311,X3470@5060

  20. #20
    I am Xtreme
    Join Date
    Dec 2004
    Location
    France
    Posts
    3,563
    who said hes posting something wrong?

    if i post one of my charts, where i measure the LLC levels with software and DMM and just add numbers, that does mean I am wrong if i don't show 100 shots of my DMM?

    i am just saying, sometimes there is no point in showing an exact picture of everything. if you saw his review before this, he took shot of his scope, an it didn't look easy to get the lines on the display to show. if you saw my shots of the scope for teh 990fxa-ud7 review you will see i hit over 100mv vpp, that isn't something that is good, but with the sniper 2 i saw much better vreg.Compared to the UD4 he is seeing better vreg with this board. if you take both independent reviews and compare you'll see that the sniper 2 has a good vrm, its just hard to show it.

  21. #21
    I am Xtreme
    Join Date
    Dec 2008
    Location
    France
    Posts
    8,832
    Quote Originally Posted by TiN_ View Post
    but there is nothing technical in scaleless numberless photos
    Donate to XS forums
    Quote Originally Posted by jayhall0315 View Post
    If you are really extreme, you never let informed facts or the scientific method hold you back from your journey to the wrong answer.

  22. #22
    I am Xtreme
    Join Date
    Aug 2008
    Posts
    5,584
    I think "power check" is misleading for engineers as there is no power units seen - lol

    voltage regulation perhaps?


  23. #23
    take me to the manbar
    Join Date
    Apr 2008
    Location
    South Africa :)
    Posts
    1,182
    Yea i guess my target audience was more people who are not yet sure of anything going on in a motherboard, was 1stly intended for gamers.

    power check is just a name nothing more, i did basic tests (i didn't want to put too much technical info in there because people will just stop reading cause they don't know what is going on). Hence the simple pictures.

    Totally agree with you though TiN, there is no way i can compete with your level of knowlege. The reviews you did on UD7 and Maximus were perfect, but less views than it should have.

    Writing a review is like making a movie,

    if you want to see a thriller don't go watch a romantic comedy and say it's not scary enough.

    thanks for the interest tho
    Last edited by Vivi; 09-27-2011 at 01:21 PM.

  24. #24
    I am Xtreme
    Join Date
    Dec 2004
    Location
    France
    Posts
    3,563
    id call this a power check, it fits the bill.

  25. #25
    Xtreme Legend
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    17,468
    nice job, I like the way you present the info, clean and simple for most ppl to "get" without having to get a level 1 certificate in EE to understand what the fluck some nerd is talking about heheh
    Team.AU
    Got tube?
    GIGABYTE Australia
    Need a GIGABYTE bios or support?



Page 1 of 2 12 LastLast

Posting Permissions

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