Some thermal paste analysis from Team Australia

Discussion in 'Extreme and Water Cooling' started by youngpro, Dec 7, 2010.

  1. youngpro

    youngpro Member

    Joined:
    Jun 18, 2006
    Messages:
    1,451
    alright guys, probably a fairly dull subject but i want to take a look and find what the best options were for me with thermal paste.. i am not the guy that will speak about a product i don't believe in, and i will never speak without doing appropriate testing for myself..

    i have tested a number of thermal pastes that i had available to me over the past few weeks, i had a few criteria

    Criteria:
    * a paste that worked right off the bat.. i dont have time to cure a paste for 1000 hours, i want top performance straight away
    * a paste that can handle the high heat load of gulftown 980x and overvolted AMD thuban X6 cpu's

    Test platform:
    ASUS RIIIE
    Gulftown 980X 1.4 vcore
    Noctua NH-D14

    Please forgive that I only test a few thermal pastes, I tested what I had available to me at the time of testing..

    My testing consists of the following, install application of paste, boot system and load my settings of 4000mhz, 1.4v and idle in windows for 5 minutes, then LinX load for 30 minutes, then another 5 minutes recovery idle temperature.

    Results:

    [​IMG]

    [​IMG]

    [​IMG]

    Brief Analysis:

    Taking a quick look across the pastes, I think its easy to discard the Thermalright.. The thermalright is also very sticky and it goes everywhere if you get it on your hands.

    I was impressed with the results of the arctic silver, which is an old favourite that many people use but under the LinX load testing the clear winner is the Antec Formula 7 paste.

    It is very interesting that the Antec paste performed the best, I was sent this paste with the information that it required no cure time. I think this proves a strong benefit of the Formula 7 paste.

    For a long time now I have used the Noctua NT-H1 and Ceramique, and recently OCZ Freeze. But personally I will be moving to the Antec Formula 7 product for my overclocking from now on...
     
  2. rbdirty

    rbdirty Member

    Joined:
    Jun 6, 2008
    Messages:
    539
    Any sub zero action?
     
  3. eva2000

    eva2000 DDR1/DDR2/DDR3 Addict

    Joined:
    Jun 27, 2001
    Messages:
    21,902
    Location:
    Brisbane, Australia
    interesting that you have yet to use MX-2/MX-3 or MX-4 yet .. would be interested in such results, i mainly use MX-2 and MX-3 now a days :)
     
  4. Linkin

    Linkin Member

    Joined:
    Apr 25, 2010
    Messages:
    6,324
    Location:
    Sydney
    Hope you know that AS5 needs 200 hours to cure ;) And why no IC Diamond 7? It's been better than arctic silver every time i use but the only place I can get it ebay :thumbdn:
     
  5. OP
    OP
    youngpro

    youngpro Member

    Joined:
    Jun 18, 2006
    Messages:
    1,451
    as i mentioned mate, this is only analysing pastes i had at my house, i didn't go out and get anything special.. :) and also as mentioned, i know these pastes take time to cure, which is why antec formula 7 suites me best.. please refer to the criteria at the top..

    so many pastes George LOL, i have about 20 more coming, but it is difficult to test them all in the same manner, have to make sure ambient temp, etc, is all the same!

    no subzero action yet, that will come though :)
     
  6. gelfling

    gelfling Member

    Joined:
    May 22, 2004
    Messages:
    153
    Location:
    sydney
    be interesting to see how these hold up against Indigo Xtreme, Coollaboratory Liquid Ultra and the Shin Etsu compounds
     
  7. Little Man

    Little Man Member

    Joined:
    Sep 4, 2008
    Messages:
    4,601
    Location:
    3147
    What method of application did you use?

    Were they single mounts or did you remount a few times and get an average?

    Very nice, interesting seeing antec of all companies to come up with the goods.
     
  8. OP
    OP
    youngpro

    youngpro Member

    Joined:
    Jun 18, 2006
    Messages:
    1,451
    only 2 mounts each and average this time.. last time i did 3, but this time i was a little lazy,

    application method very much like this

    [​IMG]

    enough to give a even but thin spread across IHS
     
  9. zeropluszero

    zeropluszero Member

    Joined:
    Aug 6, 2008
    Messages:
    6,037
    Location:
    3132
    im not sure i approve of your TIM application method, i7/AMD/DC 775 i use a moderate dot in the middle, around 5mm.

    i have coolabratory liquid pro, MX-3, MX-4, Shin Etsu, Ceramique, AS5, Noctua something, Prolimatech, and others.
    if you want to test them im happy to sort something out with you. PM me.

    for the record, with all that i have, ive been using only MX-4 on my latest remounts since i got it. seems great.
     
  10. Huggy_Bear64

    Huggy_Bear64 Member

    Joined:
    Apr 24, 2010
    Messages:
    3,092
    Location:
    Melbourne, VIC, 3127
    I love how clean your socket area is :lol:
     
  11. ysu

    ysu Member

    Joined:
    Sep 27, 2005
    Messages:
    1,097
    Location:
    Central Coast
    nice :)

    I wonder how does it look like when you remove the heatsink after the test?
     
  12. bennyg

    bennyg Member

    Joined:
    Dec 16, 2005
    Messages:
    3,598
    Location:
    Melbourne, Oztraya
    While I approve of the conclusion that AS5 is still one of the best around ;) not sure I approve of the app method either.

    Not to moan, good on you for putting in the time and effort, but when you're talking 1-2C differences a tiny difference in coverage could make your results meaningless.

    I'd recommend a centre glob that's spread by the heatsink when you put it on - assuming the gap is the same you could tell how much you'd put on my how far it's spread on the IHS.
     
  13. OP
    OP
    youngpro

    youngpro Member

    Joined:
    Jun 18, 2006
    Messages:
    1,451
    i used to do that for along time but i have had better results with multi-blobs,

    but after spending an hour tonight testing both TIM application methods tonight my conclusion is, as long as its not too thick, it makes absolutely no different to the performance of the paste..

    my temps were about dead on identical, just a nice very thin layer of thermal paste across the IHS
     
  14. BlueRaven

    BlueRaven Member

    Joined:
    Jul 29, 2010
    Messages:
    4,191
    Location:
    Sydney
    I just ran out of OCZ Freeze (which I've been pretty impressed by) and was wondering what to buy, so thanks for posting this thread, very timely.

    TKR's got it, $12 / 1.5gm :thumbup:
    http://www.thekoolroom.com/index.php?route=product/product&path=39_53&product_id=481

    I'm tempted to try either the IC Diamond or an Indigo Extrreme kit, but I might wait for some more opinions in this thread first.
     
  15. squall_leonhart

    squall_leonhart Member

    Joined:
    Jan 4, 2007
    Messages:
    268
    Location:
    Australia
    IC7 is highly overrated, and the fools touting massive differences are either comparing against bad applications or stock goop.
     
  16. Louis Cyphre

    Louis Cyphre Member

    Joined:
    Jan 13, 2009
    Messages:
    238
    Location:
    Perth 6021
    I've read a number of your reviews & found them pretty good, but man... Are you into Impressionism or something? Surely your photo is in jest!?

    Is it Mr Squiggle on LSD? An upside down drawing of Miss Janes IUD?

    It's begging for dry spots. I've always used a squirt in the middle & let the pressure sort it out.


    Keep up the good work though.

    One thing I'd be interested in seeing is how the el-cheapo white paste (the stuff from Dick Smith or Altronics called 1hunglow or something) compares to the top end goop.

    Have a good Christmas

    SB
     
  17. jmart38

    jmart38 (Banned or Deleted)

    Joined:
    Jun 2, 2007
    Messages:
    6,912
    Location:
    ಠ_ಠ Brisbane
    Am I the only one who actually spreads the paste out? lol

    I just put a bit of glad wrap over my index finger and spread it out evenly...
     
  18. Frag_Lord

    Frag_Lord Member

    Joined:
    Nov 23, 2004
    Messages:
    1,263
    Location:
    Adelaide
    Thats what your heatsink/waterblock, etc does for you.
     
  19. ravencs

    ravencs Member

    Joined:
    May 5, 2009
    Messages:
    831
    Location:
    Sydney 2089
    No i also spread it out, but at the same time i do the random dots method.
    I have had quite bad results with the 'center blob' method using very thick pastes link shin etsu(sp?).

    Those two tubes lasted about 30 minutes on the desk before they were thrown into CBF'ed pile and the AS5 used.
     
  20. Ashpool

    Ashpool Member

    Joined:
    Feb 24, 2003
    Messages:
    3,352
    Location:
    Ye Olde Melbourne Town

Share This Page