Ergebnis 1 bis 11 von 11
  1. #1
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard Xigmatek Dark Knight 1283V CPU HSF review

    Xigmatek Dark Knight CPU HSF Review

    Intro.

    Xigmatek are a relatively new name to the world of PC Hardware cooling but have built up a wide range of products and have been given positive reviews by reviewers world-wide. For a new company, availability is good, which is no small feat- especially in this economic downturn.

    Xigmateks focus is on aircooling, where they have no less than 13 CPU coolers. In addition to quantity, they also have innovation- Their HDT (Heatpipe Direct Touch) design allows for the best possible thermal transfer from CPU to heatsink, which should provide that last scrape of stability when running on the edge of what aircooling can offer- especially useful for high heatloads like Intels I7 chips.

    On the test-bed in this review is their Dark Knight (1283V) CPU cooler.





    The Dark Knight

    This cooler comes in a conventional box-y box with a perspex window to show off the fan. Artwork includes a sword and a picture of the complete HSF unit. Core I7 compatability is explicitly mentioned.


    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight006.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight007.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight025.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight026.jpg[/IMG]



    The box contents are: the heatsink itself, a 120mm Xigmatek fan, 4 springs, 4 threaded bars, backplate + insulation, AM2 crossbar, 2x mounting arms for Intel solutions, 5 fan mounting grommets, an instruction leaflet and some thermal paste.


    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight019.jpg[/IMG]



    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight011.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight012.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0205.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight018.jpg[/IMG]



    The HSF comes as standard with a sprung bolt mounting system that replaces the twist lug method used in Intels stock retention mechanism. An electrically insulated backplate is included as a screw-down base. The heatsink is tightened using four Phillips screw heads. The springs are fairly tense- not too much effort will be required to tighten them, although some effort is needed to push the threaded section down far enough to make contact with the backplate. It's still far preferable to Intels stock approach though.

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight023.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0188.jpg[/IMG]



    The fan is a smokey, transluscent colour. A mix of dark brown and grey. There are 4 white LEDs in the frame that give the unit a unique look when powered up. At full speed its... noticeable. For an overclocking session, id consider it quiet compared to many dedicated fans. For gaming it will also be fine but for watching DVDs or office work, it might annoy slightly, depending on the mobos implementation of E1ST or CnQ

    [IMG]http://i2.photobucket.com/albums/y38/K404/XigmatekDarkKnight015.jpg[/IMG]





  2. #2
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard



    Regarding mounting

    Bare HSF dimensions: 120mm wide, fin height of 110mm and depth 50mm.

    Distance from mobo to bottom fin:
    775: 39mm
    1366: 38mm
    AM2: 50mm

    For I7 and 775 mounting, the cooler can be set up either horizontally or vertically- both orientations have points to be aware of.

    For the horizontal setup, theres a high risk of the cooler not fitting around RAM or PWM cooling. For vertical attachment (more popular,) the fan must be pushed FULLY onto the rubber mounts, else, depending on the mobo layout, there might be problems with giving RAM cooling solutions enough space (as demonstrated below)

    Additionally, for vertical mounting, if the CPU socket centre is less than 60mm away from the edge of the board, the heatsink will overhang the edge, which might be a problem depending on the case used and the space available.

    For AM2 mounting, because of the size of the crossbar latch, it has to be positioned on the side opposite to the fan, else the fan will not fit.

    The most important thing needing said about AM2 mounting is that due to where RAM slots are typically placed in relation to the CPU, RAM with tall heatspreaders will probably NOT BE COMPATABLE with these coolers. As pictured below, there is a hefty overlap from the fins. I suspect this will affect most HSF combos, so Xigmatek should not be heavily penalised for this, but a scalloped design would give them a marked lead over most of the competition for AM2 support.



    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0155-2.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0152-1.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0155-1.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0158-2.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/DSCF0156-1.jpg[/IMG]


  3. #3
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard




    Test setup, notes, assumptions + drawbacks


    CPU+ Mobo combos:

    s775: Intel C2D E86 + Asus X48 P5E64 WS Evo

    s1366: Intel I7 920 C0 + DFI X58 EH6 w/ 3x2GB Crucial Ballistix

    s940: AMD 5600+X2 + XFX 8200 w/ 2x1GB Crucial Ballistix

    The AM2 setup is my 24/7 computer. The components picked for that are not really great overclockers.

    The I7 setup had settings picked to keep the RAM and Uncore MHz fairly consistent. Uncore was 3200-3300MHz, RAM was kept at 800-825MHz. RAM timings were 8-8-7-1T, 1.65v vTT was 1.44v

    The I7 mobo could not set vCore lower than 1.00v. 1.02v was the lowest voltage used.

    The I7 was not tested with a stock cooler.

    The E86 rig was not tested at stock. Its a sub-zero setup and is "tuned high" for competitive bench runs.

    All tests were done on an open bench rig. I appreciate this will have a definite effect on temperatures, but the plus side of this is that other heat sources have less of an effect on CPU temperatures due to less thermal buildup, so results do not become skewed.

    This will be the most time i've spent with aircooling since I moved to 775 in summer 2006. My test methodology will probably not measure up to that of the hardcore aircooling guys. I do expect it to be good enough to provide meaningful, comparable information.

    A K-probe was tested for a hardware-based temperature result but the results were considered very inaccurate and were not used. The distance between the probe/IHS surface and the core- a distance that could not be reduced- was considered to be the main culprit, so I had to rely on software readings.

    OCCT was used for generating a consistent heatload. The test was ran for 30 mins with a large data set under normal priority

    Speedfan was used for the software-based temperature readings for AM2 while CoreTemp was used for the Intel setups. The OCCT graphs generated for Intel gave consistently lower temperatures than CoreTemp by typically 5 degrees. I have made reference to both sets of data.

    For plotting the temperature data, the worst-case scenario data was used. ie. the lower ambient temperature recorded and the higher CPU temperature recorded.

    Arctic Ceramique was used for all testing.

    Because of the mixed ambients encountered during testing, all the results have a second data interpretation in terms of degrees Celcius above ambient. I felt this was necessary to prevent the results becoming skewed. However, the graphs are all in terms of the temperatures recorded.



  4. #4
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard



    Test Results

    [IMG]http://i2.photobucket.com/albums/y38/K404/AM22930GHz.jpg[/IMG] [IMG]http://i2.photobucket.com/albums/y38/K404/AM23132GHz.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/AM2numberswithA.jpg[/IMG]

    The BIOS only went up to 1.45vCore and I had no real interest in pushing the board to its limit



    [IMG]http://i2.photobucket.com/albums/y38/K404/All775.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/775numberswithA.jpg[/IMG]






    [IMG]http://i2.photobucket.com/albums/y38/K404/DKI7.jpg[/IMG]

    [IMG]http://i2.photobucket.com/albums/y38/K404/I7numberswithA.jpg[/IMG]

  5. #5

  6. #6
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard



    Intel I7 OCCT Graphs:
    [IMG]http://img132.imageshack.us/img132/2905/darkknightstocki7102v.th.png[/IMG] [IMG]http://img20.imageshack.us/img20/9735/darkknight3300mhzi71075.th.png[/IMG] [IMG]http://img262.imageshack.us/img262/2808/darkknight3400mhzi7110v.th.png[/IMG] [IMG]http://img132.imageshack.us/img132/9653/darkknight3500mhzi711v.th.png[/IMG] [IMG]http://img90.imageshack.us/img90/7794/darkknight3600mhzi7115v.th.png[/IMG]
    (L-R: I7@ 2660MHz, 3.3, 3.4, 3.5, 3.6GHz)


    Conclusion.

    The HSF looks great, the mounting mechanism is easy to work with and the temperatures are comfortably lower than stock. For AM2 setups, from a performance point of view- a quality aftermarket HSF is maybe going to waste as the AM2 just doesnt kick out enough heat for a new cooler to be worthwhile. Either that or the stock HSF is actually decent! For 1366 and 775 setups- the Dark Knight is a good HSF for squeezing a better overclock, without doubt.

    My thanks to Tony @ Xigmatek for the review samle and the other goodies he supplied as well

    Pricing: coming in at £37 on Scan: http://www.scan.co.uk/Products/Xigma...-Intel-AMD-CPU Its by no mean the cheapest cooler on the market but the looks, ease-of-mounting and performance make sure its worth serious consideration.



  7. #7
    AwardFabrik Crew
    Strebert
    Avatar von SeLecT
    Registriert seit
    15.09.2006
    Ort
    An der schönen Nordsee.
    Beiträge
    9.730

    Standard

    Thanks for this Review!

    I have the same cooler in my rig and i tested the Dark Knight with i7 too. My results are nearly the same.

    Here a prime screen with 3,6Ghz:

    [img]http://www.pctunerup.com/up/results/_200907/th_20090706235558_36Ghz2.jpg[/img]

    Cooler was the Dark Knight too, but i've changed the fan, because the original was too loud for mee.
    AwardFabrik - a tribute to hardware

  8. #8
    AF-hwbot-team Avatar von PLAGANOS
    Registriert seit
    21.07.2007
    Ort
    Emsdetten
    Beiträge
    6.379

    Standard

    very detailed review thx

  9. #9
    AF-hwbot-team Avatar von chriFFer
    Registriert seit
    03.11.2007
    Ort
    Hessen -> Haiger - L.A.
    Beiträge
    1.158

    Standard

    Very nice review . Great work.

  10. #10
    AwardFabrik Crew
    Mr. 7 fu***** IB-Hertz!
    Avatar von angoholic
    Registriert seit
    18.12.2008
    Ort
    Tiefkühltruhe
    Beiträge
    2.548

    Standard

    Great and very detailed work!
    "websmile-hört-auf"-Counter: 6 (Stand 12.11.2014)

  11. #11
    BenchTec UK
    Registriert seit
    05.08.2008
    Beiträge
    30

    Standard

    Some afterthought testing


    All the results in the original review were achieved with the fan set up to push air through the heatsink. I reversed the fan so it pulled air through the heatsink and got a 2 degree drop in temperature.

    Pushing air over
    [IMG]http://i2.photobucket.com/albums/y38/K404/Pushairover.png[/IMG]

    Pulling air though
    [IMG]http://i2.photobucket.com/albums/y38/K404/pullairthrough.png[/IMG]

Berechtigungen

  • Neue Themen erstellen: Nein
  • Themen beantworten: Nein
  • Anhänge hochladen: Nein
  • Beiträge bearbeiten: Nein
  •