NZXT Kraken G10 GPU Water Cooler Review on an AMD Radeon R9 290X

Jump To:

AMD Radeon R9 290X Water Cooled Temperature Testing



So, does water cooling the AMD Radeon R9 290X help lower GPU temperatures? Does it ever!  Over our 12 minutes benchmark run we hit 94C on the GPU with the Sapphire Radeon R9 290X retail card with the AMD reference cooler. With the NZXT Kraken G10 mounting bracket and the NZXT Kraken X40 water cooler mounted up to the card we hit just 44C on the GPU when running with the Extreme preset on the NZXT Kraken software program. This is a 50C degree temperature drop, which is insane and this is a perfect example of how water cooling is superior to traditional air cooling methods.


Here is a look at a GPU-Z screen shot we took of the max sensor values after playing Metro: Last Light at 1920×1080 for about half an hour straight.


The AMD Radeon R9 290X is fully dynamic when it comes to voltages and core clock speeds thanks to AMD PowerTune 2.0 technology. The AMD Radeon R9 290X is also known to throttle in high-heat situations, so we also wanted to see if water cooling the Radeon R9 290X would help increase the core clock speeds at 1000MHz, which is their peak speed. Our testing found that during the benchmark the Radeon R9 290X with the reference GPU cooler averaged 991.9MHz and with the card water cooled we averaged 992.3MHz. We were expecting to see the clock speed pegged at 1000MHz, but for unknown reasons that is not the case. We thought it might be because the GPU load was not at 100% at all times, but when we charted GPU load versus clock speed (chart) we couldn’t see any direct association with the core clock decreases.


The NZXT Kraken G10 is not a full coverage water block, so that means the GDDR5 memory and the VRM components are no longer dissipating heat to the reference cards heatsink. In the past we have found this to be a major issue as some parts of the card can overheat and actually run hotter than before. The NZXT Kraken G10 helps combat this by having a 92mm fan on the custom designed bracket to help keep the PCB components nice and cool. GPU-Z reads the temperatures of two areas of the VRM. We aren’t sure what two VRM areas GPU-Z is reading though.  We asked AMD PR and they thought that VRM1 is most likely VDDC (the 3D core power rail) and VRM1 is most likely VDDCI (the MC power rail). VRM 1 was a bit warmer on water when at idle, but once the GPU was under load it was actually cooler. During the benchmark run the air cooled card averaged 63.9C and the water cooled card averaged 62.2C at VRM 1.


VRM Temperature 2 showed a huge decrease in temperature with the NZXT Kraken G10 + Kraken X40 water cooling combination.  The Sapphire Radeon R9 290X retail card with the AMD reference cooler averaged 78.4C and the NZXT water cooled setup averaged just 51.3C!


When it comes to GPU voltage it looks like the Radeon R9 290X with the reference cooler used an average of 1.134V during the actual benchmark run whereas the watercooled card used 1.156V.  When at an idle power state the Radeon R9 290X with the AMD reference GPU cooler averaged 0.953V versus 0.961V with the water cooler attached. Why would the cards voltage increase when water cooled? AMD informed us that with the new Powertune that clock and voltage are now defined by a relationship. So if the GPU is spending more time at higher clocks in a given situation, the associated averaged voltage will be higher. I.e. the GPU will tend to ride the clock/voltage relationship higher – It is no longer true that only clocks that are affected. Like most modern VRM designs, it looks like load line is in effect here. So the heavier the current draw demands, the lower the voltage will droop relative to the set point. If you are cooling the GPUs extremely well like we are here, the power consumption will be lower and current draw will be less. Hence the voltage will not droop as much (vs Vset) when the GPU is kept this cold.


The largest shock when testing the AMD Radeon R9 290X was the VDDC power output reading of the card in Watts.  This is basically the calculation of GPU-Z 0.7.4 that shows the power output of the entire card in Watts. This also can be understood as how much power the card is using. Legit Reviews spoke with W1zzard over at TPU about this data value and he said this is a reading provided directly by the voltage controller and that it likely is the result of multiplying the measured voltage with the measured current of the card. The Sapphire Radeon R9 290X with the stock cooler averaged 146 Watts with a peak of 171.5 Watts.  The same card with the water cooler averaged just 120 Watts with a peak of 147 Watts. The fact that GPU-Z was using 26W less power with water cooling was astounding.We showed these results to AMD and they said that at lower temperatures that there will be less leakage across insulators inside the GPU.  When talking about a temperature drop of 94C to 84C that difference is usually negligible, but since we had a 50C temperature drop that could be part of the reason we are seeing huge power savings.

One would think that with the additional circuit board, water pump and cooling fan that the power consumption would increase… Oh wait, we unplugged the reference cards blower fan and the new water cooler is plugged directly into the motherboard. It looks like we’ll have to look at the systems overall power consumption at the wall to see if there really is a power savings to be had.

On the next page we’ll cover frame per second performance in Metro: Last Light, take noise readings and overall system power consumption numbers.

Jump To:
  • Andrew

    These benchmarks are crazy, I know the reference cards run a little hotter than the Sapphire R9 290 Vapor X OC cards but didn’t know they run 10 degrees hotter. Mine overclocked in MSI doesn’t go above 75c in the summer 90+F weather. Watercooling is more expensive it’ll cost you around 400-500bucks for everything, but if you wish to keep your room cooler this is the only way to go. Use EK parts not this Karken BS.

  • Andreas Elf

    I bought an extra GPU for CrossFire, but the top GPU obviously gets too hot.
    This would have been perfect. But I realized that it’s too big, the bottom GPU wont fit with that in the way and the USB ports are poorly placed on the MB so I can’t have the bottom GPU in the most bottom PCI-E port ;'(

    I’m thinking about selling the GPU again and wait 4-5 years until the CPU don’t have anything more to give and then build a custom loop.

  • Talesseed

    I installed the g10 + x40 on my r9 290x but i’m getting 70-80 deegres on silent ?!?!

  • Jarod

    I just installed the G10 and a X60 on my r9 290. But my temps are about 50-55 running heaven benchmark on silent setting… Is there something wrong with the x60 or have I installed it wrong? I should have about the same temp as you guys 🙁

    • It also appears they’re running purely on an exposed test bench, not in a case. That’s going to give temps lower than what you can expect running it in an enclosure where air flow around the other components is going affect overall temperatures. If you have an air cooler on your CPU and have the radiator mounted in an exhaust position, you’re going to have higher temperatures.

      50-55C is actually pretty good under load, though. What temperature you get is going to depend on a few things: where in your case you have the radiator mounted, the ambient temperature in your room, the ventilation around your case, airflow where you have the radiator mounted, and so on.

      If you want cooler temperatures, I’d say to start by changing where you have the radiator mounted. If you have it mounted at the back or top in an exhaust configuration, mount it on the bottom or front (if it can reach) in an intake configuration so you’ll have fresh air from the outside going across your radiator instead of the warm air from the inside of your case. Another thing that’ll help is putting heatsinks on the VRMs and memory.

  • For those curious about using heatsinks under the G10, I found this video on YouTube that shows the kind of heatsinks you can use with this for the memory and VRMs:

  • I have 2 ASUS Radeon 7970 GHz Edition Reference Cards with a Asus Crosshair Formula V Motherboard. Would it be possible to use 2 G10 Gpu mounting kits for cooling both the cards. I wanted to use this because it is cheaper than EKWB or XSPC and according to your reviews there’s is almost an astounding temperature drop. If you have any solutions kindly reply. Thanks for the review.

  • Benjamin Mou

    i dont care if you have to write another 3,000 words, GO INTO OVERCLOCK! 😛

  • tm

    Can you list the vreg temps during this run? I’d like to know what temps were with no heatsink on them.

  • nono

    Do you think it would be possible to fit the 290 and the watercooling system in a bitfenix prodigy ?
    That would be an awesome news for some people but I have some doubts as I see the mobo position and the gpu seems so close to the side wall…

  • Lordz

    Is there room to fit heatsinks to the vrams and memory chips under the kracken?
    or does the bracket take up too much space

  • t.t

    Any chance that you guys can try to see if you can fit a GPU at the third pci-slot?
    Really want to run crossfire, but im stuck on m-atx so If this solution does take 3 slots, Im out of luck

  • Quake2

    Nice, what I was expecting abount power consumption. AND I think I does not need to go that low in temperature to get that 44watts back, it will do the same at around 65-70C. Hawaii is very efficient indeed.

  • Steve Smith

    Great review, just ordered a 290x and going to order this cooling setup too. Did you get the chance to overclock it at all?

  • john

    @Nathan: Are you sure these readings are correct?? 120 with peak 147 W powerdraw? Can you recheck those as I’ve seen some reviews placing the gtx780Ti @ 262 W – there is really too much of a difference can you somehow verify those power readings? If so … wow … just WOW! It would mean that the 290x can achieve with this kind of cooling ~40% better efficiency then NVIDIA? Even with stock cooling it would be ~30% more power efficient then the 780Ti. Considering this to sum it up the chip is 30% smaller yet 30% more efficient then the 780Ti only to be edged out by ~5% in raw performance? If this is true this is quite THE achievement of AMD in ages! Can you verify the numbers so we don’t all get too excited for nothing :D?

    • Nathan Kirsch

      John, Those are the readings with GPU-Z 0.7.4 and they are showing the readings from the software. I didn’t believe it myself and spoke to W1zzard over at TPU and AMD about it before publishing the results (why the article didn’t go up on Monday). AMD said that lowering the temperature should result in lower leakage and that appears to be the case. You can see the power consumed at the wall with my Kill-A-Watt meter on the next page.

      • john

        Well this is quite cool – got to say it… It would be interesting to see the same analysis done on 780Ti with the same type of cooling and the same type of power measurement and normalize performance to get absolute performance clock for clock(1ghz).

        Efficiency = Mean Performance / W.
        Chip quality = Efficiency / diesize.

        Also I’d like to see how far you can push this thing with this cooler. And a big question that’s been bugging me for a while now… can you OC the memory to 6ghz?

        This would be useful for investors, enthusiasts & cryptocoin miners alike…

        • 200380051

          Very interesting indeed. Exciting even. On the subject of 6ghz memory though, i beleive the reference cooler may fare better with regards to cooling. With the NZXT solution, they are barely, if not cooled at all. Some of the chips are actually pretty far away from the fan, on the other side of the GPU and pump assembly, so even adding small sinks is not going to be much help.Here’s a small benchmark of an overclocked and watercooled 290X

  • Gurg

    Anyone questioning the initial total cost outlay for the G10 bracket and separate closed loop cooler needs to realize that this solution should also work on future card purchases as well as eliminating the hassle of RMA or warranties if your fan on a card goes bad. I’ve got a used Kuhler 620 I can use for this bracket.

  • Strider

    Thank you so much for the review. You really can’t go wrong with this bracket, even when adding in the cost of having to buy a supported cooler, it’s still vastly less expensive than going full-on water cooled. The performance is also downright outstanding.

    I was actually shocked to see it kept the other components so cool as well under load. If you’re really worried about the VRAM temps, perhaps there is enough room under the bracket to add some short copper ramsinks to the VRAM modules for additional dissipation?

    I already had my mind set on trying this Kraken G10 out on a 290X, however this review makes the idea even more appealing. Thanks for taking the time to do a proper review. Now perhaps you can see how much it helps when it comes to overclocking the card? =P

    • jpg207

      I’ve seen a video on this set up and the guy managed to put some small heatsinks on the VRAM no problem and it cut the VRAM temps buy 20/30 decrees.

      • Caly Payne

        on my sapphire 7970 dual-x oc w/ boost the vram isn’t an issue as sapphire put a set of screw through vram heatsinks seperate from the gpu heat sink which for non-customized air cooling w/ the sapphire fans and heatsink i dislike but with a g10 would actually be beneficial

  • Paul Margettas

    Only this is hat this is a mess of wires.

  • Tequila_Mckngbrd

    Pretty sweet, I’ve been doing this on my own, but this setup looks a lot better. The only thing I wouldn’t do is increase the voltage on the graphic card, as the fan itself isn’t enough to cool the VRM’s on the higher end graphics cards.

    • Wilbur

      Totally agree. My suggestion would be to manufacture a full-cover block on which the kraken can be mounted. The block could be cheap and easy to make, as there would be no need for internal channels, o-rings, covers and such. Just a thin solid block. This would make it very easy to change a GPU too, without opening your loop.