KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
For our review today we are using an Intel Core i7 990X (engineering sample) processor, with the Asus Rampage III Black Edition motherboard. We have spent the last week retesting the leading coolers on this board.
While we could use an open test bench, we like to try and mirror more ‘realistic’ conditions so we use one of the best chassis currently available on the market – the Lian Li X2000F chassis, which actually needed a little work to accept the radiator.
The Lian Li X2000F is our reference chassis for cooler reviews, but it doesn't have a position for a single 240mm radiator at the rear. Thinking outside the box a little we used the front of the case with a 120mm to 140mm adapter plate, after removing two of the Lian Li 140mm intake fans.
We swapped the Corsair 120mm fans to intake air across the radiator and into the case, rather than expel out the back (as it isn't possible in this chassis). Corsair recommend that the fans are used in the reverse position, and normally we would agree. However, we want to maintain the front to back flow system as shown above.
Intel System:
Processor: Intel Core i7 990x
Motherboard: Asus Rampage III Black Edition X58
Thermal Paste: Noctua NT H1
Power Supply: ADATA 1200W
Chassis: Lian Li X2000F
Memory: Kingston T1 Modules @ 1600mhz (12GB @ 9-9-9-24)
Graphics Card: Asus GTX580
Comparison coolers:
Antec KÜHLER H₂O 920
Antec KÜHLER H₂O 620
Corsair H50
Corsair H70
Coolit ECO A.L.C.
Coolit Vantage A.L.C.
Noctua NH D14
There is no point testing these coolers at reference clock speeds as the results are all too closely matched. We therefore cranked the Intel Core i7 990x to 4.8ghz at 1.49volts.
System validation is available over here.
Above, a quick run of Cinebench 11.5 64 bit at 4.8ghz. A score over 12 points means this is pretty much as good a ‘mainstream' system as you will get for rendering or video editing duties.
The H100 has a button on the top which allows the user to switch between three cooling profiles. These vary the fan speeds according to the coolant temperature. In the image above we have selected the performance setting.
Quiet: fan speed range, 900 – 1300rpm
Balanced: fan speed range, 1300 – 2000rpm
Performance: fan speed range, 1600 – 2600rpm
We use a diode attached to the CPU, as software monitoring is often inaccurate by a couple of degrees Celcius (or more). Room ambient was maintained at 23c throughout testing. We loop Cinebench 11.5 for 30 minutes to mirror real world working conditions under 100% load.
Impressive, to say the least. It had to happen sometime, and the Corsair H100 has ousted the Noctua NH D14 (Austrian Sandwich) from the top position on the leaderboard. At the performance setting it is just over a full degree celcius better, and it matches the D14 at the balanced setting.
Wow I never even knew this was out. wicked !
Now thats what im talking about. I never thought id see the D14 beaten ! good job corsair.
I take it this isn’t asetek design?
No, the D14 is still king in my book. I mean Sure they both come with incompatibility issues but I reckon the D14 has far less than the H100, I mean the D14 fits any case that has a 140mm fan in the back it’s that simple. With this it raises all sorts of problems not present with the D14. Sure the D14 overshadows some RAM but theres good alternatives to those RAM modules also you can remove the heatsinks too.
Not only is it that here the D14 costs £70 roughly but the H100 costs around £100, so more price for a lower performing heatsink which comes with, in my book, more compatibility issues.
Yup, D14 still wins. But I reckon the next generation of readymade water cooling units may be able to match the D14. But right about then Noctua will probably release their next beast which they’re working on. Saw pics of it somewhere not sure where tho.
Actually its 80 on scan, as shown in conclusion , not 100.
I also dont think its as clear cut as that because d14 can block slots and not fit in some instances. They both have quirks, but id much rather have the h100 fitting system and space around cpu area.
I would argue that the D14 is more compatible. i have a D14 and i dont mind the memory slot being covered as I use low profile memory.
A few points I have found however with D14. I recently changed my power supply and I had to remove my motherboard to get the new power supply in.
Why?
The heatsink had blocked access to the CPU power cable port between it and the I.O plate at the back. I cut my hand trying to get the cable out without removing the motherboard and in the end gave up. I swore for about 30 minutes with this.
It is just too damn big.
Sure its great, and I love it, everuone loves it, but it is a damn pain in the ass once you get the system built and need to change components. If you want access to memory, forget it. it requires removal of the heatsink and or board. same with my power supply/
H100 would be my next purchase based on this review however my case wont accept two `120mm fans with a radiator, so i would need a new case. I might just end up with an antec 920, as it looks good and gives all the room free around motherboard.
D14 is awesome, but it is a fit once and forget cooler and I wont live with it anymore
Excellent. Another great product from corsair
They are my favourite company for a reason. No other organisation improve products so fast. Their last xoolers were great too…
This isnt asektek, its coolit designed
The dual rad performs great, but it limits case fitting…..
All the corsair cases support it. which I would expect anyway!
Ordered one today, cant wait.
Thats why I bought a corsair obsidian 800D case.I can fill it with everything I want and
still have room to store my baseball mitt and bat in it if I want too.
Actually, Corsair recommends INTAKE on all their Hydro series coolers (Intake/Push for the H100). Exhaust is not recommended (although that’s what I use on my case).
I think a bench comparison would have been far more effective than a custom setup an a case not designed for a dual 120mm rad. Any case that cannot accommodate a rad like this, by definition, cannot be “one of the best chassis on the market”. At least with a bench test everything is equal (such as air flow). A bench test with results posted as deltas would have been much better. This would have been a great home computer setup, but not so great for benchmark comparisons. I also note you mounted the rad with the reservoir on the top. Although generally this will not cause issues, some enthusiasts argue that, if all the air has not been bled out of the system, putting the res on the top may cause the air (if any) to recirculate through the system. However, many builds have mounted the res at the top without any problems while others have said they heard gurgling noises that went away after changing the orientation. (FYI).
I just bought H100 as my H70 was terrible. i7 950 @4ghz, gtx580, 700D case, with H70 idle temps were 51c on hottest core in a room temp of 25c and always went up to 90c on blend prime and high 70’s to 80’s during heavy gaming sessions (BF3)
Bought H100 and have 2 corsair fans on top and 2 Apache vipers underneath in pull push (intake) and a 140 viper on the back of the case, along with a 200 fan lodged on the inside facing the back panel to circulate air. My temps now on the low setting idle in a 25c room is on hottest core 46c. Highest fan setting on the H100 pump temps go down to 39 again in 25c room. During prime on low fan speed highest temp I got was 73, highest fan speed during prime – hottest core fluctuating between 61c and 62c again in a 25c room.
I absolutely love this H100.