As seen in our previous Small Liquid versus Big Air comparisons, we wanted to include a high-performance air cooler to show what, if any, differences there are between the two technologies. Small liquid lost before, though this time we’re also considering noise, fit, and alternative radiator dimensions.

We picked Evga’s SuperClock CPU Cooler for its single 120 mm fan—which makes it a direct competitor to Antec’s Kühler H₂O 620—as well as its moderate price and good performance reputation.

Evga supports both AMD and Intel processor interfaces using the same installation plate. The firm has not added LGA 2011 mounting hardware yet.

Holes in the SuperClock CPU Cooler’s socket support plate fit most AMD and Intel processors. Thumb nuts make the included bolts stand up, while fiber washers insulate the motherboard from contact with the nuts. Evga even specifies adhesive film for the fiber washers to stick the plate in place, though we left the wax paper on this adhesive to ease removal.
The factory-installed mounting bracket supports everything from AMD’s old Socket 939 all the way up to Intel’s LGA 2011, though that'll require screws that Evga has not yet added.

The SuperClock CPU Cooler’s heat pipe-direct design reduces latent heat and weight. Several of our previous air-cooling comparisons have dispelled wild claims about this design’s performance advantages, though anything that can cut out weight without hurting performance is a benefit to us.

Spring-loaded nuts hold the SuperClock CPU Cooler tightly against the CPU.
I completely agree. Pretty much the only reason to go with these sealed water coolers is to say that you water cooled your PC.
I completely agree. Pretty much the only reason to go with these sealed water coolers is to say that you water cooled your PC.
I agree 100%. And I do, in fact have an H70. And when I'm at work describing the epic beast of computers I build, and I show off some pics of mine; I always point to the H70 and go "Theres the radiator for the water cooler" /gasp!
I think that's a perfectly good reason to buy any product. I like that I have a choice. But Noctua's tower coolers look pretty sexy too.
Remember these are easy to install closed loop water colling setups too guys.
They are not half inch or bigger professional units designed to keep your massively overclocked sub-critical mass uber rig cool as ice and quiet too.
You need a decent pump, a large or two smaller radiators, a tank, and a lot more hose ... and some more time.
http://www.swiftech.com/
Corsair, Asetech, Antec, Thermaltake ... these are all low end units ... not enthusiast's water cooling gear like Swiftech.
Good value for money though ... I am not knocking them.
Once again you do a comparison against an inferior air cooler, so your conclusion does not hold much weight with me.
Why do you not include the Noctua NH-D14 or a Thermalright Silver Arrow, both which cost about the same as your WC loops (or a bit less) yet are delivering better performance (and noise levels) with the Antec H100 at max speed as the exception. The H100 is comparable to the air coolers.
Oh I'd rather have a Noctua NH-D14 with a 3rd NH-12 fan, but I traded an old crappy PSU for this H70 brand new. So I takes what I can gets!
Furthermore you say "Once again..." even though the air cooler won the last time this editor wrote this type of comparison. Look it up, "Small water versus Big Air", I'm sure Google can help.
Furthermore you say "Once again..." even though the air cooler won the last time this editor wrote this type of comparison. Look it up, "Small water versus Big Air", I'm sure Google can help.
I read your last article as well, like I do most articles on the site.
The problem I have with this article is your intro where you mention noise and cooling performance, yet you chose an air cooler that is cheaper than the WC loops you are testing.
Had you chosen a comparable air cooler price wise (the Ncotua NH-D14 or Thermalright Silver Arrow), all of the WC loops would have performed worse than the air cooler, the H100 would have been the exception, but it would have to run at max speed, thus negating the noise level performance.