Originally Posted by Kinaesthetic
Yeah.........no. Not even remotely worth it. Hyper 212 Evo is still cheaper than this heat-sink and performs quite a bit better.
Load numbers normalized to the delta from ambient for each cooler taken from HardwareSecrets.
Seriously? Dude, hardware secrets has had
great articles, but their heatsink reviews are some of the worst on the internet (almost as bad as Frostytech). Just look at the chart you linked. Sorry, the Hyper 212 EVO does NOT beat the Nh-D14 or Havik 140! The thermaltake Frio is NOT better than the Frio OCK!
Seriously, do you even look at what you post?
Oh, check out HardwareSecrets testing methodology too: They take an i5-2500K, and overclock it to a blazing 4ghz, to really make it hot!
Then, they put an insane
1.3v, a full ~0.1v higher than the stock voltage! They use Prime95 25.11
. Wow, seriously, where do you find an outdated version of prime95? We all know that Prime95 FFT lengths was optimized for Intel chips only around v26...
Get this: We use an arithmetic average of all the core readings
. Seriously? That's got to be the most useless way to get temp read-outs. Everyone knows you go by the max core temp for temp readings, not an average temp. Holy useless thermal test batman!
So please. Don't be posting that garbage around here. Some people might not be informed enough to realize that hardwaresecret's thermal reviews are an absolute joke. Don't come back here 2 minutes later and post a frostytech bench (who don't test on a CPU, they use a large, 5"x5" thermal plate that is outputting 150w over a wide area instead of a concentrated area like a CPU die does).
Oh, and maybe you should include the fan noise too. Testing with stock fans on heatsinks is absolutely useless. There's a big difference with the Hyper 212 EVO, which uses a ~2500RPM fan (high speed is considered ~1800rpm, for reference...). Any heatsink with a 2500RPM blademaster, is going to perform way better. You put that piece of crap Blademaster on any other heatsink, and it'll perform amazingly too.
The Hyper 212+ and 212EVO are awful heatsinks, with extremely low quality fans with terrible CFM/dBa and static pressure, ramped up to extreme RPMs so they push a ton of air (and are deafeningly loud, too) so they score well in thermal benchmarks performed by bad sites. The Hyper 212 EVO is simply a 212+ that's been lapped, it's basically the same identical heatsink to the 212+, except that it ships with a different fan in North America (in EU it's the same as 212+). At it's usual sale price of $29, it's a total rip-off, you are way better off buying a $19 212+ and spending $3 on the best ceramique thermal paste like PK-3 or Masscool or Gelid 1.5g.
The problem with testing on a low overclock, as most review sites do (that way a low end heatsink like the hyper 212 EVO or even the stock heatsink can be on the same graph as an H220, ie an h220 is going to be 50C on an overclock that'll make the stock heatsink hit 120C, ie overheat). This presents problems, as the true cooling characteristics of heatsinks come out only on high heat loads.
This is why idle temp tests are so useless. For example, water loops are hotter than air heatsinks on idle and low overclock/stock clock, and furthermore HDT open heatsink designs like the Hyper 212s perform much better on idle/low clocks than standard designs and water. Is a hyper 212 better than a $300 360mm rad closed loop? No, but if you compare idle temps or even a low overclock, you would think so.
So on an overclock of only firstname.lastname@example.org, yea, a HDT cooler like the Hyper 212 EVO is actually going to outperform water cooling, and standard heatsinks. This is why the ordering on that hardwaresecrets bench is so messed up. email@example.com is basically stock (ie 3.9ghz turbo @ 1.21v is stock on an i7!). That's why you get such absurd results like the Hyper 212 EVO beating the NH-D14.
EVO performs about 1-2C better than the 212+, and that's really only because of it's significantly stronger, deafening, crappy fan. HDT design is really cheap, so makes for super cheap heatsinks, and they are awesome for low overclocks because they'll actually outperform standard base heatsinks on low heat loads. But on any significant overclock, the EVO will overheat just the same as any other $20 heatsink. The EVO is a total rip-off at it's usual sale price of $29. It's just a lapped 212+ with a louder, crappier fan.
For future reference, some good heatsink review sites are xbitlabs, vortez, and [h]ardOCP. You really need to see an apples to apples comparison to get a full picture. A Hyper 212+ with the EVO's blademasters on it is going to outperform an H50 with a silent fan, an h50 with 2x Blademasters will perform similarly to an H80. The site might tell you the dBa of the fans, but it doesn't give you an accurate picture of heatsink performance still because heatsinks usually are bundled with crappy fans with awful cfm/dBa.
The OP's deal is pretty good, but not great. Gamer storm is the 'extreme' division of Deepcool, which is the heatsink division of Logisys, who make a lot of high quality computer accessories (isn't that weird? Most people think of them as OEM company that makes cheap office stuff). All of their heatsinks are just rebrands of Alpenfohn, a European brand that makes very high quality heatsinks, except usually with lower quality fans (that are usually blue or green).
A 6-pipe heatsink like this one is going to perform basically the same as any other 6-heatpipe single tower, ie Venomous X, TRUE, Havik, Megahalems, Archons, or 120x25mm closed loops like the h40, h50, h60, kuhler 620, which are all either asetek 510LCs or coolit Ecos, give or take a couple degrees. For $38 you can usually find high end closed loops like new coolit Eco 2's like the H60 refresh/2013 on sale, and sometimes high end heatsinks like the Zalman LQ-320, NH-D14s used, or Zalman dual towers like the CNPS14X or CNPS9900Edited by Belial - 7/14/13 at 7:13pm