Corsair H60 or comparable?

[MongGrel]

Stopped by to read a few things, just stuck the X5680 in here a few days ago and at 130W it is kinda throwing a heat problem at me while OCing.

Have it up to 4.65G messing around, not sure if it would be worth throwing a X61 in here over the NH-D14, I have it pretty customized myself. I'm getting throttling trying to push it higher ATM it looks.

Last CPU I'll be throwing in this old X58 rig, was planning on keeping it awhile still though.

 

[MongGrel]

Hmm, That H-240X does look interesting have to read around.

I could see some interesting possibilities with one of these.

https://www.youtube.com/watch?v=PYKdKVxbnp8#t=514

Probably not worth the small bump though.

 

[BonzaiDuck]

Hmm, That H-240X does look interesting have to read around.

I could see some interesting possibilities with one of these.

https://www.youtube.com/watch?v=PYKdKVxbnp8#t=514

Probably not worth the small bump though.

Depends on whether you already have a cooler for an existing system and you're looking for the step up -- in which case -- probably not worth it.

Those things are bound to cost extra. Uses the Apogee waterblock. Need to look more closely -- it's an option.

But thanks to DrMrLordX on this. I know I'd seen them before, but hadn't looked this closely -- "this closely" still being a "preliminary" assessment.

 

[BonzaiDuck]

Depends on whether you already have a cooler for an existing system and you're looking for the step up -- in which case -- probably not worth it.

Those things are bound to cost extra. Uses the Apogee waterblock. Need to look more closely -- it's an option.

But thanks to DrMrLordX on this. I know I'd seen them before, but hadn't looked this closely -- "this closely" still being a "preliminary" assessment.

Here's a question -- could be another thread, but we're talking about coolers such as the Hydro's -- the H60, the Swiftech 240X.

Right away, if you were going to build a custom-water setup, you'd look at reservoirs in addition to radiators and pumps -- some reservoirs coming ready-made with pump built-in and some with dual-pumps.

A reservoir is another sink. There's nothing much to cool it down, and heat builds up under increased load conditions until it reaches an equilibrium pretty much given by the actual cooling capacity of radiators by size and thickness and even design. But temperature would reach equilibrium slower because of the extra capacity.

With the AiO's there's no reservoir. Therefore there is little in the way of an adjustment delay due to the reservoir's attenuation of changes in temperature. So the larger the reservoir, the longer it would take under extended, continuous load conditions to reach an overall equilibrium, and under normal operating conditions, I surmise that a monitoring-program's log-file and ability to adjust polling interval would show either a reduction in the number of temperature spikes or a reduction in amplitude.

Does this make sense? If a bigger radiator is better, then is a bigger reservoir also better? And if it is, how much does it matter to either cooling or cooling perceptions?

I'm jus' learnin' about this stuff!! I've been reading online bong-cooler projects since maybe 2005.

Ah jus' haven't left the flight simulator and strapped myself into a real cockpit!!

Nooby-noobie-noobee-noob!!

 

[DrMrLordX]

Glad to have been of service by bringing the H240-X back up. Don't forget this thread:

http://forums.anandtech.com/showthread.php?t=2406786

that's another review where the nh-d15 beats several AiOs. But it can't beat the H240-X!

 

[BonzaiDuck]

Glad to have been of service by bringing the H240-X back up. Don't forget this thread:

http://forums.anandtech.com/showthread.php?t=2406786

that's another review where the nh-d15 beats several AiOs. But it can't beat the H240-X!

So . . . an easy trick should be to make the H240-X. . . beat itself. . . .

I just came across this. We can't keep up with the news on this stuff, but people had mentioned the "Macho" before. Look at the bundled accessories in the rightmost picture. I'm using one of the old blue-rubber-ducties" right now from TR. Looks like they opted for some more flexible materials:

http://www.techpowerup.com/206482/thermalright-announces-hr-02-macho-zero-cpu-cooler.html

What's the rating on the Macho against the D14? In that sort of comparison, 50-50 chance the reviewer would've benched it "with and without." So then, we could find out whether the Macho beats the ACX.

Much as I want custom-water with dual pumps and rads, the Swiftech AiO looks so cool it's almost a temptation. Think about it though. Suppose you were able to decrease stock bench tests on the Swiftech by 5 or 10C? How would that compare with mainstream water-cooling -- maybe with single radiator?

Here's another review on the Macho as compared to D15:

http://www.nikktech.com/main/articl...ro-passive-cpu-cooler-review?showall=&start=4

Read the testing methodology. They say on the one hand they use the bundled fans, or they use a standard Noctua fans for bare-bones cooler packages. They say NOTHING about the accessory ducts for the Macho Zero. I'll bet there's a pretty good chance they didn't use the ducting accessories. Only if that's the case, then the Macho Zero would likely beat the D15 by ~2 to 3C . . something.

But it won't match the ACX. So how would the ACX stand up against the Swiftech H20-240X? And how would the latter stand up on itself?

Ah . . . my mistake. The D15 is maybe 3C better in performance than the D14, and if it's just a little more than that, it will match the ACX.

But the ACX is smaller -- as small as we could expect for anything else in a tower-heatpipe.

 

[DrMrLordX]

I have my doubts about that nikktech review. They have the Noctua NH-U14S beating the NH-d14. Something about their methodology is off.

 

[BonzaiDuck]

I have my doubts about that nikktech review. They have the Noctua NH-U14S beating the NH-d14. Something about their methodology is off.

Listen, Doc!! We all have this trouble remembering all those reviews, and the rank-order may be occasionally inconsistent. But -- Yes you can! -- show that a U14S will slightly outperform a D14. So with the D15, which has the same number of pipes but twice the fins, and probably just edges out the U14S "a little."

And I could also wonder if the Macho Zero review did it justice. I DON'T think they used the ducting accessories to test it! You'd have to wonder how, for a company like ThermalRight which made a name for itself in the business, a massive cooler like that would under-perform this blingy-bling EVGA ACX I'm rocking here now. Actually, I even like the TR Macho better for its "appearance," but this ACX cooler has been amazing so far. Ducted or unducted, 5 to 6C better than the D14.

Testing those two on my own, I'd tentatively concluded that ducting only has a linear effect that may actually be identical between coolers. That, too, could have caveats, and the only way to find out would be to test the Macho and its gear. Maybe if the manufacturer sends me an evaluation sample for free. And maybe when pigs fly!

 

[DrMrLordX]

Well, I think that's the first time I've seen the U14S beat the D14, but that's just me. At least in the case of the ACX and D14, you have samples of both so you can test them for your own purposes. Making heads or tails of some review sites' results can be dicey if their methodology is off.

Just because one site or another can show a U14S beating a D14 (for example) does not necessarily mean that they will do so in my hands, or anyone else's for that matter.

I do agree that their failure to use the included ductwork during testing of the Macho is probably a poor practice. It would be difficult to use the ductwork on an open bench platform, which is often the testing setup of choice for sites (which is not always a good idea).

It's important to keep things like that in mind when looking at the D15 beating AiO coolers in tests as well. For some, the D15 is not going to be the superior solution.

 

[BonzaiDuck]

Well, I think that's the first time I've seen the U14S beat the D14, but that's just me. At least in the case of the ACX and D14, you have samples of both so you can test them for your own purposes. Making heads or tails of some review sites' results can be dicey if their methodology is off.

Just because one site or another can show a U14S beating a D14 (for example) does not necessarily mean that they will do so in my hands, or anyone else's for that matter.

I do agree that their failure to use the included ductwork during testing of the Macho is probably a poor practice. It would be difficult to use the ductwork on an open bench platform, which is often the testing setup of choice for sites (which is not always a good idea).

It's important to keep things like that in mind when looking at the D15 beating AiO coolers in tests as well. For some, the D15 is not going to be the superior solution.

There's always "new" information that paints a bigger picture as you also suggest.

I had high hopes for the D15. After Jakob Whatshisname (Dellinger?) gave his interview about the forthcoming D15 release, we waited, we watched like the Three Wise Men. then the reviews started trickling out. I'd be surprised if any of them proved the D15 could best the D14 by more than 4 or 5C.

It's also more useful that they tested the Macho against D14 and others without the duct. At least you know where it really stands on the merits of fin and heatpipe design.

I had looked at this U14S versus D14 several times -- at least a few -- during the earlier part of this year. I could be wrong, but I'm pretty sure it bested the D14 by a couple degrees or so. It has the D15 heatpipes, but only one set of fins. I had considered getting one.

For that, I'm glad I didn't. The Noctuas look "more serious" as heat-exchangers; they are "well-made." I just couldn't have imagined something like the EVGA ACX outperforming all the Noctuas, when it's freakin' three years old. But it does!

 

[BonzaiDuck]

Here's the Frosty Tech Review comparison:

http://www.frostytech.com/articleview.cfm?articleid=2734&page=5

It's in a dead heat nose-to-nose with the D14, and a degree or so behind without an extra fan. And like I said, the D15 proved better than both, but not by that much.

 

[leedreamer189]

How about Nepton 220 or 240?

 

[BonzaiDuck]

How about Nepton 220 or 240?

Two thoughts about that. First, there are certainly comprehensive comparison reviews that provide insight of a rank-order between the Neptons, other AiO's, and the heatpipe coolers.

Second, you have to convince yourself that the review isn't hyping a cooler by omitting competition you know is stronger. Generally, the larger number of coolers in the database of comparison, the more likely the review is reliable.

You can question the testing methodology, and sometimes testers have to depart from a standard approach: for instance, with coolers that don't come with fans, they choose a mainstream fan and use it for the tests.

Let's see what we can find right away on these Nepton models:

http://lanoc.org/review/cooling/6986-cooler-master-nepton-240m?showall=&start=4

http://www.ocaholic.co.uk/modules/smartsection/item.php?page=5&itemid=1480

http://www.kitguru.net/components/henry-butt/cooler-master-nepton-240m-liquid-cpu-cooler-review/6/

None of these have the comprehensive comparison more common with Frosty Tech's reviews; the comparison involves a handful of liquid and air coolers. The first two reviews don't seem to do much to get a testbed with an overclocked profile. Even so, the limited comparison list can be compared with more comprehensive reviews in which some of the coolers appear, and you can infer a more thorough rank-order.

The last review gives us something we like to see: a good 4.7 Ghz overclocked test-bed using a Sandy Bridge E processor. You can see that the Nepton 240 performs better in those tests than the NH-D14 by about 5C degrees. And it is likely the thermal wattage of the overclocked testbed is much higher than 140W.

What am I going to say about that? I can make up all or a decent part of the 5C difference by ducting and a high-CFM fan on the D14. But even without the ducting and powerful fan, it seems to me that the EVGA ACX cooler is in a dead heat (ha! -- a pun) with the Nepton. "Oh!" you say, "the ACX isn't in the review." No, but it was compared to the D14 in a Hardware Secrets review that I've verified and confirmed here on this thread. It is simple logical inference. And if you add the ducting, you get better than Nepton performance.

On the other hand, add more powerful fans with rubber isolators to the Nepton. Other more minor things can be done to it. The biggest challenge with a dual-fan AiO would seem to be whether you can get any performance at all from ducting it. You can double the fans for push pull; you can get more powerful fans like an AP-30 or Noctua iPPC 3,000 RPM; you can use rubber mounts on those fans; you can wrap the fans in Spire rubber.

but it just shows how close the AiO's and heatpipes perform together -- especially if you're willing to do more than the average consumer does with a new appliance. That takes me back to my earlier remarks: these new processors are going to show higher temperatures at the same thermal wattage with the same coolers as opposed to SB processors -- because of the die-shrink. If you aren't comfortable with load temperatures in the '80s and '90s, you can "de-lid." Or something with greater cooling capacity is required.

And the fact still remains. If a single-pump AiO fails, there's no backup. If a fan on a heatpipe cooler fails, it will continue to function. Pumps have a MTBF of some few years or so; the MTBF for a heatpipe is 1 million years. This isn't to say you shouldn't buy AiO's. It's not to say you MUST use custom water cooling. It's just an appraisal of the situation, even with AiO's with good, reliable pumps.

Does that satisfy you?