Corsair H60 or comparable?

[thejohnfist]

Hello all! Let's start off with the facts necessary here. I'm currently running an Intel i7 2600k on the OEM cooler, yeah I know...

That being said, it's slowly giving up it's ability to keep the CPU below 70C on factory clock speeds. I'd like to get something cooler, and potentially quieter, with a little room for any future light overclocking.

I've been eyeing the H60, but I've never done liquid before, and am not entirely sure where to get good info on the matter.

Again, ideally looking for small, liquid, and quiet. Help!

 

[lehtv]

In what circumstances is it exceeding 70°C? If it's idling that high, then it's not properly seated and it should overheat and crash in the slightest workloads, and you need to reseat it. If you're stress testing it, then 70°C is nothing, you can exceed 80°C at load and it'll still be fine for an Intel CPU. 70°C while gaming seems typical to me, nothing unusual.

In any case though, upgrading the stock cooler to a quieter one makes sense. For light overclocking, you can't go wrong with Hyper 212 EVO $35, although a Thermalright True Spirit 140 $55 or a Thermalright HR-02 Macho $53 will perform better while staying quieter. The 212 EVO makes sense if you aren't THAT picky about noise levels and temperatures, it's still a substantial improvement over the stock cooler. Whichever you get, make sure the cooler fits in your case.

I don't recommend liquid coolers at all, they have pump noise and typically high RPM fans. At every price point, there's a better air cooler available.

 

[tarmc]

the h60 and 75 are good options also kraken x41? is great as well

 

[thejohnfist]

I don't recommend liquid coolers at all, they have pump noise and typically high RPM fans. At every price point, there's a better air cooler available.

Whatever software I have that comes with my board (AI Suite I think? Asus software.) likes to alert me that my processor is reaching 70-73C while gaming.

If liquid isn't going to be at least as quiet as the stock fan then I may look into some passive cooling options, but I only use a mid tower so I'm sure options are limited.

 

[BonzaiDuck]

Whatever software I have that comes with my board (AI Suite I think? Asus software.) likes to alert me that my processor is reaching 70-73C while gaming.

If liquid isn't going to be at least as quiet as the stock fan then I may look into some passive cooling options, but I only use a mid tower so I'm sure options are limited.

Seems I post a lot to these forums, and offer opinions which seem to be atrophying by my own observation. I've had my 2600K system for 3+years, and followed a strategy of "custom" air-cooling going back to 2004.

The reason I would have avoided picking one-fan-and-radiator coolers like the H60, though, is the higher risk of a single pump failing. At least with thoughtful fan-deployment, the failure of any single fan isn't going to sufficiently cripple a tower-heatpipe cooler or some variant thereof. At least with the H60, there'd be a decent chance of cooling on par with a Hyper 212, but how far or how far short, I can't say.

So I have an NH-D14 in my current Sandy rig. There is only one (non-Noctua) fan hanging in the middle of it, but the case exhaust fan is potentially powerful and ducted to the D14's rear.

Now I've recently discovered that there was actually a heatpipe tower that benchmarked with thermal results better than the D14, released several months after the latter. It had been called the EVGA "SuperClock," but the maker renamed it the ACX cooler. You can get it at the Egg for maybe less than $55.

Frosty Tech and Hardware Secrets reviewed it on a controlled thermal platform. In that review, the cooler bested the STOCK D14 (and Noctua fans) by 6C additional reduction under load conditions.

I've yet to test it myself. And I would usually consult multiple reviews before picking a cooler, but I trust this review that I'd seen. It's modestly bigger than the 212/+/EVO -- more like a Megahalems, but the same shape -- comes with one fan (which I'd likely also replace, as is my habit.)

I'm waiting for parts to replace another system which is LGA_775. I'll still be using an 1155 board and processor. Will be able to say more about the EVGA next week.

At the bottom of it, I'm pretty sure by comparison of tests with another forum member that my ducted D14 is neck-and-neck with an H110i. [Sounds crazy, I know! But I have no reason to delude myself.]

If the ACX proves out, it means that heatpipes are still viable as options over AiO's. There may be another round of competitive market entries: better, bigger AiOs, and less likely significant improvements to heatpipe designs. I think we're approaching a limit for the latter.

But for a 2600K? You'd be good to go with an ACX even if my results next week are mildly disappointing.

 

[XavierMace]

Sorry, this post is a bit of a book.

I don't recommend liquid coolers at all, they have pump noise and typically high RPM fans. At every price point, there's a better air cooler available.

I don't know what kind of pumps you've used, but I can't hear the pumps on my X61 and X41 over the 3 SP140 Quiet Edition fans at anything above 40% speed on the fans. All of the above are completely drowned out by the GPU fans with every GPU I've owned (GTX460, GTX465, GTX670, R9 290X). That's why I went liquid on the GPU too. My tower sits on my desk at elbows length from me and you're hard pressed to hear it over ambient noise. Guru3D's review of the H110 shows basically identical sound levels to the 212 EVO. Technically the H110 was 1db quieter under load but 12c cooler.

The reason I would have avoided picking one-fan-and-radiator coolers like the H60, though, is the higher risk of a single pump failing. At least with thoughtful fan-deployment, the failure of any single fan isn't going to sufficiently cripple a tower-heatpipe cooler or some variant thereof. At least with the H60, there'd be a decent chance of cooling on par with a Hyper 212, but how far or how far short, I can't say.

I've now personally owned 6 single pump boxed setups, 1 custom and 5 AIO, (and installed 3 times that many) and the only pump failure I had was on Thermaltake's first POS all in one system back in like 2000. Stick with an Asetek based system and don't give it a second thought. My Corsair H100 and both of the above mentioned Kracken's are on 24x7 99% of the time. The H100 has been running since Dec 2011. Do you install a secondary pump and radiator in your car? I certainly don't and I don't drive around worrying about it.

The X61 + X41 combo keeps my 3770K (@4.6Ghz) under 40c and the R9 290X under 50c while gaming. Low 30's idle. Just a push setup on the fans. I didn't have to duct my case, or play musical chairs with the fans until I found a good setup.

10 years ago, I would have agreed with you guys. That Thermaltake put a sour taste in my mouth about Liquid Cooling at the time. But AIO Liquid Cooling has come a VERY long way. I'm not arguing you can't make very effective air setups. But when you're trying to do a high end air setup you usually start worrying about where the fans mount, what fans to use, ducting, etc, etc. The AIO you pretty much just mount and forget about it.

OP: If these guys don't talk you out of going Liquid, here's my thoughts on the matter.

If you look at MOST of the AIO liquid cooling setups, they are mostly rebranded Asetek systems. Each "manufacturer" slaps their own logo on it and gives you some different accessory options. Since they are all typically using the same pumps (in each "tier", for example H60 vs X40), from a reliability standpoint they are all going to be the same. It's just a matter of accessories/features. Please note however that the different "tiers" sometimes have different pumps. For example the NZXT X40 and X41. In addition to the X41 being 140mm instead of 120mm, the X41 has better pump and a 6 year warranty vs 2 year (Corsair is 5 years across the board). Just something to consider. I highly recommend the NZXT's for the following reasons:

1) Tubing. The tubing on the Corsair H100 is very inflexible and is too short for many setups. Regarding the length, I get that you don't want a bunch of extra tubing to deal with, but it pretty much requires you to have a case where the radiator is mounted directly next to the motherboard. Regarding the inflexibility, it just further compounds the length issue and makes mounting a bit of a PITA because you're fighting this tubing when trying to get everything situated. However I do believe the flexibility issue was fixed with the H100i/H110. They still have shorter tubing though..

2) Cabling. The H110 has you plug the pump into the CPU Fan port then the radiator fans onto the mobo. What if you're running a MATX or ITX board? You may not have enough fan headers. The H100i has you plug the fans into the CPU block. Frequently this results in fan cables that can't reach the pump (I'll get to why that's important in a minute). The Kraken's (both the X61 and X41, possibly their whole product line) have a braided (Corsairs cables are not) cable coming off the block for the fans. Single slot models have 2 fan plugs, Dual slot have 4 fan plugs. The use of the braided cable reduces the cabling going to the motherboard (so looks cleaner) and gives you enough extra length to prevent cable reach from being an issue in most setups.

3) Management. Both companies have software to manage your cooling. Corsair has "Link", NZXT has "CAM" both have Pro's and Con's. They both provide software fan control offering preset settings or full manual control. This is great and allows you to tweak fan speeds as needed. For example in my environment, I keep them set to 60% most of the time. This makes the fans nearly inaudable but still provide sufficient cooling. The X61 has plugs for 4 fans, so this allows me to manage all 3 radiator fans, plus the one lone case fan from the same place and without cables going to the motherboard. Corsair offers other products (lighting and select power supplies) that also use "Link" which gives you access/control over those aspects. So, if you have/want those products, that's a Pro for them. NZXT's CAM doesn't provide those options. However it provides 2 features that "Link" does not (to the best of my knowledge). CAM shows pump speed and liquid temp (rather than just component temps). Liquid temp is a key aspect of liquid cooling and a great troubleshooting tool. The X61 has an LED in the pump and you can use the software to link the LED color to the liquid temp. If your component is getting hot but the liquid isn't, you have a mounting problem. If your liquid is getting hot, you need more airflow over the radiator. So that's a Pro for NZXT. They also offer a mobile app for your phone that gives you the same features, however I don't have experience with that. Now the cons... The H100i uses Link. It's larger brother, the H110 does not. So if you want all the features, you have to go with a smaller radiator/fan setup. For both noise and cooling, bigger is better. It's an odd decision by Corsair. You can use CAM with the cheaper offerings from NZXT, some features may just be disabled (namely the pump lighting). The downside though is that CAM is a bit of a resource hog when you have it up and running on the screen. It drops down significantly when you minimize it. Hopefully they improve that at some point.

I've been liquid cooled on my rigs for 5 years straight at this point, I wouldn't go back.

 

[tarmc]

i do have to agree, the pump on my h70 made a bit of noise, and the stock fans with alot of aio coolers arent great. but i replaced the fans on the h70 i have with noctua (nf-p12)
and its pretty quiet and does a great job cooling the i7 920 its on right now.
main rig has a h110 with 2 bitfenix fans on a fan controller and its very quiet
what case are you using? is space limited?

 

[BonzaiDuck]

I had mentioned this EVGA ACX heatpipe tower -- once called "EVGA Superclock." There are some caveats about my reading of a Hardware Secrets review. The comparison with the NH-D14 differs because the single Superclock/ACX system has one stock fan; the D14 had two. The test speed on the D14 is about 1,250 RPM; the ACX fan is more like 2,200.

How this plays out in severe load-test performance -- for me -- depends on 120mm airflow between 2,200 and 3,600 RPM. For someone else, who doesn't "do stuff" like I do, you don't want to use such a fan.

Thanks also to XavierMace for YOUR response. Important to dispel myths about the AiOs.

There's no doubt that the AiO's give near-maximum possible cooling as opposed to heatpipe coolers, but they're not rated under test conditions corresponding to "stuff I do" that others don't or won't. Likely, someone who orders an AiO won't seek any further "enhancements" other than doubling the fans for push-pull.

generally, I've been able to match an H110 performance (using a single pair of fans) with my D14's with an AP-30 and a rectangle of foam-board with layers of Spire foam-rubber. The comparison involved Prime95, near-same processors and same load-voltage.

So therefore, my D14 would still outperform a [puny :biggrin: ] H60 or H70 whatever. ] That doesn't make them bad choices; it just means that they're probably good alternatives to a lot of better-than-average heatpipe coolers, like the 212-EVO.

The proof of the pudding is in room-ambient, clock-speed, voltage and the average-maximums in Celsius of the four Tj sensors under high-load conditions.

I think you'll see a lot of ~240mm radiatiors with two fans which show maybe 6 or 7 Celsius better than the D14 under the usual test-lab conditions, but I've closed that gap. And you'd be surprised at how quiet it is with that Gentle Typhoon. There's no pump-noise, either.

My thinking about this will not match that of others. But if I go with air-cooling -- which I do -- I make improvements. And if I'm going to do water-cooling (which I will by mid-2015), it has to be custom parts.

 

[AnitaPeterson]

I'm using a Corsair H60 to cool my FX-6300, and I haven't regretted my decision - not for a second! Excellent performance, quiet, exactly what I needed. I'm really pleased with that machine.

 

[XavierMace]

Bonzai, don't get me wrong, I know the performance of those things. I've got a pair of D14's in my storage system cooling a pair of Xeon E5645's. Noctua makes very good components and if I had a case that didn't have room for a radiator, that would be my recommendation as well. Of course those things have significant space concerns as well, just in a different area.

However, the fans on those are typically running at ~1500rpm. Right now, my fans are at 750rpm.

I would love to see some actual numbers on your setup though. Again, I'm not questioning their performance. Just curious what kind of numbers you are seeing. I would also mention the fans Corsair includes with the H110 aren't the greatest. Replacing them helps a lot.

Regarding pump noise, I don't understand why that keeps getting brought up. If I take the side of my case off and put my ear next to the pump, yes I can hear a low hum when the pump is operating at high speed. With the side of the case on, and the 3 140mm fans at 750rpm I can't hear it at all over the fans. Am I missing something here? Like I said, I've done quite a few builds and never found the pump noise to be loud enough to even notice unless I was specifically listening for it. If I had a good mic, I'd take a recording.

 

[lehtv]

Guru3D's review of the H110 shows basically identical sound levels to the 212 EVO. Technically the H110 was 1db quieter under load but 12c cooler.

And $80 more expensive. What was your point again?

For less cash than the H110, a D15 will perform the same while making less noise. Check techpowerup's review

 

[XavierMace]

And $80 more expensive. What was your point again?

For less cash than the H110, a D15 will perform the same while making less noise. Check techpowerup's review

The whole point in upgrading your cooling is to further cool your processor. You don't see a point to 12c cooler under load?

In regards to techpowerup's review, they're doing something wrong to get temps that high on the AIO setups. For example:

http://www.hardocp.com/article/2014...1_aio_liquid_cpu_cooler_review/3#.VHA-k4vF9zt

That shows the X61 (which should be nearly identical to the H110) 8c cooler than the D15 and that's with a very poor radiator setup. Again, I completely agree the D15 is a kickass air cooler. I just don't see it matching a good AIO in most situations. For reference, this is what I get out of my 3770K @ 4.2 with an X61 and the fans at 50% (850rpm):

My numbers simply don't match up with theirs. In fact I can't it really doesn't seem to line up with their other benchmarks of it:

http://www.techpowerup.com/reviews/Corsair/H110/6.html

Their review of the NH-D15 is using a 4770K (84W TDP). Their review of the H110 is using a 3960X (130W TDP) yet the H110 is benching 20c cooler on the second test. That doesn't add up. However one thing rarely mentioned is ambient temp. Here's one that does:

http://www.hardwaresecrets.com/article/Corsair-H110-CPU-Cooler-Review/1736/6

That shows the H110 taking the win and the X61 has better fans. But since noise keeps getting brought up....

http://www.silentpcreview.com/article1408-page6.html

That's the only review I see with both the D15 and X61. TLDR: At the same noise levels the X61 is slightly cooler with the stock fans. With better fans, the gap widens. Even at max speed the pump only makes 19db which is quieter than the Noctua fans.

 

[BonzaiDuck]

Bonzai, don't get me wrong, I know the performance of those things. I've got a pair of D14's in my storage system cooling a pair of Xeon E5645's. Noctua makes very good components and if I had a case that didn't have room for a radiator, that would be my recommendation as well. Of course those things have significant space concerns as well, just in a different area.

However, the fans on those are typically running at ~1500rpm. Right now, my fans are at 750rpm.

I would love to see some actual numbers on your setup though. Again, I'm not questioning their performance. Just curious what kind of numbers you are seeing. I would also mention the fans Corsair includes with the H110 aren't the greatest. Replacing them helps a lot.

Regarding pump noise, I don't understand why that keeps getting brought up. If I take the side of my case off and put my ear next to the pump, yes I can hear a low hum when the pump is operating at high speed. With the side of the case on, and the 3 140mm fans at 750rpm I can't hear it at all over the fans. Am I missing something here? Like I said, I've done quite a few builds and never found the pump noise to be loud enough to even notice unless I was specifically listening for it. If I had a good mic, I'd take a recording.

There is a subset of all the things I had to do to make the D14 match temperatures of an H110 installed as it would have been in a benchmark review, or using the boxed instructions and parts with no special "mods" if any are possible. Those things might improve the AiO by a few more degrees, and then "the tie is broken."

Another member had a well-done setup using the AiO for the CPU and additional custom parts (I think) for the graphics cards. The H110i exhausted through the top. I don't think he replaced the fans -- as I would do immediately with the D14.

You can install the Noctua cooler "as is" with the bundled fans, and it represents an accommodation by Noctua to a "noiseless" market. For me to reach that standard of noiselessness, it took me another day or so applying 4-ply pieces of Spire Acoustic foam-rubber on a part that I designed and created -- which, push come to shove -- could be discarded without affecting anything but cooling in the computer. Anything, that is, except the noise from my AP-30 exhaust fan. I need the part with the Spire rubber to staunch that noise at its source and quell any "echoes."

In correspondence I had with Noctua tech-support, they fully realized that you could squeeze more C from the bargain with more CFM and other modifications. Face it, though. People have to be willing to do all that . . . DIY stuff. I just went ahead and did it.

 

[essential]

I have a kraken x41 on my i7-4790k and it keeps it in the high 20's low 30's idle, and that's on it's "quiet" profile setting, which really is quiet. I'm very happy with it so far.

 

[BonzaiDuck]

I have a kraken x41 on my i7-4790k and it keeps it in the high 20's low 30's idle, and that's on it's "quiet" profile setting, which really is quiet. I'm very happy with it so far.

For XavierMace: Maybe someone could formulate a "Theory of CPU cooler reviews." The Frosty Tech summary for what could easily be 150 different coolers with tests at 150W and 85W shows the X60 besting the D14 by 3C degrees. I think I could best the X60 in my D14 by as much as 5 or 6C, but I'd have to do "a lot of stuff" to scrape up a C here and a C there.

I may have the screenies, but too lazy now to search for them.

With a room-ambient of 80F (Fahrenheit), LinX "affinitized" or IBT (all eight threads) would show an average-of-maximums on HWMOnitor to be either 73C @ 4.6Ghz or 75C @ 4.7. Voltage @ 4.6 under loaded droop was 1.35, and around 1.37 for 4.7.

Prime95 sFFT gives about 68.5C at that room ambient. The duct-box and the AP-30 limited to a top speed of 3,600 RPM (top-end: 4,200) was worth a 5C drop without the ducting using a less powerful fan. Further, it became apparent that the fan could be capped at 3,200 to the same effect; any increase in the "average-of-maximums" was unmeasureable.

After a half hour or so, those were the temperatures I got with the rather balmy room ambient.

What else do I do? I pressurize the case with low-rpm 200mm fans; I lapped and ground off all the nickel on IHS and HSF base; and I used Diamond paste. Not everyone wants to do that, and you can do some of it with an AiO -- which, as I say, would be a tie-breaker.

For Essential: The Kraken X40 falls short of the D14 by just under 1C difference -- or warmer. The Kraken X60 bests the D14 by about 3C. These are all the 150W tests -- the rank order is about the same with the 85W tests. But I've already got the X60 beat, and probably then some. PROVIDED!! That the X60 user isn't making little custom mods to make the X60 more effective! But just replacing the fans on some of these AiOs would have a positive effect.

Here, we're now dealing with acoustic issues. But I also have that licked, too, and I think it could be done with a day's work on an AiO.

Folks! Here's my thinking!! If you somehow "find" some . . . DIY solution . . that nets a temperature benefit, you may likely try to stick with it. But the string is going to run out.

The AiO folks are likely going to introduce models that marginally perform better, hopefully aren't any bigger and fit in the same population of computer cases. If I were going over to Morongo Casino and they had a game wherein I could bet on the future of heatpipes versus the future of AiO's -- I'd probably bet on the latter.

And the other thing. NO-body really "wants" to take such time and attention to squeeze another 7 to 10C out of an air-cooler. I just chose to do it.

I'm just faced with a mild dilemma like the commuter on the train: Should I buy a ticket to Podunk and transfer to the train to Kansas City? Or should I just buy a ticket on the train to KC?

I'm planning to water-cool -- really -- I AM!! But I think I just want to go "custom" in the parts and installation aspects. Maybe I could squeeze another 5 to 10C out of that choice over AiO. Maybe more. For the moment, I'll have more trouble fitting an AiO that's acceptable to me than an interim heatpipe cooler.

It's definitely a whole new ballgame, whichever path one chooses. Then, again, who can say what sort of contraption will be released in the next six months? Something could pop into commercial view that would make my Morongo bet a loser . . . And then it would all change again, depending on any new AiO's we might see.

 

[lehtv]

The whole point in upgrading your cooling is to further cool your processor. You don't see a point to 12c cooler under load?

My original point was that at each price point where you have both an air cooler and a water cooler available, the air cooler will perform better in terms of the balance of temperatures and noise levels. Since H110 is at a completely different price point than 212 EVO, it has absolutely nothing to do with what I said, hence "what was your point again?".

H110 is also more expensive than D15, so it's to be expected that in some reviews you'll find it performing better. TPU's review may or may not be representative of the general consensus, it was just the first one I clicked into.

 

[BonzaiDuck]

My original point was that at each price point where you have both an air cooler and a water cooler available, the air cooler will perform better in terms of the balance of temperatures and noise levels. Since H110 is at a completely different price point than 212 EVO, it has absolutely nothing to do with what I said, hence "what was your point again?".

H110 is also more expensive than D15, so it's to be expected that in some reviews you'll find it performing better. TPU's review may or may not be representative of the general consensus, it was just the first one I clicked into.

With water, you have more room to compromise on parts so they fit in your case, or within your budget. Acceptable performance is => any air cooler. There are extreme solutions -- bong and Volenti coolers, or unusually large external radiators like the MO RA3. So there's more room for compromise, and a lower acceptable temperature range.

More options: Radiator thickness and size, choice of waterblocks. Special fittings, single or double-loop.

With air, you likely don't -- or shouldn't -- have a problem "fitting the parts to your case." But the efficiency of an air-cooler depends on many things, and some of them add up to size. You would know for sure that it would fit in your case unless you had some . . . . stubby . . "mini-midi-midtower" that limited your choices.

There might be ways to reduce noise with multi-fan radiators; there are ways to increase CFM through either a heatpipe cooler or a radiator.

If you overclock, your desire for benchmarks and wringing blood out of a stone would long-ago pushed you into water-cooling or even phase-change. If you want to adhere to a specification or extrapolated spec limit on your chief cause of heat (VCORE), you may find some very good overclocks for 24/7 that can show manageable temperature under enhanced, improved or "custom" air-cooling.

But with air-cooling this way, every little grain of rice counts. I've heard folks recently voice aversion to lapping their Haswell-E or 4790K integrated heat spreaders. People somehow prefer greasy, easy-to-apply TIM formulas. Or you can choose liquid metal or nano-diamond. If you can keep up with AiO alternatives, it can also be made quiet despite the decibel rating that never reaches the user's ear.

 

[XavierMace]

With air, you likely don't -- or shouldn't -- have a problem "fitting the parts to your case." But the efficiency of an air-cooler depends on many things, and some of them add up to size. You would know for sure that it would fit in your case unless you had some . . . . stubby . . "mini-midi-midtower" that limited your choices.

I wasn't going to go there as it's somewhat besides the point, but I've had more fitment/size issues with my Noctua's than any of my AIO setups. I've dealt with more than one full size tower they simply won't fit in. They barely fit in my 4U server cases as well.

To all: I'm just going to conclude with this since we clearly aren't going anywhere here. Yes, with work, an appropriate case, and testing you can probably get a good air setup to match a stock H110/X61. But there's nothing stopping you from upgrading the AIO setup with better fans. I have no way of testing sound output although I've already linked reviews that do and show the X61 to be no louder than the air setup even with stock fans. But I've already posted multiple reviews showing those air coolers aren't matching the good AIO's for cooling.

Lehtv makes mention he found one link they shows they are equal and those results don't even match up with that sites own other tests of the AIO's. And if cooling isn't your ultimate goal, why are you spending $100 on cooling in the first place? They make passive heatsinks for a reason and they cost a fraction of that.

I've also posted screenshots of my temps as well as multiple sites showing the air isn't matching liquid. At the risk of being rude lehtv, feel free to supply some actual data to support your opinion. I've given you multiple links showing your wrong, you come back with the equivalent of "no I'm not, because I said so". Plus you're stuck on the H110 and I've already explained why I would go with the X61 over the H110.

 

[BonzaiDuck]

Yes, with work, an appropriate case, and testing you can probably get a good air setup to match a stock H110/X61.

A lot of work, only shortened by previous experience. You can save yourself the trouble by getting an AiO -- just to break even.

But there's nothing stopping you from upgrading the AIO setup with better fans.

True. But it's more than that. You're going to find better airflow with more powerful fans. Some of the good fans can run to 2,200 RPM with little noise. But the radiators amplify any sound at all. So you'd pay particular attention to fittings.

Early this year, I had a discussion with WoodButcher, who drops in here occasionally, and we speculated about using AP-30's for a radiator that would accommodate two or even three. There might also be some other tricks.

I guess I'd rather have a rig with a bay-reservoir and two D5 pumps -- one inflow and one outflow.

Of course, coolers like the D15 takes up a lot of space-in-da-case. But there's a lot more challenge, I think, to fit two radiators in a midtower case.

And also, as I said, more products are likely to be introduced in the future. Larger AiOs -- certainly.



I have no way of testing sound output although I've already linked reviews that do and show the X61 to be no louder than the air setup even with stock fans. But I've already posted multiple reviews showing those air coolers aren't matching the good AIO's for cooling.

Lehtv makes mention he found one link they shows they are equal and those results don't even match up with that sites own other tests of the AIO's. And if cooling isn't your ultimate goal, why are you spending $100 on cooling in the first place? They make passive heatsinks for a reason and they cost a fraction of that.

I've also posted screenshots of my temps as well as multiple sites showing the air isn't matching liquid. At the risk of being rude lehtv, feel free to supply some actual data to support your opinion. I've given you multiple links showing your wrong, you come back with the equivalent of "no I'm not, because I said so". Plus you're stuck on the H110 and I've already explained why I would go with the X61 over the H110.[/QUOTE]

 

[BonzaiDuck]

I tried to record sound on my cellphone at 3' and 6' from my sig system, but the faint hum behind the air-turbulent white-noise seems unrecordable. Point about this: If you can muffle the noise at the source leaving only the white-noise, the system is as quiet gaming full-bore as it is at idle.

I have some screenies on my Z68 system-- I'd have to scour through folders and sort through them. Also -- 30 pages of overclocking notes.

But you'd still want screenies for the H110i system, same or near-same processor, same load voltage, same speed or no more than 100Mhz apart at same voltage.

I PM'd a member who's still happier than a pig in s*** with his Sandy Bridge system, and we exchanged notes. He had originally said "doesn't exceed 50C." I inquired some more, and he gave me his temperatures for the Prime95 test. So that was the basis of my comparison. "~68C @ 1.35V and 4.8" versus my 68.5C @ 1.37V and 4.7Ghz. I had the 2600K, he had the 2700K; He lives in Montreal; I live in So Cal and my room-ambient for that bench was 80F.

I don't "need" to make claims about my air-cooler. I could also find benchies I took which showed reductions in load temperatures for "no ducting" versus ducting. That's worth about 5C at 140W thermal power. Diamond paste is worth 2C. Lapping off all the nickel plate to flat copper is worth at least 2C. Pressurizing the case and forcing all the air therein to go through the CPU cooler is worth something.

I wouldn't be using these coolers WITH THESE PROCESSORS if I hadn't recorded the data for myself -- screenies or not. This isn't a competition between AiO and Heatpipes. I am saying that the AiO cooler-makers seem to produce a product that just stays ahead of better heatpipes enough to justify the purchase. If you want better now, you go to custom water parts. And part of the problem: the AiO makers know full well if their product fits in fewer cases than before, it's not good for their bottom-line. So that's another limitation on AiO capability.

Finally, what the AiO makers DON'T do is stay ahead of the surfers who use "special wax" on their boards. Most people don't want to fiddle with it.

More will follow on this, probably in a thread about how "wonderful is the CM Hyper 212 EVO." I was going to test an old one I have with my duct and fan setup -- with and without. But since I don't like replacing coolers on a system still "under construction," I got curious about another air cooler purported to best the D14 by 6C. Did I make the mistake to say any of those air coolers was "even" with a KRaken X60? That would be wrong, because they would lag behind with default installation IN THE REVIEWS. I can't make claims like that, and I must have been thinking of something else.

And just a parting thought. Water-cool enthusiasts like to show the complexity of their rigs. The laws of Physics say that properly configured water-cooling is more efficient. That's a given. the showcase projects have dual loops, plenty of LED-light, custom parts, chrome fittings. If it's "custom," it's a DIY project. So? I've done some DIY projects with my air-coolers.

 

[BonzaiDuck]

All praise to XavierMace for post #6. Yes -- this is a bit of a book. [This is a footnote.]

 

[DrMrLordX]

I'm surprised nobody's mentioned the 220-x or 240-x. Is it the price? You can't beat the 240-x with any air cooler.

I still prefer air for numerous reasons, most of which are more subjective than objective. There's something to be said for just strapping the radiator to the heat source and cutting out the water loop. Sadly, heatpipes have some interesting limitations, which I do not yet fully understand or appreciate. But it seems that they have a maximum heat load. Water loops have enormous heat capacity, so water is obviously better for high heat flux.

 

[BonzaiDuck]

I'm surprised nobody's mentioned the 220-x or 240-x. Is it the price? You can't beat the 240-x with any air cooler.

I still prefer air for numerous reasons, most of which are more subjective than objective. There's something to be said for just strapping the radiator to the heat source and cutting out the water loop. Sadly, heatpipes have some interesting limitations, which I do not yet fully understand or appreciate. But it seems that they have a maximum heat load. Water loops have enormous heat capacity, so water is obviously better for high heat flux.

Can you link some "showcase" or reseller pictures for this "220-x" or "240-X" cooler? I don't know which manufacture to look for, and a search on your model-name abbreviations as-is turn up nothing.

Lord! The writing is on the wall!! The die-shrink and lithography mean less effective cooling of any kind because of the reduced area of processor contact. the only thing to do is to use some sort of water-cooling, and any "hybrid-air" enhancements such as a Sabertooth-like mobo duct would improve things, but not for the processor.

And if we thought we no longer had to fret over TIM choices, we were wrong. Even for the simple physics of water-cooled effectiveness and efficiency, a delidded Haswell would have more promise. At least with the E cores, there's nothing to do about it, nor expected shortcomings of "new, improved polymer TIM." But you'd still want to pick a TIM and make "processor preparations" (like lapping) to get every last grain of rice -- each Celsius degree improvement.

 

[xthetenth]

Swiftech makes the H-220X and H-240X. It's a very good looking product in all regards. If I were trying to disperse more heat I'd probably have one in my case this second, but I don't really feel I need something with pump noise to make when most of the time my fans barely spin up when I game (and low noise is my real goal).

 

[DrMrLordX]

https://pcpartpicker.com/part/swiftech-cpu-cooler-h240x

there ya go duck

edit: no pics there, but follow the embedded links and you'll get what you want from NCIX. Or . . .

https://www.google.com/search?q=h24...niv&sa=X&ei=xGt_VMu5G8SjgwTJhoJY&ved=0CCYQsAQ