Partners
The Games selection
kids :
Bob
Throw bubbles so as to make the ones that appear in the game disappear. For this, use the Right / Left arrow keys to duck or move about, and the...
|
crazy :
PC Breakdown
What is worst than a Fatal Error occuring during a game you did not save? Unleash your rage at your PC in this game. Blow it to pieces, it feels so...
|
Sponsored links
Small Water Versus Big Air, Part 3: Cooling Questions Answered
- Email |
- Print |
- Comments (50) |
- Share
Table of contents
- 1 – The Equalizer(s)
- 2 – Test Hardware
- 3 – Cogage True Spirit Fan Configurations
- 4 – Corsair H50 Fan Configurations
- 5 – Rosewill FORT120 Fan Configurations
- 6 – Test Results
- 7 – Conclusion
- 8 – More on this topic
Our quest to find an advantage—any advantage—in today’s maintenance-free compact liquid coolers began with CoolIT’s Domino A.L.C. sealed liquid system and the low-cost Cogage True Spirit 120mm tower. Unfortunately, the liquid cooler only performed as well as the air cooler when its custom-wired fan was forced to operate at its maximum speed of 2,800 RPM, versus the air cooler’s maximum fan speed of 1,600 RPM. The enormous increase in cost and noise for sub-par cooling performance put the Domino A.L.C. at an equally big disadvantage compared to the simpler sink and fan.
Our quest continued with Corsair’s H50 sealed-liquid system and Rosewill’s FORT120 air-cooling tower, where the air cooler’s fast 2,400 RPM fan gave it a cooling advantage comparable to its resulting acoustic disadvantage. Readers complained that the tradeoff between cooling and noise could only be addressed by using the same fan on both coolers, while Corsair pointed out that we weren't using the configuration the way it was designed, blowing cooler air through the radiator. In contrast, air coolers must typically use warm case air to cool the processor, which is a disadvantage not properly reflected on an open bench. End users typically don’t run open systems, and the H50 was designed to use the case advantageously.

This leaves us with a few additional configurations to test. How much better will the Corsair H50 and the Cogage True Spirit function with Rosewill’s FORT120 more powerful fan? How much worse will the Cogage True Spirit and Rosewill FORT120 perform in a closed system? Today we test each configuration inside a traditional oversized mid-tower case, using two different fan configurations for each unit.
Get ready. All of your questions are about to be answered.
Best prices for tested products
- 1 / 3
- Next
-
Sponsored links
Related articles
-
Part 2: Little Water, Big Air--Corsair H50 Versus Rosewill FORT120
Do sealed liquid coolers provide a compelling performance alternative to air? Liquid-cooling proponents cite the benefits of a larger radiator placed further from the heat source, saying it provides superior results. Meanwhile, fans of air-cooling...
-
Ambient temperatures of 23.2° to 23.6° Celsius were subtracted from sensor readings to provide the “above ambient” temperatures shown below. An extra-warm, overclocked Core i7 processor under the load of eight Prime95 threads provided the heat....
-
With four U-shaped heatpipes and a 6” distance from the base to its top fin, Rosewill’s FORT120 resembles many other 120 mm tower coolers. Various manufacturers will play fast and loose with the facts by calling their four pipes eight (counting risers...



Ditched my Domino A.L.C. for a Xigmatek Thor's Hammer with 2 X Scythe fans. Strapped it on a Core i7 920, bumped it to 4.0Ghz and never looked back.
Interesting. This is good to know for a future build, since cheap water cooling was always a temptation for me.
I was looking at water, then I decided if I really want to do this I'm going to have to spend at least 200 if i want a good, effective, upgradeable system. So I'm going to get the megahalem or the thermalright TRUE extreme. and stay with some sick air.
Air is still a better value unless you value noise or lack there of.
Water offers lower noise @ a slightly less extreme overclock, but who runs 4.0Ghz plus everyday.
still using my old thermaltake big typhoon with a few mods - sealed the gaps on the sides for more air pressure and using a 12cm "thick" fan from a dell tower (crazy) and the same type fan to extract air - works a treat
kinda proves that when your going water cooling, do it PROPERLY not a pre made kit
if i was to do water cooling, i would go all the way with a modded car radiator, drum for a water sump and a few powerful decent sized pumps to start off with to keep everything sweet, none of this "barely better then stock" bs.
The corsair h50 is NOT a water cooling solution. Not even close. At best, call it an "optimized" air cooler. The only situation where you would want one is if you need to install a cooler in a tight space. Otherwise, it's higher cost really ruins any value it has.
Get a Swiftech H220 in there and it will beat the air coolers pretty well, besides that there are no out of the box water cooling setups that can actually beat high end air coolers by anything meaningful.
Good article though, your best articles are when you take the time to answer these odd questions that are commonly asked by the enthusiast.
As many times as I see bottom-PSU cases like the Antec 300 recommended in builds in the Forum, the lingering question for me becomes, "Suppose I do have a bottom-psu top-panel-fan case. Would that make a difference?" Or, is there ANY situation where the cooling performance of this type of liquid cooler is actually superior to a big air cooler?
Thanks for listening to the feedback and doing follow ups like this TS. Really gives the community a reason to become involved knowing our unanswered questions get addressed.
^Absolutely.
It would be good to see a comparison of the H50 with a push-pull fan setup.
I have an old-style Antec SLK-B case with a side-port fan (intake) that blows air into and through two push-pull 1500 RPM fans with the H50 radiator sandwiched in-between; these fans blow out, not in. The temp drop from a single fan to dual fans is around 7c degrees.
The biggest advantage of the H50 is noise, or lack thereof. My Tenma sound meter records less than 30db within 1ft of the case. Can't say that for any of the other HSFs I've tried (mostly stock).
still prefer the liquid cooler as it reduces the need for an exhaust fan, only I should have to match it to a powerful processor for good use of it.
i have the thermaltake v9 case with 230x230 large top exhause fan, and psu is at the bottom of the case, would the h50 do better for me?
Otherwise fine article but who would use 250w consuming freak CPU in their computer in everyday life? I'd love to see the results of using stocks-speed instead over clocking.
Air is still a better value unless you value noise or lack there of.Water offers lower noise @ a slightly less extreme overclock, but who runs 4.0Ghz plus everyday.
Air can be very quiet, but it requires that you keep fan speeds low. A good 120 mm fan running at 1000 rpm or less is just about inaudible, but such low speeds mean you can't overclock a whole lot. I'd guess based on my experience that you wouldn't want to pump more than about 150-160 W through a decent 120 mm tower heatsink like the ones used in the review if you keep the fans
@vinshon
IMO you should be able to build a custom water setup in that case. Custom water will cool much better than the h50 anyday. Will it cost more than the h50? Yes but if you're gonna do water do water, these little prebuilt kits really don't cut it when comes to shedding heat or noise levels
@vinshonIMO you should be able to build a custom water setup in that case. Custom water will cool much better than the h50 anyday. Will it cost more than the h50? Yes but if you're gonna do water do water, these little prebuilt kits really don't cut it when comes to shedding heat or noise levels
Yep, I started out with a little pre-built kit but now I use a custom built loop to cool my CPU and northbridge/southbridge. Keeps the temp way way down on both. I have a 780i and the northbridge & southbridge have terrible stock cooling and the CPU is a quad core overclocked to 3.75GHz and it still stays icy cold.
Cooler Master Hyper Z 600 with dual scythe ultra kaze 38mm x 120mm. Nuff said'
yeah i would like to see you guys do some loop testing with some danger den stuff. I have a custom loop with an old ehiem pump, danger den waterblock, and a dual 120mm heatercore, and it stomps temps. I reach the limit of the cpu way before I tax the loop.
I just ditched the Domino when a "self-contained" pump burst and spewed coolant all over the VGA's, motherboard, and PSU. It was a nifty little thing, but the Zalman 9700 replacement I got seems to cool better and cost a lot less. Having tried both big air and little water, I really don't see the benefit of little water. If you want liquid, just go full out with "big water" in my humble opinion.