Swiftech H2O-120 Compact and Corsair Nautilus 500: Is Water Better?
by Wesley Fink on September 17, 2007 6:00 AM EST- Posted in
- Cases/Cooling/PSUs
Overclocking
As cooling solutions do a better job of keeping the CPU at a lower temperature, it is reasonable to expect the overclocking capabilities of the CPU will increase. In each test of a cooler we measure the highest stable overclock of a standard X6800 processor under the following conditions:
CPU Multiplier: 14x (Stock 11x)
CPU voltage: 1.5875V
FSB Voltage: 1.30V
Memory Voltage: 2.20V
nForce SPP Voltage: 1.35V
nForce MCP Voltage: 1.7V
HT nForce SPP <-> MCP: Auto
Memory is set to Auto timings on the 680i and memory speed is linked to the FSB for the overclocking tests. This removes memory as any kind of impediment to the maximum stable overclock. Linked settings on the 680i are a 1066FSB to a DDR2 memory speed of DDR2-800. As FSB is raised the linked memory speed increases in proportion. The same processor is used in all cooling tests to ensure comparable results.
The Swiftech H2O-120 Compact and Corsair Nautilus 500 both topped out at 3870 MHz - near the top of tested air coolers but below the 3.94 GHz of the Thermalright Ultra-120 eXtreme, and the 3.90GHz of five tested air coolers. Once again test results - this time in overclocking - are near the top but do not match or exceed the best air coolers tested at AnandTech.
It was quite a surprise to find overclocks on the P35 motherboard also topped out at 3.87 GHz, particularly since we could not duplicate the 14X base multiplier on the ASUS P5K Deluxe motherboard. We achieved 3.87 GHz by choosing the 333 strap and setting a FSB of 352 with the 11X multiplier. This turned out to be the highest stable overclock on the P35 board with either water cooler - the same results found on the EVGA 680i board.
As stated many times, the overclocking abilities of the CPU will vary at the top, depending on the CPU. This particular CPU does higher FSB speeds than any X6800 we have tested, but the 3.9GHz top speed with the Tuniq is pretty average among the X6800 processors we have tested with Tuniq cooling. A few of the other processors tested with the best air coolers reach just over 4.0GHz, but the range has been 3.8 to 4.0GHz. Stock cooling generally tops out 200 to 400 MHz lower, depending on the CPU, on the processors tested in our lab.
As cooling solutions do a better job of keeping the CPU at a lower temperature, it is reasonable to expect the overclocking capabilities of the CPU will increase. In each test of a cooler we measure the highest stable overclock of a standard X6800 processor under the following conditions:
CPU Multiplier: 14x (Stock 11x)
CPU voltage: 1.5875V
FSB Voltage: 1.30V
Memory Voltage: 2.20V
nForce SPP Voltage: 1.35V
nForce MCP Voltage: 1.7V
HT nForce SPP <-> MCP: Auto
Memory is set to Auto timings on the 680i and memory speed is linked to the FSB for the overclocking tests. This removes memory as any kind of impediment to the maximum stable overclock. Linked settings on the 680i are a 1066FSB to a DDR2 memory speed of DDR2-800. As FSB is raised the linked memory speed increases in proportion. The same processor is used in all cooling tests to ensure comparable results.
The Swiftech H2O-120 Compact and Corsair Nautilus 500 both topped out at 3870 MHz - near the top of tested air coolers but below the 3.94 GHz of the Thermalright Ultra-120 eXtreme, and the 3.90GHz of five tested air coolers. Once again test results - this time in overclocking - are near the top but do not match or exceed the best air coolers tested at AnandTech.
It was quite a surprise to find overclocks on the P35 motherboard also topped out at 3.87 GHz, particularly since we could not duplicate the 14X base multiplier on the ASUS P5K Deluxe motherboard. We achieved 3.87 GHz by choosing the 333 strap and setting a FSB of 352 with the 11X multiplier. This turned out to be the highest stable overclock on the P35 board with either water cooler - the same results found on the EVGA 680i board.
As stated many times, the overclocking abilities of the CPU will vary at the top, depending on the CPU. This particular CPU does higher FSB speeds than any X6800 we have tested, but the 3.9GHz top speed with the Tuniq is pretty average among the X6800 processors we have tested with Tuniq cooling. A few of the other processors tested with the best air coolers reach just over 4.0GHz, but the range has been 3.8 to 4.0GHz. Stock cooling generally tops out 200 to 400 MHz lower, depending on the CPU, on the processors tested in our lab.
Facebook
Twitter
RSS
58 Comments
View All Comments
psychotix11 - Monday, September 17, 2007 - link
But water cooling has never been about the best price for the buck. The same can be said of any high end parts. Extreme intel, amd FX, 8800gtx, SLI, none of this makes sense for the price you pay.Water cooling falls right into that category. If you really want it to show it's teeth load up an 8800 SLI system, build a custom loop at the 300 and 500 price point, water cool everything, and then compare temps.
Obviously it's going to cost a lot more, but this is a given.
You're article seems to hint that air cooling is better then water cooling, this isn't the case at all. All it shows is that high end air cooling is a better value then a water cooling kit designed for a novice on a budget.
As for "might" and 300, that's wrong as well. My apogee gt + lian DDC, + mcr 220 2x yate loons push, pentosion did drop my temps over a tuniqu tower. The water blocks on the 8800's with an mcr 320 moved even more hot air out of my case and temps dropped more. That CPU loop costs less then 300.
So, DDC pump $65, mcr 220 45, apogee gt 45, and about 30 bucks for fans, tubing and anti corrosive. Brings you to about 200, hardly the 300.
EODetroit - Monday, September 17, 2007 - link
Next time water cool the video card as well, and compare it against the CPU and video card on Air. OC both the video card and CPU and see how far they go. Then measure system noise. I agree with your conclusion that if you're only going to water cool your CPU, that you might as well go with a new air cool instead. But the CPU isn't the only expensive, hot, noisy thing in the case. In fact, the GPU is for many or most of us more expensive, hotter, and noisier.You shouldn't pretend the GPU isn't there. Water cool it as well and compare it THEN against air cooling.
Lonyo - Monday, September 17, 2007 - link
You forgot another element of one of your watercooling setups.It's external.
PLEASE PLEASE PLEASE try looking at other temperatures, e.g. graphics card, motherboard, HDD, when doing these tests.
The heat has to go somewhere. With an air cooler it stays inside the case before passing by the graphics card and getting pushed out the back of the case.
With an external watercooling setup it gets transferred outside without going over the graphics card/motherboard, and hence internally other components beside the CPU should in theory be cooler.
My view is that the main advantage of watercooling is the low CPU temps combined with lower internal case temps, since I wouldn't think of running water with an internal radiator.
Nickel020 - Monday, September 17, 2007 - link
I don't understand how you reached the conclusion, that good air cooling is better than sub 300$ water cooling by testing two sub 150$ water cooling kits.I have recently upgrade from a lapped Scythe Infinity to a 300$ wattercooling setup and at a maximum fan speed of ~1300 rpm (two fans dual radiator) the watercooling beats air by 10K but is little bit noisier since the fans sit at the top of the case. At 7V it's less noisier than the Infinity and still about 5K cooler.
Please test a real 200+$ watercooling setup before drawing such conclusions. The test clearly shows that cheapw atercooling isn't worth it, but starting at about 200$ watercooling will beat air cooling if you choose good components.
Here's a link to jstu about the ebst setup for 250$:
http://www.petrastechshop.com/pecoel.html">http://www.petrastechshop.com/pecoel.html
The waterblock is one of the ebst available, the Swiftech radiators are great for the price and with a modded top the Laing pupm is very quiet and has a excellent performance.
I'd very much like to see Anandtech finally review some proper water cooling!
PS: The review is otherwise very nice, finally a good review that compares low end watercooling with top-end air cooling.
poohbear - Monday, September 17, 2007 - link
dude, that link shows it's $250!!! i can get the thermalright 120 extreme for $65... the choice seems like an absolute no brainer to most.
retrospooty - Monday, September 17, 2007 - link
"'d very much like to see Anandtech finally review some proper water cooling!PS: The review is otherwise very nice, finally a good review that compares low end watercooling with top-end air cooling."
Agreed. It would be great to see a full review, including a few high end kits, cooling graphics cards as well, since that IS how they are used.
This is a high end "best of the best" air cooler against 2 low end "mediocre" water kits.
Wesley Fink - Monday, September 17, 2007 - link
We have also tested two kits closer to $300 in price from top water cooler manufacturers and they did not beat our top air cooler results. The Infinity was not a top performer in our test results unless it was driving a dual push-pull fan configuration. Stock it was slighly worse than the two water coolers we tested in this review.Also, Intel, Swiftech, Thermalright and others have now found that a convex base improves cooling performance mounted to current Intel CPU caps which are manufactured not flat to improve cooling performance. Lapping your Infinity prbably did no harm but lapping a thermalright would likely decrease performance. Swiftech in the past was a strong advocate of flat mirro-finish bases, but they now use the intentionally bowed plate on their top water blocks to improve cooling and overclocking.
Nickel020 - Monday, September 17, 2007 - link
Well the problem with watercoolig is that no big manufacturer offers a good kit, you absolutely need to mix and match parts from different manufacturers.Go to the forums here or at xtremesystems.org and everybody will confirm that a *well chosen* 300$ watercooling will beat any air cooling. You just need the right components, like in the ones in that kit.
And I know about the lapping/bowing, I lapped my CPU as well and got a 13K lower temperature with both lapped than before. A bowed cooller usually has the advantage since almost all heatspreaders are concave, therefore a convex cooler is needed to make a good contact. But if you lap both CPU and cooler you're usually better off if you do it right.
I would love to see you review a watercooling setup with good components, there almost no reviews on the net comparing good watercooling to top-end aircooling.
yyrkoon - Monday, September 17, 2007 - link
I think the real questions are . . . is it really worth the Hassle/danger/cost.
Even a $300usd water cooling system is not a good water cooling system, and one that uses less than 1/2 tubing/waterbloccks etc, is not a good water cooling setup.
Anyhow, I am not an advocate of water cooling, and I have been saying air is better than water for a long time now (couple of years), not because of cooling performance alone, but because of the three things mentioned above . . .Hassle/danger(or risk if you like)/costs. Building a top notch cooling system would set you back about $400 a year ago, now, probably more like $500-$600, but I wouldnt know for a fact, beause water cooling for PCs is like 4x4's with a 128 inch lift kit, its made to compensate for something, perhaps a small . . . CPU( or more likely something else) ?
poohbear - Monday, September 17, 2007 - link
well i'll be damed, i never considered watercooling cause it was too exotic and expensive, but it seems there's no point considering it to begin w! the best aircoolers are quieter AND run cooler than watercooling. Thanks for showing this fantabulous change of events anandtech.