Noise

For many enthusiasts upgrading cooling the goal is maximum stable overclock, and they will live with the inconvenience of a louder system. For other users silence is the most important factor, and these users will forgo maximum overclocking if that increases system noise levels.

There are virtually no power supplies that do not include a fan. While Zalman and a few others do make an expensive fanless power supplies, we have not seen a fanless unit larger than 500W, or one that would be used for seriously overclocking a system. With that in mind the noise level of the system with all fans turned off except the power supply was measured. The power supply used for the cooling test bed was the OCZ PowerStream 520, which is one of the quieter of the high performance power supplies.

We have also measured the Corsair 620-watt and Mushkin 650-watt power supplies which are reported to be quieter than the OCZ. Both the Corsair and Mushkin are indeed quieter at idle or start up speed. However, as soon as load testing begins and the PSU fan speed kicks up the measured noise level is almost exactly the same as the OCZ PowerStream 520 watt power supply.

We are currently in the process of evaluating "quieter" power supplies for an update to our cooler test bed. We will make changes to that test bed as soon as we are confident in the noise measurements and test procedures with a variable speed quiet PSU. We plan to evaluate additional power supplies and configurations in our upcoming 120mm fan roundup, at which point we will complete the transition to a revised and lower noise cooler test bed.

The noise level of the power supply was 38.3 dB from 24" (61cm) and 47 dB from 6" (152mm). The measured noise level of the test room is 36.4 dB, which would be considered a relatively quiet room with a noise floor slightly below the OCZ PowerStream 520 PSU.

We measured noise levels with the Thermaltake MaxOrb and the Scythe Andy Samurai Master with the stock Scythe fan at its single speed and the embedded MaxOrb fan at Low and High speeds. Results were then compared to the other coolers/fans tested in this category. Measured noise levels in this chart should be considered worst case. Measurements were taken with an open side of a mid tower case 6" and 24" from the HSF. Real world would be a completely closed case resulting in a further reduction in noise.

Noise Level - 6

Noise Level - 24


The measured noise levels at 6" and 24" are below the system noise floor with the quiet Scythe fan on the Andy Samurai Master. The Thermaltake actually specifies even lower noise levels than the Scythe, but at both low and high noise it was above the 24" noise floor at 40.7dbA at low speed and 45.8dbA at high speed. The MaxOrb fan is definitely audible, but the noise is not particularly irritating in frequency. If you are very sensitive to noise you should be aware you will likely hear the MaxOrb fan with an open case.

At the 6" measurement the MaxOrb was below the noise floor at low speed, and significantly above the noise floor of 47dbA at high speed with a measured 52.5dbA. The Tuniq Tower 120 on high, the Zalman 9700 on high, and the Monsoon II Lite stressed were noisier than this, but this is still one of our highest measurements. Again the noise frequencies are well placed and not particularly irritating as there is none of the high-pitched whine that some fans generate, but the MaxOrb is definitely not silent.

Overclocking Conclusion
Comments Locked

50 Comments

View All Comments

  • Wesley Fink - Tuesday, June 5, 2007 - link

    We now use a standard quality silver-colored paste in all our reviews. But in our experience and testing the thermal paste makes absolutely no difference in performance. The paste fills gaps and needs to remain fluid and not dry out.

    A famous review from several years ago showed toothpaste and Kraft Vegimite equal or superior to AS3 - at least until they dried out. The point was use a decent paste but don't obsess on it - as it really makes no difference. This is particularly true now that all the current CPUs use heatspreader covers and you no longer are mounting on a tiny die as you were on AMD Socket A.
  • rjm55 - Tuesday, June 5, 2007 - link

    The review Wesley is talking about was at the Australian site Dan's Data at http://www.dansdata.com/goop.htm">http://www.dansdata.com/goop.htm
  • Chunga29 - Monday, June 4, 2007 - link

    Whoa! Two articles today? Heh. I was just finishing the 2900 fiasco article and suddenly this shows up. [Big grin!]

    Only problem with the Hexus article is that overclocking isn't even a factor. I'm a bit confused as to their results, as they have the Zalman 9700 performing in the top of the tested HSFs, trailing the Tuniq by 1C. The AT results show Tuniq leading by 2C at load, but even more interesting is that the AT results are about 10C cooler on every HSF (at least where they overlap).

    Oh... WAIT! The Hexus reviews are using a Pentium D Extrene 840!? What in the hell is up with that? Why don't we do some testing with the original Athlon FX-51 while we're at it? While most people won't purchase a $1000 CPU like the X6800, at least Core 2 Duo is very popular and many are OC'ing E6600 chips to similar levels (and beyond). Perhaps the VX is the best HSF when it comes to cooling Pentium D setups... but that's about as far as I'd take the Hexus roundup.

    Honestly, I'd like only a few things from AnandTech cooling reviews:

    1) Testing on AM2 as well as 775.
    2) Temps of various components.
    3) More HSFs! Heh. A roundup would be great.

    Two HSFs in one ariticle is a start, but how about four or five similar HSFs at a time? Half the material is just filler (copy paste from previous articles), so other than the spec sheets, intro, conclusion, and charts there's not much going on here. Am I the only one that skims (at best) the middle pages?
  • DrMrLordX - Tuesday, June 5, 2007 - link

    Hexus used a Pentium 840D because it's a stupidly hot processor that's cheap. Pentium Ds can easily put out as much heat as Kentsfields.
  • Chunga29 - Tuesday, June 5, 2007 - link

    I understand that. The problem is that performance with a Pentium Extreme 840 doesn't really tell you anything about how the cooler will perform with other processors. For what it's worth, a heavily overclocked Core 2 Duo X6800 (or better yet a QX6700) is also a stupidly hot processor. I'd wager that the output of QX6700 easily exceeds the 840EE.

    Testing with outdated equipment is meaningless, regardless of how hot the old parts get. That's why it would be ideal to see current AM2 performance as well. I remember a LOT of reviews hailing the Zalman 9500 as the greatest cooler ever, and while it appears to do great on AM2 it's pretty average on an overclocked C2D. I hope you can see that the same could easily be true of comparing performance with 840EE to current C2D and AM2 chips.
  • DrMrLordX - Wednesday, June 6, 2007 - link

    Actually, uh, your point isn't really valid. A heavily OCed Smithfield can easily pump out heat like a heavily OCed Kentsfield. Heat is heat as long as its distributed properly by the IHS and properly-applied TIM.

    The age of the equipment in question is irrelevant unless a particular HSF has a mounting mechanism that is somehow better-suited to one socket or another, not that that's an issue when dealing with Smithfields and Kentsfields (both s775 processors).

    Using a Smithfield to test the limit of modern HSFs is entirely appropriate, given its massive heat output that can still rival that of Kentsfields and its socket which is identical to the socket-type used with Conroe and Kentsfield processors.
  • Stele - Monday, June 4, 2007 - link

    quote:

    Honestly, I'd like only a few things from AnandTech cooling reviews:
    1) Testing on AM2 as well as 775.
    2) Temps of various components.
    3) More HSFs! Heh. A roundup would be great.


    I second that, though especially no.(2). That would help decide between a traditional downdraught HSF blowing air onto the motherboard vs. the tower ones.

    quote:

    We do not use auxiliary fans in the test cooling case

    It might be useful to have a proper cross flow of air through the case too, e.g. one 120mm fan each for intake and exhaust. This would not only be more reflective of regular casing setups but would also ensure a level playing field - otherwise the tower ones may potentially have a greater advantage over the downdraught ones (since they can exhaust hot air out of the case through the case rear) than real-life setups may show.

    How about including the Enzotech Ultra-X in a future review? Seems to be quite a good performer by all accounts! :P
  • Wesley Fink - Monday, June 4, 2007 - link

    The MaxOrb is reputed to be at least the equal of the Big Typhoon VX in performance and it is a more recent design.

    We have an Enzotech Ultra-X in for testing. You will see it in a future review.

    We had planned a big spread on board component temperatures, but when the down-facing fans didn't even come close to the heatpipe towers in cooling efficiency or overclcoking it became a moot point. IF the down fans really cooled better they would cool more efficiently and/or allow a higher overclock. Neither is the case in our tests. We run our top OC speed with a pretty hefty chipset voltage increase, and if the dowh-facers cooled components better we would get a better overclock, or at least an equal OC. Instead OC is average at best with all 3 expensive down-face designs.
  • DrMrLordX - Tuesday, June 5, 2007 - link

    Reputed by who? Would it really kill you guys to review the Big Typhoon VX, or an older Big Typhoon with an FM-121?

    The Big Typhoon has had at least one other good showing here:

    http://forum.abit-usa.com/showthread.php?t=103462">http://forum.abit-usa.com/showthread.php?t=103462

    It beat the Tuniq Tower 120 pretty solidly there as well once it had a decent fan in it. Note that at least one of the fans that allowed the Big Typhoon to defeat the Tuniq Tower 120 has approxmiately the same cfm rating as the fan included with the Big Typhoon VX.
  • Chunga29 - Monday, June 4, 2007 - link

    Addendum: it might be interesting (although it's too late now) to test with an E6600 instead of the X6800... maybe switch to a Q6600 in the future? When AM2 gets quad core (if it gets quad core?), I'd like to see head-to-head comparisons of HSfs on both platforms. Maybe retest the current leaders for reference (Ultra-120/eXtreme, Tuniq, Hyper6+, and a couple others?)

Log in

Don't have an account? Sign up now