A lot of good information. Thanks. The main problem I have with fans is that in the several dozen computers I've built over the years, all of them have been silent once built. Stick your ear to the exhaust fan and hear nothing; However, 4-6 months later they are all loud or obnoxiously loud. Is there any where or any way to find out how loud fans are later in life?

Lot of good testing and information provided with each fan, fine work right there.

It's definately an article which I will be referring to for the next couple of years, though I do have a couple of questions.

What was the ambient sound level in the environment where the fans were measured? Did fans like the Nexus make a noticeable noise (on ear) and what was the accuracy of the sound meter (I've read somewhere the sound meter has an accuracy of 1,5 dB (94dB@1kHz), though what's the accuracy around 30-35dB).

I was also surprized to see the Noctua NF-S12B and NF-P12 performing far below expectations, while there are also some good quality reviews available about the same fans. Though if I look to the results, I can conclude that the Akasa 120mm White LED fan performed better (higher CFM with lower noise level at most measurement points (voltage levels)). What about this? Is it really an expensive piece of average fan with a good marketing story behind it?

This test also shows that the Nexus 120mm is once again confirmed as one of the most silent case fans available in the market, I guess I'll certainly have a look on this fan for my next case update.

@bdcrazy
This happens with most Sleeve Bearing fans which are mounted horizontally. If you've mounted your fans vertically, sleeve bearings are still supposed to last a little bit longer, though still not as long as ball bearings and other good bearings out there. This only applies to general situations however, it could be that some manufacters use better bearings than others, while the bearing belongs to the same type...

oh man i went reading all the pages to the end, to bad the 14cm fans dident perform better then the 12cm fans.
I will definitly check this review again when im gonna buy new fans for my system.
Thanks Xbitlabs!

Hats off for another rather excellent article. This is great information!

Exhaustively fantastic article as usual. Just a clarification: the power consumption chart was at each fan's max speed, not the power consumption when they were all at the same dB level, right?

I do have request: If ever you were to do an update to this fan roundup I would love to see the other 3 S-Flex models included. The 1200rpm model is about the only one that is ever tested.

30dba minimum is hurting the silent fans in my opinion. it is not difficult to get your computer noise level under 30dba, it only gets difficult somewhere under 20dba.

it would be a lot more interesting to see the minimum noise levels of the silent fans.

I would bet the only reason the Scythe FDB fans have a longer mtbf rating is due to the fact they rate theirs according to a temperature of 25c, and Thermalright rates their mtbf according to a temperature of 40c. Thermalright is being more honest with the rating.

Also, in the review about the Thermalright fans, it mentioned the 1000rpm model being slightly louder at the same speeds, but that isn't what the charts display, it shows it as slightly quieter.

This is the absolute finest fan test I've ever seen, and your overall methodology is unsurpassed in my experience, excellent work!

Two of your graphs, the "Maximum airflow and noise level" and "Airflow at 33dBA noise level" are very useful in determining overall fan performance, and should be standards in your fan testing. Simply adding newly tested fans to them would create a small but highly useful database on fan performance.

Even if your noise level numbers were not absolutely correct regarding SPL, (if what is correct could be known and agreed upon) the numbers relative to each other should be correct, which is IMO the most significant factor.

While you may not be making all the manufactures out there happy, you certainly have made myself and your readers happy, thank you. That is, as long as your readers have made good purchasing decisions relative to your test data (I've done well and not so good too....)

Question, in the Audio equipment world, a difference of 3db in SPL is considered to be the minimum change that will be audible to most people. Do you agree with that?

Also, regarding the subjective fan noise profile, "voice", or "style", how significant do you feel the subjective noise profile is between two fans that otherwise produce the same measured noise level?

Again, excellent work gentlemen.