17-Way roundup of 120mm fans @ Madshrimps

rpsgc · Post by **rpsgc** » Fri Mar 24, 2006 12:58 pm

Here:
http://www.madshrimps.be/?action=getarticle&articID=421

Fans tested:

AC Ryan Blackfire4
AcoustiFan AF120C
AcoustiFan AFDP-12025
Aerocool Turbine 1000
Arctic Fan 12
Coolermaster ALU AAF-B12-E1
Coolermaster LED Silent TLF-S12
Coolermaster Ultra Silent SAF-S12-E1
Coolink SWiF-1201
GlobalWin 1202512L
mCubed X12
Nexus D12SL-12 (black/white)
Papst 4412 F/2GLL
Scythe S-FLEX SFF21D
Scythe S-FLEX SFF21E
Scythe S-FLEX SFF21F
Spire FD12025C1E3

JVM · Post by **JVM** » Fri Mar 24, 2006 2:16 pm

Interesting, but the Nexus orange fan runs 100 rpm slower than the black & white version??? And the Nexus B & W runs 917 rpm at 12V?

jmke · Post by **jmke** » Fri Mar 24, 2006 2:56 pm

that's what the motherboard fan header tells me at least; I was surprised as well that the Orange version showed different results; I've asked Nexus about this, waiting for an answer:)

Butcher · Post by **Butcher** » Fri Mar 24, 2006 4:39 pm

Could just be manufacturing variance, but that's over 10%, which is pretty high!

shades_of_blue · Post by **shades_of_blue** » Fri Mar 24, 2006 7:44 pm

That sound file for Scythe S-FLEX SFF21D sounds smooth and even the audio spikes shown were pretty even, very nice.

m0002a · Post by **m0002a** » Fri Mar 24, 2006 8:59 pm

The GlobalWin seems similar in overall performance to the Nexus (but maybe more durable since it has a ceramic bearing), but sells for only about $6.00 here:
http://www.svc.com/1202512l.html

jmke · Post by **jmke** » Sat Mar 25, 2006 2:02 am

that's a very nice price for a HQ fan, Papst is more expensive afaik

kenyee · Post by **kenyee** » Sat Mar 25, 2006 7:48 am

Nicely done review. I didn't even know about the Scythe fans...they should make a thermistor controlled one.

Now we just need Madshrimps to do a roundup of 80mm and 92mm fans

JazzJackRabbit · Post by **JazzJackRabbit** » Sat Mar 25, 2006 7:53 am

I'm disappointed in scythe fans. E sounds horrible with hollow low frequency noise, D is much better but it still has the same hollow sound, although a lot less not. Overall D sounds about the same as Nexus, however nexus spins 200RPM faster, so nexus is a clear winner here.

JVM · Post by **JVM** » Sat Mar 25, 2006 7:57 am

JazzJackRabbit wrote:I'm disappointed in scythe fans. E sounds horrible with hollow low frequency noise, D is much better but it still has the same hollow sound, although a lot less not. Overall D sounds about the same as Nexus, however nexus spins 200RPM faster, so nexus is a clear winner here.

As I recall, the Scythe D spins at 844 and Nexus at 917 in testing.

JazzJackRabbit · Post by **JazzJackRabbit** » Sat Mar 25, 2006 9:14 am

JVM wrote:
JazzJackRabbit wrote:I'm disappointed in scythe fans. E sounds horrible with hollow low frequency noise, D is much better but it still has the same hollow sound, although a lot less not. Overall D sounds about the same as Nexus, however nexus spins 200RPM faster, so nexus is a clear winner here.
As I recall, the Scythe D spins at 844 and Nexus at 917 in testing.

That's MB readout, I wouldn't trust it too much.

rpsgc · Post by **rpsgc** » Tue Mar 28, 2006 9:13 am

Why bother posting things... someone will just come around the corner and post the same s**t without even searching and that's it. Dead thread.

DrCR · Post by **DrCR** » Wed Mar 29, 2006 7:24 pm

Not to knock on MadShrimps too much as I have found much of their material useful in general, but this is kinda run-of-the-mill type stuff to be posting at SPCR. I give the reviewer a lot of credit for going through all the hassle he did, but most of us here are looking for uber-silence geek material.

jmke · Post by **jmke** » Wed Mar 29, 2006 10:23 pm

and the globalwin fan, scythe fans and nexus which are mentioned in 99% of the latest threads in this forum are not for geeks?

when you start putting your ear next to a fan in order to hear if the motor noise is disturbing or not at low fan voltage, you've entered "uber-silence-geek-world" fyi.

run-of-the-mill would not even include dBA result just RPM.

Butcher · Post by **Butcher** » Thu Mar 30, 2006 3:29 am

jmke wrote:run-of-the-mill would not even include dBA result just RPM.

A few years ago maybe. Quiet computing is becoming more mainstream though.

nutball · Post by **nutball** » Thu Mar 30, 2006 3:50 am

DrCR wrote:I give the reviewer a lot of credit for going through all the hassle he did, but most of us here are looking for uber-silence geek material.

You mean eight pages of review per fan, including a report on whether or not the fan makes for a hot date and picks up the tab in the restaurant?

jmke · Post by **jmke** » Thu Mar 30, 2006 4:00 am

Butcher wrote: A few years ago maybe. Quiet computing is becoming more mainstream though.

only a handful of review sites use dBA meter in their fan reviews... now. not years ago, but now.

their idea of a fan review is posting 30 pictures from all angles; installing inside their case, sometimes they will post their CPU temp (not before or after, just their current cpu temp) and say something like "silent and good". More sites even skip the whole testng part, post pics and specs and call the product "good".

here are a few examples, I used the Arctic Fan 12 for this little pratical test

no temp, no noise, in short: no testing
- http://www.frozenmod.de/test-arcticfan.php
- http://www.driverheaven.net/reviews/Art ... ng%20fans/
- http://www.burnoutpc.com/modules/wfsect ... icleid=226
- http://www.doomedpc.com/?q=node/68
- http://3dfusion.de/artikel/arctic_cooli ... Fan_Serie/
- http://www.rbmods.com/Articles/Arctic_c ... fans/1.php

some temp tests:
- http://www.pc-modders.com/modules.php?n ... icle&sid=6
- http://www.crazy-oc.de/crazy-oc/index.p ... ews&rid=66
- http://www.absoluteinsight.net/2748

some dBA tests, no comparison, not very detailed, small temp test:
- http://www.oc-squadron.com/index.php?op ... mitstart=3

now if you say that run-of-the-mill reviews included RPM, TEMP, NOISE tests, you might want to clarify what you understand by "run-of-the-mill"

it makes me wonder why I would invest hours of time and material in testing a product, if I might as easily just post a few pics , list the specs and call it a day?

quikkie · Post by **quikkie** » Thu Mar 30, 2006 4:58 am

it sounds (forgive the pun) like some people don't realise just how much effort goes into reviews like this...

Thanks for the effort jmke!

Weldingheart · Post by **Weldingheart** » Thu Mar 30, 2006 6:46 am

DrCR wrote:Not to knock on MadShrimps too much as I have found much of their material useful in general, but this is kinda run-of-the-mill type stuff to be posting at SPCR. I give the reviewer a lot of credit for going through all the hassle he did, but most of us here are looking for uber-silence geek material.

how about creating silentpcfanreview.com?

jmke,could you please put numbers from original Big Typhoon fan(Hong Sen?)to the list for comparison purpose?

diver · Post by **diver** » Thu Mar 30, 2006 8:44 am

Its not all that easy to test fans or coolers, for that matter. There are a lot of variables. The cooler they used has very tightly spaced fins and does not work that well at low air flows, according to Madshrimp's own tests. They tested with an open case. How about a closed case test, but that brings in the exhaust fan as a factor. I wonder if the various fans would interact differently with a CAG.

In the end you have to hope the results are good enough to give some guidelines as to which are the good ones and which are the culls. But, I would not jump to conclusions and say A is better than B because it was 1 db quieter while providing a 1C lower CPU temperature under these circumstances.

rpsgc · Post by **rpsgc** » Thu Mar 30, 2006 8:48 am

quikkie wrote:it sounds (forgive the pun) like some people don't realise just how much effort goes into reviews like this...

Thanks for the effort jmke!

This man knows it! Hear hear!

jmke · Post by **jmke** » Thu Mar 30, 2006 10:38 am

this review was posted today, what do you think? http://www.pro-clockers.com/article.php?id=127
Sharkoon Silent Eagle 2000 120mm fan version.

jmke · Post by **jmke** » Fri Mar 31, 2006 5:45 am

this review confuses me: PC-Experience.de google translate

they used a $2000 dB meter and were able to "measure" 11dBA from certain fans... no offence but.. what the ?????

# Aerocool Turbine 1000
# Aerocool Turbine 2000
# Aerocool Turbine 3000
# Enermax Warp
# ichbinleise 12/1000
# Papst 4412 F/2GLL
# Papst 4412 F/2GL
# Papst 4412 F/2GL + NMT-2
# Revoltec AirGuard
# Scythe S-Flex SFF21D
# Scythe S-Flex SFF21E
# Yate Loon D12SL-12

temp tests were done with SI-120 inside a case.

I can't post attachments here; sorry to have to link off-site, I've made a small comparison table between their results and mine of the "same" fans
http://www.madshrimps.be/forums/showthr ... adid=22871 (scroll down).

Tibors · Post by **Tibors** » Fri Mar 31, 2006 6:26 am

jmke wrote:this review confuses me:
[...]
they used a $2000 dB meter and were able to "measure" 11dBA from certain fans... no offence but.. what the ?????

If you lookup the technical data for the SPL meter they used you'll find it can't measure that low. This is not strange since the thermal noise in the microphone and the electronic noise of the circuitry inside the SPL meter for most sound meters add up to ~15dB. So measuring below that is impossible, even if your ambient would be lower.

My guess is the table shows how much they measured above the ambient noise. This is of course mathematically incorrect to do, but......

Can you change that url in your last post to something with [ u r l = h t t p : ...]blah[ / u r l ]. The way it is now it causes sidescrolling

jmke · Post by **jmke** » Fri Mar 31, 2006 6:34 am

putting the dBA issue aside, there are still some things not quite right, if you look at the Turbine 1000 vs Papst at 12V

*in their test the Turbine cools better than Papst, is louder than Papst and spins slower.
* in my test the Aerocool cools less than Papst, is more silent and spins slower.

(less/more somewhere I think I screwed up the cool english language in this post)

cmthomson · Post by **cmthomson** » Fri Mar 31, 2006 9:30 am

jmke wrote:this review was posted today, what do you think? http://www.pro-clockers.com/article.php?id=127
Sharkoon Silent Eagle 2000 120mm fan version.

The Sharkoon web site shows two 120mm versions, 1000 RPM and 2000 RPM. The 1000 version is rated by them at 36.7 CFM and 19 dBA.

Have they sent you these to review yet? I'm very curious how they compare to the Nexus and the AcoustFan DustProof 120mm fans, which move 37 CFM at about 10V and 6V respectively.

According to several reviews, the Sharkoon fans have the speed locked; changing the voltage does not change the RPM. Pity.

jmke · Post by **jmke** » Fri Mar 31, 2006 9:37 am

how can you lock the speed of a fan; what happens if you give it only 7v, it stops working?

I'm not testing any more 120mm fans for a while, until the next roundup maybe, test setup is now being re-used for CPU roundup. sorry;)

nomoon · Post by **nomoon** » Sun Apr 02, 2006 6:22 pm

Thank you for supplying the data in the article! I created this graph the first day that you posted, but I just got around to posting it. There is a lot of a data in the article, so I created the following to compare a few of the fans. When comparing fans at a fixed voltage, some fans will be louder, though they might be pushing more air, which makes it hard to compare without a graph. I effectively grouped all of the Scythe fans together by not connecting them with lines. Based on the data below, it looks like the Nexus and GlobalWin fans beat the Scythe fans by a small margin. If you can control your fan speed, then maybe the Nexus and GlobalWin fans offer identical performance. I suspect that low speed noise performance is dominated by bearing noise, and high speed performance is dominated by turbine blade design.