I think it's time we approach fan reviews differently - Overclock.net - An Overclocking Community
Forum Jump: 

I think it's time we approach fan reviews differently

Reply
 
Thread Tools
post #1 of 72 (permalink) Old 06-20-2020, 11:03 AM - Thread Starter
Never Finished
 
airisom2's Avatar
 
Join Date: Aug 2011
Location: Within the Milky Way
Posts: 2,212
Rep: 279 (Unique: 216)
I think it's time we approach fan reviews differently

Whoops, this actually turned out to be a wall of text. Oh well

I mean, a couple of reviewers are kinda already doing it, but the majority still go off of mfg specs which is a poor indicator of real-world performance.

1. RPMs don't matter
With all of these unique fan blade designs, motors, bearings, frames, etc. one fan at 1100rpm may move as much air as another at 1500rpm if not more.

Instead, airflow at a fixed noise level ("noise-normalized" as popularized by Gamer's Nexus) will give a more accurate indicator of fan performance. Increasing those datapoints will give you a more comprehensive breadth of the fan's performance.

2. MFG specs don't matter (airflow)
They can be a poor indicator of how the fan performs in real-world usage. These fans are measured in free air, usually at 12v. We know that dust filters, solid front panels, that hexagon case mesh material and whathaveyou, and especially radiators will shape the performance of the fan. Depending on the fan type, those restrictions could take out a chunk of your airflow.

Instead of relying on those specs, having airflow measurements through various obstructions would paint a better picture of the fan's airflow performance. For instance, fan a may have 120cfm in free air, but it loses 30 going through a dust filter. Fan b may also do 120cfm in free air, but loses 45cfm through a dust filter. Or a radiator. Or a front panel. Those are the numbers we need.

On a side note, I'm also curious about how the airflow shape contributes to thermals. For instance, those silverstone AP fans with the neat looking grill on the backside shapes the exhaust airflow so that it travels straight. Your typical fan blows out air in a V-shape. You have the most pressure around the edge of the fan, and there is a deadzone with turbulence near the middle. If you're trying to push some air towards your graphics card or other component in the case, wouldn't a fan with a more focused airflow path be more desireable? Though, with all of the stuff going on in people's cases nowadays, that's probably a wash, especially if you can brute force it with more fans. Again, a side note.

3. dB doesn't matter
The noise a fan makes is much more than just a peak dB result. You'd think a fan that's advertised for 32dba would be quiet until you hear the ball bearings growling at 1000rpm, the turbulence of the intake fan next to a grill or radiator, or the harmonics and resonances the fan produces. I remember reading some amazon review of the a1225 about some guy returning the fans because of the sound signature. Could have been a shill review, but it's still a very important point, possibly more important than airflow metrics if you've a keen ear.

Instead, I think we should adopt something similar to speaker and headphone measurements. Some of you audiophiles on here know how some headphones have their peaks and dips in the frequency range, and that resonance could make or break the headphones without equalization. There is no equalizer for fans; what you hear is what you get.

So, being able to point out specific noises in a frequency chart, and how it relates to what you'd hear in person would be great information to have. For example:

Fan a: This 3dB peak at 500hz is the DBB growl, this peak at 2kHz is the motor noise resonance. In fan b, the fan at 500Hz is 6dB quieter due do it using FDB, and the rubber isolators have dampened the resonances from the frame which are pretty common at 4kHz. However, you can see here in the 700-1kHz range that the mesh screen in front of the intake fan is creating 4db more noise on fan b than fan a, so it's recommended to use this fan in a push configuration or with a bit of a gap in pull configuration to reduce the turbulence.

And then, you put a sound clip on top of that to really put things into perspective, and maybe some spectrometer results of the fans on a cooler or in a case vs. open air to see the tonality differences. Wouldn't that be nice to have?

4. Static pressure doesn is too ambiguous


But lol yeah. Static pressure isn't really something that is important nowadays when you have an airflow:noise graph. If it can push a lot of air through a radiator, then that's all that matters because that's where your temps are going to be based off of (considering well ventilated case, optimized airflow path, blah blah). While static pressure has been used as an indicator of how well it can perform in restrictive environments, it doesn't do a good enough job of how much that mm/h2o metric actually affects performance. And in the grand scheme of things, the combination of the above points is all you really need to gauge the performance of the fan. That mm/h2o number would just be a reference point as to why a P12 is performing better than an S12A on a radiator. By itself, it doesn't mean much, but when benchmarked it really puts it into perspective, and that's what I feel we need in these fan reviews.

5. Temperatures don't matter
When I see youtube reviewers reviewing fans by putting them in a case or air cooler, I wonder how well they're actually controlling all of those variables. Do you know how much is going on inside a case, that ONE case to get those temperatures? It's too personalized, and honestly I think it's an unnecessary datapoint.

Again, I think an airflow:noise graph is all that's needed because the more heat you can dissipate from the heat exchanger, the better your temperatures will be. A drop in temperatures will be the effect, and the cause of that is the 40fpm increase in airflow at 30dba versus the other fan.

6. PWM and Voltage scaling matter
I don't see this being measured enough. Some fans don't scale linearly with PWM, and if you're trying to target a specific percentage range that gives you a good balance of noise and performance for your auto config, then having a linear rpm spread would be ideal instead of having the fan jump from 1300 to 1800rpm with just 15% PWM increase.

The same applies to voltage adjustment, or basically undervolting. Some fans exhibit clicking noises when undervolted or have high starting voltages that can make them unusable if you're trying to keep the noise down. Granted, anomalies like this aren't as common as they used to be, and I'm sure the engineers have tested these metrics, but I still think it's important to validate them regardless.

However, and this is a big however and likely the reason why this isn't tested more often, is the fact that motherboards nowadays aren't the same when adjusting fan speeds. Some fans can't even hit their advertised speeds because of motherboard fan control, or the PWM curve changes, or you have a limit on how low the starting voltage is. But I guess that's why you use purpose-built measuring equipment for it instead (read: expensive).

But yeah that's about it. You have VSG on thermalbench doing the nice airflow:noise charts as well as pwm/v scaling. VSG also does the fan reviews on TPU, but aren't as good as the ones on his website. Overclockers.com has some old fan roundups that have fans tested with a dust filter, air cooler and radiator, but lack noise normalization. SPCR has some nice measuring equipment, but the scope of their tests are small. A number of youtubers have some microphone shots of the fan's sound signature, though that's all you can really take from those reviews. None that I know of do smoke tests. Cooling technique has his machine and good sound clips as well as probably the best spectrometer readouts, though he only measures free air and the spectrometer results don't really define any quirks of the fan.

The point is, nobody has put all of them into one review, and you have people having to reach in the nether-regions of the internet trying to supplement what one review doesn't have, or just trying to find any quantitative data on a particular fan. And we all know that doesn't work due to different testing methodologies, QC variance, measuring equipment, and so on.

So, what do you guys think? Is it time for fan reviews to be held to a higher standard, to take the next step like frame-time measurements were for graphics cards? Are current reviews good enough for Joe Blow?

1680 V2 4.5GHz 1.33v | G.Skill Trident X 4X4GB 2400MHz | R4BE | Titan Xp + Morpheus II
Wasabi Mango UHD550 | Rosewill Rise | Thermalright LGM RT
Logitech G600 | G440 | Tt Meka G-Unit Cherry MX Black+Double O-Rings
JBL LSR308s | Temblor T10 | Custom headphones | TEAC-UD501+Passive Preamp | Emotiva BasX A-100
OCZ Vertex 4 256GB | 51TB DAS | SS Prime Ultra Titanium 1KW
airisom2 is offline  
Sponsored Links
Advertisement
 
post #2 of 72 (permalink) Old 06-21-2020, 03:57 AM
New to Overclock.net
 
doyll's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 26,642
Rep: 1696 (Unique: 938)
Static pressure rating is one of the more important specs manufacturers give us, because it directly reflects fan's ability to overcome resistance. I would rather have it at 100rpm intervals so we can see how

Like you said, problem is specs are generally only at full speed while most of the time our fans are at idle with even heavy load pushing them to maybe 75-80% with top 20-25% rarely used except in extremely hot weather with dirty filters and system at near 100% load on both CPU and GPU.

Other sites that do good fan testing are CoolingTechniques and Hardware.Fr. Both use data collected at 10-12 points from minimum rpm to full speed video with audio from startup to full speed. Only thing missing is airflow and noise levels through resistance like grill, filter, radiator create.

Phanteks Owners Club Ways to Better Cooling
i7 980 @ 3.55GHz =PH-TC14PE w/2x TY-143 fans =Crucial Ballistix 3x4GB =GA-X58A-UD5 =ASUS GTX580 DirectCU II =Enermax Modu84+ =Define R2 w/3x TY-140 case intake fans; all PWM controlled by CPU fan socketPhanteks Enthoo Primo MoBo Rampage III Extreme CPUi7 980X
@ 4.0GHz =R1 Ultimate w/2x TY-143 fans =Dominator GT 6x2GB =Rampage III Extreme =ASUS GTX580 DirectCU II =TX850 =Enthoo Primo w/ custom castor base

Last edited by doyll; 06-21-2020 at 10:48 AM.
doyll is offline  
post #3 of 72 (permalink) Old 06-21-2020, 04:20 AM
Shooting down fallacies
 
ciarlatano's Avatar
 
Join Date: Aug 2013
Location: NYC Metro
Posts: 7,847
Rep: 451 (Unique: 298)
@airisom2 - So....essentially you are saying we need Thermalbench back?

I agree with you 100%, excepting perhaps the particulars on the noise since it is so subjective.

But, all of it means nothing if users don't take the time to understand it. 99% of the users here look at SP ratings on the box and have absolutely no idea what they mean. Look at all of the people who have changed out Noctua fans to iPPC thinking they will be getting more static pressure at 1000 rpm because the box has a bigger number. It never clicks that that number is only representative of the maximum rpm. The multiple results in what you suggest could be mind melting for some. All I can think of is "Spinal Tap" and the vacant look while he says "this one goes to eleven".

Yeah....the little girl hentai avatar is really creeping me out....
ciarlatano is offline  
Sponsored Links
Advertisement
 
post #4 of 72 (permalink) Old 06-21-2020, 05:06 AM - Thread Starter
Never Finished
 
airisom2's Avatar
 
Join Date: Aug 2011
Location: Within the Milky Way
Posts: 2,212
Rep: 279 (Unique: 216)
Quote: Originally Posted by ciarlatano View Post
@airisom2 - So....essentially you are saying we need Thermalbench back?
naw man noise testing should be measured with the fan being hanged by a rubber band in a box covered in matress tooper foam.

But yeah TB did cover most of those points and already has a good arsenal of fans, but some aspects are still in the air.

1680 V2 4.5GHz 1.33v | G.Skill Trident X 4X4GB 2400MHz | R4BE | Titan Xp + Morpheus II
Wasabi Mango UHD550 | Rosewill Rise | Thermalright LGM RT
Logitech G600 | G440 | Tt Meka G-Unit Cherry MX Black+Double O-Rings
JBL LSR308s | Temblor T10 | Custom headphones | TEAC-UD501+Passive Preamp | Emotiva BasX A-100
OCZ Vertex 4 256GB | 51TB DAS | SS Prime Ultra Titanium 1KW

Last edited by airisom2; 06-21-2020 at 05:10 AM.
airisom2 is offline  
post #5 of 72 (permalink) Old 06-21-2020, 09:49 AM
New to Overclock.net
 
Memmento Mori's Avatar
 
Join Date: Nov 2016
Location: Katana Orgy...
Posts: 176
Rep: 2 (Unique: 2)
Quote: Originally Posted by ciarlatano View Post
@airisom2 - So....essentially you are saying we need Thermalbench back?
is there any "replacement" for it? (similar web?)

Memmento Mori is offline  
post #6 of 72 (permalink) Old 06-21-2020, 10:41 AM
New to Overclock.net
 
doyll's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 26,642
Rep: 1696 (Unique: 938)
Quote: Originally Posted by Memmento Mori View Post
is there any "replacement" for it? (similar web?)
I listed a couple in 2nd post this thread.

Phanteks Owners Club Ways to Better Cooling
i7 980 @ 3.55GHz =PH-TC14PE w/2x TY-143 fans =Crucial Ballistix 3x4GB =GA-X58A-UD5 =ASUS GTX580 DirectCU II =Enermax Modu84+ =Define R2 w/3x TY-140 case intake fans; all PWM controlled by CPU fan socketPhanteks Enthoo Primo MoBo Rampage III Extreme CPUi7 980X
@ 4.0GHz =R1 Ultimate w/2x TY-143 fans =Dominator GT 6x2GB =Rampage III Extreme =ASUS GTX580 DirectCU II =TX850 =Enthoo Primo w/ custom castor base
doyll is offline  
post #7 of 72 (permalink) Old 06-21-2020, 10:46 AM
Vandelay Industries
 
maltamonk's Avatar
 
Join Date: May 2015
Location: US/UK
Posts: 1,708
Rep: 83 (Unique: 62)
Missing one major factor......PRICE!
maltamonk is offline  
post #8 of 72 (permalink) Old 06-21-2020, 10:52 AM
New to Overclock.net
 
doyll's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 26,642
Rep: 1696 (Unique: 938)
Quote: Originally Posted by maltamonk View Post
Missing one major factor......PRICE!
Prices vary so much around our big blue marble it makes little to no difference. When we can't find data about how product performs price doesn't mean much.

Phanteks Owners Club Ways to Better Cooling
i7 980 @ 3.55GHz =PH-TC14PE w/2x TY-143 fans =Crucial Ballistix 3x4GB =GA-X58A-UD5 =ASUS GTX580 DirectCU II =Enermax Modu84+ =Define R2 w/3x TY-140 case intake fans; all PWM controlled by CPU fan socketPhanteks Enthoo Primo MoBo Rampage III Extreme CPUi7 980X
@ 4.0GHz =R1 Ultimate w/2x TY-143 fans =Dominator GT 6x2GB =Rampage III Extreme =ASUS GTX580 DirectCU II =TX850 =Enthoo Primo w/ custom castor base
doyll is offline  
post #9 of 72 (permalink) Old 06-21-2020, 11:04 AM
Intel in 2020 LOL
 
Join Date: Oct 2011
Posts: 1,128
Rep: 46 (Unique: 22)
Quote: Originally Posted by doyll View Post
Prices vary so much around our big blue marble it makes little to no difference. When we can't find data about how product performs price doesn't mean much.
At least keep in mind the MSRP or something. A $50 fan being better than a $10 fan shouldn't surprise anyone. On the other hand, if the $10 fan does 80% of what the $50 fan does, that means the people who overspend are straight up morons, just like everywhere else in life.

AMD Bliss
(15 items)
CPU
AMD Ryzen 5 3600 - 2019PGT
Motherboard
MSI X470 Gaming Plus Max - 1.0.0.6
RAM
2x8GB G.Skill Trident Z 3600C15 @3800 14-15-12-28-40 1T 1.52V
Hard Drive
Samsung 970 Evo Plus 250GB
Hard Drive
2x Samsung 860 Evo 500GB
Hard Drive
2x Crucial MX500 1TB
Power Supply
Super Flower Leadex II Gold 650W
Cooling
be quiet! Dark Rock 4
Case
Phanteks P350X
Monitor
Asus VG245HE
Keyboard
Logitech G710+
Mouse
Logitech G502 Hero
Mouse
Logitech G502 Proteus Core
Mousepad
Logitech G440 hardmat
Audio
Asus Cerberus
CPU
Intel Core i5 4460
Motherboard
Gigabyte B85M-D3H-A
GPU
Sapphire RX 580 Nitro+ 8GB
RAM
2x8GB Kingston HyperX Fury @1600 CL10
Hard Drive
Samsung 860 EVO 500GB
Hard Drive
WD Blue 1TB
Power Supply
Super Flower Golden King Pro SF-500P14PE 500W 80+ Platinum
Cooling
Intel Stock Cooler
Case
Zalman Z11+
Operating System
Windows 10 x64
Monitor
Asus VG245HE TN 1080p 75hz
Keyboard
Logitech G710+
Mouse
Logitech G502 Proteus Core
Mousepad
Logitech G440
Audio
Logitech X-540 5.1
CPU
Intel Core 2 Quad Q6600
Motherboard
Gigabyte P43-ES3G
GPU
Asus R9 270X Top 2GB
RAM
2x2048MB Kingston DDR2 800mhz
Hard Drive
Samsung 840 EVO 120GB
Hard Drive
WD Blue 1TB
Power Supply
Sirtec High-Power 80+ 400W
Cooling
Some 92mm cooler
Case
Zalman Z11+
Operating System
Win 7 Ultimate x64
Monitor
Several LG/BenQ 17" 1280x1024 75hz
Keyboard
Logitech G710+ MX Brown
Mouse
Logitech G502 Proteus Core
Mouse
Razer DeathAdder 2013
Mousepad
Logitech G440 hardmat
Mousepad
Razer Goliathus Speed WoT Edition
▲ hide details ▲

Last edited by rares495; 06-21-2020 at 11:14 AM.
rares495 is offline  
post #10 of 72 (permalink) Old 06-21-2020, 12:03 PM
Vandelay Industries
 
maltamonk's Avatar
 
Join Date: May 2015
Location: US/UK
Posts: 1,708
Rep: 83 (Unique: 62)
Quote: Originally Posted by doyll View Post
Prices vary so much around our big blue marble it makes little to no difference. When we can't find data about how product performs price doesn't mean much.
Price is one of the main considerations of any product sold. Varying prices just means that variable will change the outcome of the equation. By the same token some fans perform differently in different scenarios....again those variables need consideration. That's why we have SP and AF editions...those variables change, but are still important. And again since we are the consumers of these products price matters ofc unless we get them for free...then it doesn't matter unless we resell it/them.
maltamonk is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off