Cooling Scythe KazeFlex low-profile fan review.

3lfk1ng

King of Cable Management
Original poster
SFFn Staff
Bronze Supporter
Jun 3, 2016
906
1,713
www.reihengaming.com
Huge thanks to Scythe for sending us review samples of their KazeFlex fan lineup. Please let us know your thoughts below!

Scythe, the heavy hitting underdog of the HSF world, has once again hit us with another surprise entry into the SFF market with their low profile 120mm fan linup. At just 17mm thick, it will be interesting to see how they compare to the competition.

Read the review here
 

Kooki

SFF Lingo Aficionado
Mar 30, 2016
129
56
Excellent review! It even includes sound samples.

I see you included a Voltage x RPM data.
Did you perhaps also measure the lowest voltage need to start each fan?
 

CountNoctua

(no relation)
Jul 11, 2019
214
263
Impressive results, and I like the sound signature. Might be the best overall option for slim fans. I have to see if this would fit in my PSU, would be a good replacement if it does.
 

blubblob

SFF Lingo Aficionado
Jul 26, 2016
104
127
I love my Big Shuriken 2B to bits, so please don't take this as a diss on Scythe.
But I would be careful with a conclusion, from a few hours out of the box experience, that Scythe fans are on the same quality level as Noctua fans.

Noctua fans have in my experience never been the absolute quietest or best performing fans but the best well-ageing ones. My nearly 4 years old NF-A15 (with >2 years of warranty to go!) still sounds just like the one I bought just a few month ago, while I had to swap the (very impressive when it was new) Scythe Slipstream Slim less than 2 years after I got it because it sounded like it was grating plastic.

The Kaze-Flex Slim seem to have a Fluid Dynamic Bearing, which is a clear upgrade from the Sleeve Bearing of the Slipstream Slim, but that does not necessarily guarantee a longer lifetime. Especially in start-stop conditions (fanstop, semi-passive, etc. pp.) that are common nowadays, FDBs are unstable and unlubricated until they are up to speed.

If google translates the already available Japanese listing of the Scythe Kaze-Flex Slim correctly they still only come with a disappointing 1 year warranty (and a completely useless MTBF value at 25°C).

Again: The fans might be awesome. I'm not saying they are guaranteed to fail. (Especially If the plan is on changing the build just six month later anyway ;D ) Just be careful with jumping to conclusion on fan noise with very fresh fans.


Sidenote:
The review leaves me a bit confused regarding how you control the fans. Aren't they all PWM fans? I was under the impression that controlling PWM fans via voltage had an impact on the acoustical profile and performance of a fan (aside from obviously higher minimum starting RPMs). Some (bad) PWM circuits click for example.
 

Testifier

Average Stuffer
Oct 16, 2017
55
103
Well, we have always known that Scythe is comparable to the like of Noctua. The problem is mainly due to their packaging aesthetic IMHO. In a way, it's even worse than Noctua color. I mean you can literally walk through a PC hardware store's isle for fans and thought that these are some cheap-ass Chinese fans which aren't worth your time, meanwhile your attention is being directed toward the boxes of Noctua, Corsair, CoolerMaster, NZXT,...
 

3lfk1ng

King of Cable Management
Original poster
SFFn Staff
Bronze Supporter
Jun 3, 2016
906
1,713
www.reihengaming.com
Thank you for the replies everyone.

@blubblob
All great points.

As much as Noctua may want to advertise their proprietary SSO/SSO2 tech, ultimately, the design and functionality is exactly the same as what's used in an FDB, just higher quality bearings/magnets/plastics I suppose. Overall, the Scythe is rated for 14 years worth of spinning vs Noctua's 17. At that point, when comparing these two designs, the 1v2 year warranty will matter very little and what will matter here is noise and performance for the maybe 2-3 builds that you will likely use these fans in, before some other hot new fan get's buzzworthy praise on the web.

FDB based fans are terrific alternatives to the more expensive Noctua's(which I use in all my builds) in that the bearings do last an extremely long time, without issue, and will produce the same noise profile until the day they die. For those that care about price: performance ratio, while maintaining a near-silent operation, they just cannot be beat. For those that want the Chiron, no matter the cost, look to the Noctua lineup for the best performance -$30 for a "fan" is a tough pill to swallow for most. I still have a set of Scythe S-Flex fans that use Sony FDB's and they still work and sound great to this day. Scythe's S-Flex series were considered top-tier fans long before Noctua ultimately took the throne away from them. Meanwhile, my old Bitfenix Specre Pro fans are really starting to show their wear after all these years (which where also really well reviewed at some point in history).

The fans were all controlled using a Lamptron FC5 to ensure constant current output. The Lamptron was NOT powered by the Molex from a PC but instead plugged into a Coolerguys 100-240v AC 12v 4pin Molex 2A power adapter to help ensure a more stable power delivery. The fans where tested apples to apples for the first three graphs to ensure an accurate comparison of each profile. Any and all "whine" produced wouldn't be picked up at 30", just the sound of the blades chopping/buffeting through the air (which I had intentionally set to blow away from the MIC in the audio captures). The performance capture (the final graph) was performed using the CPU fan header on the motherboard, using the default fan-profile setting in BIOS, in order to produce real-world test results at 100% load sustained. Hope this helps.


@fabio
These are all grey market fans at this time. I will reach out to Jerry and see if they have any plans to bring them to other retail channels elsewhere.
 
Last edited:

fabio

Shrink Ray Wielder
Apr 6, 2016
1,885
4,325
Thank you for the replies everyone.

@blubblob
All great points.

As much as Noctua may want to advertise their proprietary SSO/SSO2 tech, ultimately, the design and functionality is exactly the same as what's used in an FDB, just higher quality bearings/magnets/plastics I suppose. Overall, the Scythe is rated for 14 years worth of spinning vs Noctua's 17. At that point, when comparing these two designs, the 1v2 year warranty will matter very little and what will matter here is noise and performance for the maybe 2-3 builds that you will likely use these fans in, before some other hot new fan get's buzzworthy praise on the web.

FDB based fans are terrific alternatives to the more expensive Noctua's(which I use in all my builds) in that the bearings do last an extremely long time, without issue, and will produce the same noise profile until the day they die. For those that care about price: performance ratio, while maintaining a near-silent operation, they just cannot be beat. For those that want the Chiron, no matter the cost, look to the Noctua lineup for the best performance -$30 for a "fan" is a tough pill to swallow for most. I still have a set of Scythe S-Flex fans that use Sony FDB's and they still work and sound great to this day. Scythe's S-Flex series were considered top-tier fans long before Noctua ultimately took the throne away from them. Meanwhile, my old Bitfenix Specre Pro fans are really starting to show their wear after all these years (which where also really well reviewed at some point in history).

The fans were all controlled using a Lamptron FC5 to ensure constant current output. The Lamptron was NOT powered by the Molex from a PC but instead plugged into a Coolerguys 100-240v AC 12v 4pin Molex 2A power adapter to help ensure a more stable power delivery. The fans where tested apples to apples for the first three graphs to ensure an accurate comparison of each profile. Any and all "whine" produced wouldn't be picked up at 30", just the sound of the blades chopping/buffeting through the air (which I had intentionally set to blow away from the MIC in the audio captures). The performance capture (the final graph) was performed using the CPU fan header on the motherboard, using the default fan-profile setting in BIOS, in order to produce real-world test results at 100% load sustained. Hope this helps.


@fabio
These are all grey market fans at this time. I will reach out to Jerry and see if they have any plans to bring them to other retail channels elsewhere.
Thank you, I really appreciate it! I was looking for those since the release of the BS3!
 

blubblob

SFF Lingo Aficionado
Jul 26, 2016
104
127
I don't want to be to nitpicky, but...
As much as Noctua may want to advertise their proprietary SSO/SSO2 tech, ultimately, the design and functionality is exactly the same as what's used in an FDB, just higher quality bearings/magnets/plastics I suppose.
As far as I know FDBs generally don't have a spinup-stabilizing magnet at all. The (also proprietary :D ) Matsushita FDBs Scythe is using are design wise about as far removed from a SSO2 bearing as they are from a sleeve bearing.

Overall, the Scythe is rated for 14 years worth of spinning vs Noctua's 17. At that point, when comparing these two designs, the 1v2 year warranty will matter very little and what will matter here is noise and performance for the maybe 2-3 builds that you will likely use these fans in, before some other hot new fan get's buzzworthy praise on the web.
The rating of the Scythe is for operation at 25°C - no fan hub directly on a 40+°C heatsink inside a case, with its own motor generating additional heat, will operate at that temperature. The Noctua rating on the other hand is for a temp range of -10°C to +70°C. They are in no way comparable. It's also 1 vs. 6 years warranty, not 1 vs. 2.
The MTTF is generally a bad indicator for fans in the first place. A fan might be screeching like a banshee after 12 month, as long as it won't stop turning for another 13 years it's within spec. That it won't matter to most people is still true though. :)

The performance capture (the final graph) was performed using the CPU fan header on the motherboard, using the default fan-profile setting in BIOS, in order to produce real-world test results at 100% load sustained. Hope this helps.
Under 100% sustained load, were all fans maxed out? I'm trying to figure out if it there is a way to at least somewhat match the decibel measurements with temperature values. Obviously the flow impedance of the heatsink will change the noise considerable too, but it would be at least an indicator.
At the moment one fan might be spinning at 60% while another one is maxed out just because the default fan curve is set to ramp the fan from 50% to 100% between 70°C and 80°C.
 
  • Like
Reactions: Kooki

rfarmer

Spatial Philosopher
Jul 7, 2017
2,588
2,702
Thank you, I really appreciate it! I was looking for those since the release of the BS3!
These are the ones I bought off of ebay, I realize now they have a different product number than the ones reviewed here but seem to be the same as the ones on the BS3.
 

Poblopuablo

King of Cable Management
Jan 14, 2018
816
465
Don't mean to derail from the topic, but @rfarmer can you review the sf120m vs the nf-a12x12 pwm. Both are ~30, and the new sf120m (released yesterday I believe) have better specs then the noctua A12x25, but I wonder how the compare... It would be awesome to see!!

Thanks for the review, I'll definitely be keeping these in my line of sight for future slim 120mm fans!!
 
  • Like
Reactions: CountNoctua