The QUIETEST Cooler I Have EVER Used!

The QUIETEST Cooler I Have EVER Used!


This may be the most insanely quiet CPU cooler
I have ever tested. Howdy howdy guys ponchato here, and today
we’re gonna take a look at the Ninja 5 from Scythe. Thanks to Scythe for sending this over for
review, and let’s get started. The 60 US dollar Ninja 5 is a very large,
like… really, very large dual fan tower cooler. With both fans installed, it comes out to
138mm wide and a very hefty 180mm long. That’s almost the entire length of a microATX
motherboard, not including the rear I/O connections. Despite its enormous footprint though, Scythe
made the smart choice of limiting the height to only 155mm, which means it will fit into
the vast majority of ATX and microATX cases. It’s also compatible with AM4, LGA 1151, and
LGA 2066 sockets. Scythe doesn’t give it a specific TDP rating
but I would estimate it to be about 200 watts. It has 6 heat pipes and allows for RAM up
to 55mm in height in both the front and rear, though that may be limited by how high you
can position the fans on this cooler in your case. The two fans are Scythe’s Kaze Flex 120mm
PWM, with an RPM range from 0 to 800 and a fluid dynamic bearing. The fans have rubber mounts on the corners
to isolate them and reduce noise and, in typical Scythe fashion, do not have any LEDs. I just want to emphasize again, this thing
is enormous, and uses all the space available to a CPU cooler. Installation is surprisingly easy for a cooler
this size. In fact I’d go so far as to say this is one
of the easiest installations I’ve performed. Scythe really nailed it with the AM4 socket,
and they include the funny looking long screwdriver you’ll need to mount the cooler, so you don’t
have to go digging for an extra long screwdriver yourself. Install starts with removing the AM4 mounting
brackets, leaving behind the stock backplate. Four plastic and rubber spacers get placed
over the backplate mounting holes, and the two universal brackets are screwed into the
backplate. Naturally, I put these on backwards first
because reading directions is for the weak. After orienting them the right way, a dab
of thermal paste goes on the center of the CPU, I took the protective cover off the base
of the cooler, and I sent it down on the processor. The cooler is held down by a cross bar to
the mounting brackets, and Scythe does remind you to alternate tightening them down so the
bar doesn’t get misaligned or bent. With the cooler mounted, the last step is
to mount the fans and plug them in with the included PWM Y-splitter cable. As you can see here I installed the rear fan
backwards, with it pushing air toward the front instead of pulling, but I corrected
that before recording data. I’m very good at installing things backwards. And now let’s go to the benchmarks. My test bench is a Ryzen 3 1200 overclocked
to 4.1GHz at 1.35V, the graphics card is a passively cooled GT 1030 from MSI, and thanks
to Seasonic for the 850W Focus Plus power supply. Both the graphics card and power supply can
run fanless, meaning the only sound coming from this system is from the CPU cooler itself. Load temperatures are taken with the CPU running
Prime95, and cooling results are reported as deltas; degrees above ambient temperature. First we’ll look at the idle results. These are the noise levels and temperature
deltas with no programs running and the fans at their minimum RPM. Since the Ninja 5’s minimum RPM is 0, this
cooler is completely silent at idle, something I really wish more companies would design
their coolers to do. Idle delta isn’t very relevant in the real
world, but for curiosity’s sake I include it here as well, and the Ninja 5 does run
a little warm. Note that “warm” here is about 34 degrees
Celsius, so like I said, not too relevant. Now we’ll look at the load results. These measurements are taken with Prime95
running and the fans at full speed to show the cooler’s upper limits. Since the Ninja 5 tops out at only 800 RPM,
it is quiet. Very quiet. In fact, it is by far the most quiet cooler
I’ve tested, more than 8 decibels lower than the next. At under 34dB, it is only barely audible when
sitting exposed on a desk next to you, but once it’s in a case or paired with a non-passive
GPU or power supply, in all likelihood you won’t be able to hear it at all. And it does this without sacrificing thermal
performance; only a 35 degree delta above ambient, corresponding to a CPU temperature
of about 55C in my studio. That’s impressive. Now my favorite graph, temperature vs noise. This is under load with the fan speed varied
from 100% down to the minimum speed the cooler can run without thermal throttling the processor. As you can see, the Ninja 5 in yellow is simply
quieter at any given temperature delta than any other cooler, with the gap only narrowing
at very low fan speeds where everything is quiet anyway. Now here’s the chart of cooler scores, a
combined rating of temperature and noise. You can think of this as a metric to compare
coolers running at their optimal speed, balanced between cooling performance and noise level. Owing to its insanely good acoustics, the
Ninja 5 takes the crown here with ease. It’s so significantly quieter than any other
cooler that, even though it only has upper-midrange thermal performance, the overall picture is
dang near perfect. Finally, cooler score per dollar to show price
to performance. As expected, the Ninja 5 falls in the middle
of the pack here. It is an extremely good cooler, and $60 isn’t
outrageously expensive, but it isn’t cheap either. If you’re focused on getting the absolute
best bang for your buck, less expensive coolers are almost universally a better choice, but
then you need to ask yourself: is it really worth saving $20 to miss out on dramatically
better noise or cooling performance? That’s a question only you can decide. And your wallet, I guess. Overall, the Ninja 5 is a fantastic cooler,
especially if you’re like me and you pay particular attention to how much noise your PC produces. Installation, though you’ll likely only need
to do it once, is very straightforward and fast – it took me just over 10 minutes, and
that’s only because I put the brackets on backwards to begin with. But the thing I love most about the Ninja
5 is its noise level. It is insanely, ridiculously quiet. 8 decibels quieter than the next closest cooler
I’ve tested, and more than 10 decibels quieter than AMD’s stock Wraith Stealth. The Ninja 5’s gigantic heat sink really does
work. The only potential downside here is RAM clearance,
but if you have standard or close to standard-height memory (which I pretty much always try to
get), it’s a non-issue. Oh and on a final note, Scythe’s manuals are
among the best, with very easy to understand instructions and illustrations. They do a really good job on those. The Ninja 5 goes for $60 and if you want to
pick one up for yourself, follow the link in the description. Hit subscribe and click the bell icon to get
notified of new videos as soon as they’re up. If you liked this video hit the like button,
if you want to see more hit subscribe, and I wanna hear from you: with CPU coolers, do
you care more about thermal performance or noise? Let me know in
the comments below! Thanks for watching, I hope I helped, and
I’ll see you in the next video.

42 thoughts to “The QUIETEST Cooler I Have EVER Used!”

  1. Wow the sound difference between the Cryorig A40 ultimate and the Sycthe is nearly 30dBA…

    I would definitely sacrifice 3-4C for that sound difference. I feel like sound generated is something that is overlooked in today’s pc building community as people only look towards performance.

  2. I've been seriously considering the Mugen 5 RevB for my 2700x. The Wrath Prism is so pretty but friggin loud. The GSkill Trident Z RGB RAM I have limits my options a bit since it's so tall.

  3. Thanks Ponchato – really great presentati
    on as usual. The quality of review and camera work of your stuff never ceases to amaze me. Very timely, this one, as I was about to buy a Mugen 5 Rev.B for a new build about to commence, based on your review of same. I was blown away to see the cooling and dba on this Ninja thing. Now I don't know what the hell to buy as this Ninja ticks even more boxes. I do however see 2 negatives of the Ninja over the Mugen 5 rev.B: 1) the size hides all the glory of my new ASUS X299 Prime II MoBo and a few little LEDs I will install – it looks like it will completely fill ALL the airspace in front of my gloriously visual new MoBo. 2) the WEIGHT of Ninja, compared with Mugen 5, is horrendous. I can only imagine over time what the effects that a tubby Ninja cooler hanging off the side of my precious new MoBo might be. Hairline cracks have been known to kill a motherboard dead and the weight and vibration (granted, minimal vibes with Ninja) may indeed take a toll over time. Motherboards were not designed for much more than the weight of a low-lying GPU at most, certainly not a hulking Ninja monster on the delicate CPU area. Surprising they don't make a rear MoBo support bracket like ASUS has on their X299 TUF board – THAT would negate ANY & ALL concerns re weight hanging off the board on one side. So, the Mugen 5 weighs a heap less and has very close dba and heat levels AND is visually the best looker of the 3 main Scythe coolers. I think I will go with the Mugen 5 Rev.B, on reflection, as all things considered it seems to be the sweet spot and best looker. It is also cut very high for 100% full RAM clearance – quite a good design feature. It even has plenty of space left around the edges to show off my Prime II. Just don't try to get a Mugen 5 rev.B here in Australia as the company does not return emails sent through their web page (4 sent over 4 weeks!) and of the two Australian distributors, one doesn't want to know them (little shop in Perth) and the other, New Egg, doesn't return emails either. Some companies, it seems, manage to stay in business despite themselves! I am now forced to buy through Amazon USA (God help me) and get it shipped to Australia at a full cost of what will probably be upwards of AU$150. Just unbelieveable and I wouldn't do it unless I was convinced the Mugen 5, of all others & brands, is the best air cooler for me. Based on the recent observed behaviour, the company doesn't deserve my business but it seems a great product so I'm not going to cut off my nose to spite my face. PLUS, Ponchato recommends it! That's enough to get it over the line for me. This guy knows his stuff.

  4. Good job there, man! Love the cooler and the video!
    I used to be a hardcore thermal performance > noise kind of guy a few years back, but as I age I tend to prefer quieter systems over ultimate thermal performance. I even tried running all my case fans on 5 volts but the airflow was so weak that my power supply fan would ramp up and be insanely loud during load. Back to 12V and everything stayed nice and relatively quiet.
    My current CPU cooler is an old but still very capable 3R System IceAge Prima Boss II. It's not very popular but it does a great job, and it comes with extra rubber pins for adding a second fan (not included) for push-pull. You don't see that often nowadays.

  5. Very good review, as always. I had the 212 Evo and I upgraded to the Noctua NH-D15 a few months ago. I must say that Noctua cooler is huge. I did measure out my case to make sure it would fit, but when I finally got it, I was blown away with how large it really is. Personally, I'm not too concerned with noise levels as I am with performance. I was looking at AIO cooler also, yet I was too paranoid about it leaking. I have received some good input from others not having any problems with their coolers. I think the next time I upgrade though, I plan on doing liquid cooling too see how it works out for me. Where was your wireless wrist strap?

  6. Great to see someone reviewing this cooler. It seems to have slipped under the radar, most people are going with the NH-D15, Dark Rock 4 Pro, or R1 Universal. The Ninja 5 can compete with them all easily, and it can fit in more cases than they can.

  7. I'd be interested to know how well this monster would perform with more powerful fans.
    I watched your Cryorig review a little while ago and was on the verge of buying it, but am now considering this beast and sticking on some spare fans I have (Corsair SP120 – 2350 RPM).
    With the stock coolers this thing matches the cooling performance of the legendary Noctua NH-D15, with other fans it could crush it!

  8. I kinda wish someone would test scythe coolers with pressure optimised fans like the Arctic P12-P120.
    I also wish Scythe stuff was more availlable, it was the shit when I bought a Mugen 2 and a Ninja 2 in 2008… Now it's almost impossible to track down despite being reviewed all over Youtube.
    Keep it Up ponchato

  9. I am always obsessed by CPU cooler noise, I think the best way to solve the problem is put your PC far away use a 5 meters HDMI wire connect your screen.

  10. There's no way you're getting idle temps that low on an air cooler. Do you live in a house size refrigerator?

  11. Great review, nice presentation and well thought, i really like the graphs and comparative you do, thanks for sharing it.

    I do have a question though, around 3:10 you present a graph comparing noise and temperature. IIRC the Mugen5 and Ninja5 use the same fan, if they don't, disregard my query, but if they are the same, almost always 1 fan will be quieter than 2 fans, at least driven at the same rpms, so i dont understand how the mugen5 reflecting 43dba while the ninja5 is showing 33.9. What i would have expected is the mugen5 to be quieter (1 fan) but slightly higher temps (ninja5 has more area of dissipation), unless you didn't tested the fans at the same rpms.

  12. Can someone confirm the numbers? This seems impossible – such an amazing thermal result and and even more amazing noise one? Wow

  13. Interesting that the fuma is 16db louder than the ninja at 100% fan speed, seeing as how they both use the same kaze flex fans I find it a bit strange. Can fin design alone really result in such a massive noise difference or did one have particularly loud or quiet fans I wonder.

  14. I need this. I just built a new system with a Ryzen 7 2700 X cpu and the AMD Prism cooler that came with it. Slick LED lights, but very loud under load. Too loud.

  15. It can not be inaudible in load. I've got scythe choten. It has the same scythe flex fan but with 1200 max rpm. And its definitely audible after 800 rpm. Yes its really quiet but not inaudible.

Leave a Reply

Your email address will not be published. Required fields are marked *