The "different sound" is also something to watch out for in other 120mm Noctua fans. It may not be very loud, but I personally find it irritating. The fancy fan design (pun intended) has some problems, unfortunately.

Good to know the heatsink is so efficient.

P.s. On the second page, you doubled up in "punched notches punched."

Thanks for the excellent review. Although I'm a regular reader of SPCR articles, it's first time I'm posting.

However I wonder if something was wrong with the specific fan. Despite I don't like Noctua P12s (have a few of them laid in a drawer), I never noticed the "aircraft flying overhead" noise. Maybe is only audible when fan blows over this heatsink...

Looks very solid. It's too bad they don't sell it without the fan, wouldn't mind a bit more reasonable price (and selection of fans).

This must be fairly new? Only two places I can find it are Xoxide ($70, out of stock) and pcToys($90, in stock :p).

Excellent review, thanks


I’ve been looking forward to this review with such great anticipation that it now feels like Christmas (not to put that image in your head during the summertime and everything, duh Walle) so it was nice to see that their latest creation turned out so nicely, something which I thought it would do yet you never really know before you chaps put the product thru its paces or you purchase it yourself and since I’ve been planning on using this cooler for a HTPC build (not mine though) it feels great, more so since it now appears that I can begin ordering parts and start building, kept the build on hold until a review was available. But, not nitpicking here Mike, I’m missing some findings, or thoughts around its effectiveness to actually cool the circuitry around the socket, I’ve read the review twice trying to make certain that I didn’t miss a possible mentioning but I couldn’t find one so I’m sticking my neck out there, did you make a quick check Mike? if you did I would very much appreciate some feedback.


“ We interrupted testing twice, believing that a plane was in the vicinity â€

Excellent review, thanks.

I started using Zipang, but soon found I had to replace the fan at extra expense. The $69.95 retail price puts the Noctua price-point modestly higher than Zipang + Fan + LGA-775 kit.

The non-parallel mounting screw issue would really bother me after paying $70 for a premium unit. There is no excuse for sloppy work at that price point. I figure Noctua will get this corrected in short order.

The wide fin spacing should allow for more breeze on the VRM and NB components. For me, that is vitally important. The tight fins of the other down-drafters do not allow much breeze with a slower fan speed.

It does seem odd that the measured noise (32dB) would be over 12dB above that of the rating. Looking at the review of the NF-S12-1200, though, that was measured 8dB above the rating. Perhaps Noctua uses some questionable methods of rating their fans?

Looks like this is the best performing (air) cooler to fit inside Antec NSK2400/Fusion.

I'd like to see it compared to Enzotech Extreme-X. In my own tests, Enzotech was the only "down blower" on par with some of the best tower heatsinks.

Great article. I've found very interesting the fan comparison (maybe even more than the heatsinks comparison itself )

There is a typo through all the article: The noctua fan model is NF-P12 and not NH-P12 (I think that NF stands for Noctua Fan, and NH stands for Noctua Heatsink).

Manufacturer's noise ratings, when done by the engineering dept. rather than marketing, are performed in rare and expensive anechoic chambers.

Lacking a couple mil, Mike's measurements are made in a "live" (normal) room, meaning there are echos and that increases the sound level. I'm sure Mike would appreciate any contribution of an anechoic chamber in the Vancouver CA area should anyone find this a matter of concern.

Thing is, the measurements of the Nexus fans were spot on. The Scythe fans reviewed appear to be rated below the ambient noise level when tested, so it's understandable that they would test higher than advertised. Furthermore, if this Noctua review was done in the past week or so, it appears Mike was able to drastically reduce the ambient levels in his lab.

I just think it's odd. I'm not trying to say Noctua is pulling a fast one on us. I think everyone here expects to replace stock HS fans anyway.

The ambient levels did not decrease, but our new SLM resolves down way below the old B&K SLM. The latter seemed to have an internal noise limit of 15 dBA; with the new one, we've seen as low as 12 dBA in the lab so far. Yet when a fan measures 20 dBA with one, it also does so with the other. Which confirms that the new SLM has less internal noise, and so is able to show lower levels when they exist.

I have to caution about the negativity regarding the stock fan. When it is undervolted with the supplied 7V adapter, it's as quiet as the Nexus and cools perfectly well, so... The negative comments in the review pertained more to the overly high airflow capability of the fan, which didn't do any good for cooling anyway -- at least not for our test CPU.

Scythe offers five models of the Slipstream, 500RPM to 1900RPM nominal. S-Flex now has 4 models, 800 to 1900. Noctua offers a munificent one model, throwing in a coupla resistors to provide 3 total nominal RPMs at 12V. Noctua could do worse... but they could do better, too.

The decision to offer only a single fan model was probably due to Noctua's sales projections. I doubt they expect the new fan to be a big seller.

Well... The only difference is that Scythe puts the resistors inside the fans. The way Noctua does it is actually better in terms of seller inventory and user convenience.

From the Article (unbolded in article)




It's 15, right?

Corrected -- it's 15.

In my system (a TT Armor case with the 250 mm lateral fan) this cooler is

shoulder to shoulder with the TRUE, using the Noctua fan NH-P12.

With a Tamb= 22ºC and E6850 @ 3700 MHz @ 1,36 V (CPUZ) (1,42 V in BIOS)

I have a TLoad (Orthos Small FFt´s)= 52ºC/52ºC (reported by RealTemp 2.63) and a Tidle= 28ºC/28ºC

Hi mariosimas,

I’ll need to ask you what your take is on the coolers ability to cool the circuitry around the socket.

Thanks.

Wow, the P12 fan really isin't as good as i thought. Can you review the fan like you've done with so many other fans, Mike? Would really appreciate it.

It really wasn't said that it isn't a good fan. It's just even more powerful than one would need. With the adaptors, I don't think that is a problem.


I'm either going to buy a Xigmatek 1283 + Antec P182 for absolute silence or go for a CM590 + 140mm yate loons all over + this Noctua for relative silence and lots of airflow to keep my overclocked-all-day-long-folding@home-components cold enough.

So I was glad to see this review

What I can say to you is that the sensor reporting MB temp in Everest

(I have a Asus P5N-T Deluxe) reports a diference of 3ºC using the Noctua

comparing with the TRUE.The Noctua being lower. This sensor must be closer

to the socket.

Thanks for your help mariosimas, much appreciated.

" LGA775 mounting arms attached. They could also have been mounted differently, across the path shown by the red lines, which rotates the heatsink 90 degrees. Which is the better way? It just depends on the details of your board. With AM2, the heatsink can only be mounted one way, dictated by the orientation of the heatsink retention bracket mounting on the board. "

I think it's been proven that heatsinks works best in when the pipes goes horizontaly rather than vertically?

Hello,

I think the fact that both this HS and the Thermalright HR-1 Plus both have vaulted to the top of their respective classes, shows that (among other things) soldering the heatpipes to the base and to the fins makes a big difference. Contrast this to the Ninja Rev B and Copper...

Actually, I don't think it is going to cost a couple mil. Mike's plans to address this is already underway.

Very nice review, koodos, etc. (obligatory but still genuine compliments)

Like some of the others, I'm really more intrigued by the fan than by the heatsink - a higher pressure but still quiet fan has the potential to improve the performance of some of those tower heatsinks with closely spaced fins. In particular, it might be nice to see how the Thermalright Ultra 120, Xigmatech, and Noctua's own NH-U12 (which comes packaged with the new fan now) fare with the Noctua fan versus the Nexus.

Would there be a significant difference here? It's probably not worth the effort if it's only one or two degrees, but I'd still like to know.

Nice heatsink, with a high quality fan thrown in as well!
In fact i would prefer them (and all others) to have 2 versions for sale for all heatsinks with one lacking a fan. Nobody does this, and yet it wouldnt really cost them extra.
At the price though it isnt cheap.

The NH-U12P does quite a bit better than the NH-U12F. It was reviewed by AnandTech a while ago, and it did good compared to the results of their earlier test of the NH-U12F. Note that they ran both fans at stock speeds, and they call that silent.


Since when does separate packaging and separate logistics in general not cost you extra?

I think the difference, if there are any, would only show up if you choked the airflow to the case or heatsink, or with very tightly spaced fins. All of these instances are anathemas to SPCR, so I doubt we'll explore it.