Noctua NH-C12P: A Top-Down Cooler Rises to the Top

Want to talk about one of the articles in SPCR? Here's the forum for you.
IMOM
Posts: 77
Joined: Sat Mar 27, 2004 10:55 pm
Location: CA, USA

wrong link on the Thanks to Noctua

Post by IMOM » Tue Jun 17, 2008 1:59 pm

Mike,
I clicked on the Our thanks to Noctua for the review sample link at the end of the review and it takes you to the scythe website. Just thought I let you.

Other question if this hasn't been asked... how well do you think it would perform in NSK2480/FUSION case? Thanks,

FartingBob
Patron of SPCR
Posts: 744
Joined: Tue Mar 04, 2008 4:05 am
Location: London
Contact:

Post by FartingBob » Tue Jun 17, 2008 2:25 pm

Modo wrote:
FartingBob wrote: 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.
Since when does separate packaging and separate logistics in general not cost you extra?
For a company the size of noctua (or bigger ones like TR) an extra box design would add very very little cost to each product, as they are used to producing different products and can handle that efficiently. It could appeal to the price conscious that already have fans and dont fancy spending $5-20 more for something that will perform worse than the fan they have now, and Heatsinks like most of TR's which come without fans would be more appealing to the masses that dont know what fans are good and bad.

andyb
Patron of SPCR
Posts: 3307
Joined: Wed Dec 15, 2004 12:00 pm
Location: Essex, England

Post by andyb » Tue Jun 17, 2008 3:53 pm

For a company the size of noctua (or bigger ones like TR) an extra box design would add very very little cost to each product, as they are used to producing different products and can handle that efficiently. It could appeal to the price conscious that already have fans and dont fancy spending $5-20 more for something that will perform worse than the fan they have now, and Heatsinks like most of TR's which come without fans would be more appealing to the masses that dont know what fans are good and bad.
Even better to sell the product without a fan, but to pass on a reduced cost (via the retail channels) of any Noctua fan bought at the same time.

This would make everything easier for the customer, and not make a lot of difference to the retailer, but most importantly Noctua would benefit.

Anyway, its an excelent product, shame about the price, the HDT-S1283 is half the price but comes with a crap fan (which I have never even plugged in BTW), if it was cheaper it would look a lot more attractive, if it was a lot cheaper because it didnt come with an overpriced fan then it would sell by the thousand and not the hundred.


Andy

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Tue Jun 17, 2008 9:27 pm

FartingBob wrote: For a company the size of noctua (or bigger ones like TR) an extra box design would add very very little cost to each product, as they are used to producing different products and can handle that efficiently.
Would you mind counting for me the products Noctua has on the market, compared to Scythe or Thermalright. No, really, count them...

charonme
Posts: 131
Joined: Tue Apr 11, 2006 4:17 am
Location: Slovakia

too high + better than xp120?

Post by charonme » Tue Jun 17, 2008 11:52 pm

three things:
1.) I'm disappointed that it's so high (but I like that it has somewhat wider fin spacing). We are in the age of htpcs and if someone wants a big HS, he would get a tower. Those who reach for the top-downs probably need them low I guess. At least I do, that's why I'm using the XP120 - haven't found a lower one at that performance (except shuriken, but I'm not sure about the comparison with xp120 yet).

2.) Why wasn't it tested in a completely passive mode and what about the K/W figures?

3.) How does it compare to the XP120 and scythe shuriken at low speeds/passive?
Last edited by charonme on Wed Jun 18, 2008 12:39 am, edited 1 time in total.

floffe
Posts: 497
Joined: Mon May 08, 2006 4:36 am
Location: Linköping, Sweden

Re: too high + better than xp120?

Post by floffe » Wed Jun 18, 2008 12:22 am

charonme wrote:2.) Why wasn't it tested in a completely passive mode and what about the K/W figures?
I can answer this one. No heatsink SPCR has ever tested has been able to cool their test CPU without some forced airflow. That includes the HR01+ that was reviewed recently, which is designed to be passive. The test bed has no fans nearby, so requires one to be used on the heatsink. Also a top-down cooler would be less likely to benefit from airflow from nearby fans compared to a tower heatsink, due to being at 90 degree angle to the airflow.

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Wed Jun 18, 2008 1:24 am

Actually, the problem of no system airflow is a big one only for tower heatsinks (and only in this test bed). The NH-C12P and other similar heatsinks can make their own airflow with convection, so it would be interesting to try it out. I think such tests would be very much appreciated by HTPC builders.

Kaleid
Posts: 254
Joined: Mon Oct 11, 2004 9:43 am
Location: Sweden

Post by Kaleid » Wed Jun 18, 2008 6:53 am

Too bad that it already comes with a fan. Many of us already have enough of them at home and would rather save a bit of money.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Wed Jun 18, 2008 6:55 am

Modo wrote:Actually, the problem of no system airflow is a big one only for tower heatsinks (and only in this test bed). The NH-C12P and other similar heatsinks can make their own airflow with convection, so it would be interesting to try it out. I think such tests would be very much appreciated by HTPC builders.
We've actually put the board up sideways in some passive HS tests in the past to simulate a tower case. Didn't matter, none of the HS in the past ever kept the CPU temp safe under our test load. The bottom line is that if there is no fan in the vicinity, none of the HS we've tested will keep the CPU cool enough. If you need a fan anyway, why not put one on the HS? It's actually most needed there -- the hot components around the board also need the airflow.

To really do passive cooling tests, we'd really have to put the board in a case and run it vertically or horizontally depending on the HS. There would have to be at least an exhaust fan... and the distance between that fan and the HS would determine to a large degree whether the HS would keep the CPU cool enough.

Sorry, I just don't see passive cooling tests to be worth the time and effort -- it's way too dependent on system and config details. If you're interested in running a HS without a fan directly on it, it's already obvious which ones are worth trying it with: The ones that perform very well with very low airflow, and which have very wide spacing between fins.

Besides, we run several single and multiple fan PCs in the lab that not only run cool enough -- one has done so for years 24/7 -- but are also inaudible. The old one mentioned measure around 15 dBA/1m. With 3 fans... admittedly, barely spinning.

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Wed Jun 18, 2008 7:19 am

MikeC wrote: We've actually put the board up sideways in some passive HS tests in the past to simulate a tower case. Didn't matter, none of the HS in the past ever kept the CPU temp safe under our test load.
I see. Thanks for the info. :) Didn't really think the problem was that bad. I guess a closed case is even more important than I thought.
MikeC wrote: If you need a fan anyway, why not put one on the HS? It's actually most needed there -- the hot components around the board also need the airflow.
Because in a HTPC, a fan often won't fit if the heatsink is taller than stock height. And that's true even in a relatively high HTPC case, like the Antec Fusion. This is why some testing with proxy airflow would be useful IMO (changed from no airflow, as per your statement about such situations).

pony-tail
Posts: 488
Joined: Sat Aug 23, 2003 4:39 pm
Location: Brisbane AU

Post by pony-tail » Thu Jun 19, 2008 4:12 am

I have not posted on this forum for a while !
I have 2 M2N32-SLI based machines one with a 5000+BE and one with a 6400+ both using TR ultra 90 coolers (which cool the cpu adequately ) I was looking at this cooler as a replacement for the U90s as they do not cool the northbridge and vrms and my northbridge temps get up into the high 60s .
Has anybody on this forum used one of these coolers on an Asus M2N32-sli board ? The macine are in gen1 Sonata cases so it should have enough space . It is just the clearance on the heatpipe northbridge cooler that I have concerns about .
I am currently using 40mm Sunon fans cable tied to the NB heatpipe cooler to keep it cool as a temporary solution (for the last year or so ) but the buzzing is annoying . I was hoping that this cooler would provide adequate cooling for the motherboard components because the TR unit does not .

osmium76
Posts: 24
Joined: Mon Nov 05, 2007 7:44 am
Location: Italy

Post by osmium76 » Fri Jun 20, 2008 1:10 am

I've just returned from a competition so I'm taking rest because I'm very tired, so I didn't read the full article and this thread but I red the sound performances of the Noctua P12 and in my opinion your sample should have problems because I've bought 3 of them and they are quieter than the SilenX iXtrema rated for 26 decibels... I tried Noctua P12s after red a review of an Italian site where the reviewers said the P12 model is quieter than the S12 model. I don't have any S12 fan but I can ensure the P12s are surely A LOT under 33 decibels (I have mounted 2 inside my case...). It would be very interesting ask Noctua to send other samples of this fan.

Bye bye!

mariosimas
Posts: 21
Joined: Wed Nov 02, 2005 3:26 am

Post by mariosimas » Tue Jun 24, 2008 11:35 am


MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Jun 24, 2008 12:03 pm

mariosimas wrote:interesting review from XBit Labs:

http://www.xbitlabs.com/articles/cooler ... -c12p.html
It's a well done review, but their focus is quite different from ours. They urge Noctua to make a high performance HS w/o regard to noise.

Also the results they obtained with the Thermalright SI-128 is way better (at all fan speeds) than what we got with ours, which didn't even come close to the Noctua's performance. Makes me wonder whether it's their sample, our sample, their testing or ours?

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Tue Jun 24, 2008 12:29 pm

MikeC wrote: Also the results they obtained with the Thermalright SI-128 is way better (at all fan speeds) than what we got with ours, which didn't even come close to the Noctua's performance. Makes me wonder whether it's their sample, our sample, their testing or ours?
Maybe it was the SI-128 SE, but they shortened the name for some weird reason?

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Jun 24, 2008 1:46 pm

Modo wrote:Maybe it was the SI-128 SE, but they shortened the name for some weird reason?
The two models look pretty similar except for the extra holes on the fins in the SE. I have a hard time believing they'd make so much difference.

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Tue Jun 24, 2008 9:51 pm

MikeC wrote: The two models look pretty similar except for the extra holes on the fins in the SE. I have a hard time believing they'd make so much difference.
I'll know for sure by the end of the year, when I try it my self. ;) For now, I only have the words of a SPCR member that made me suspect Xbitlabs referred to the new version.

mariosimas
Posts: 21
Joined: Wed Nov 02, 2005 3:26 am

Post by mariosimas » Thu Jun 26, 2008 12:15 pm

MikeC wrote:
mariosimas wrote:interesting review from XBit Labs:

http://www.xbitlabs.com/articles/cooler ... -c12p.html
It's a well done review, but their focus is quite different from ours. They urge Noctua to make a high performance HS w/o regard to noise.

Also the results they obtained with the Thermalright SI-128 is way better (at all fan speeds) than what we got with ours, which didn't even come close to the Noctua's performance. Makes me wonder whether it's their sample, our sample, their testing or ours?
do you have a review with the TR SI-128 SE ?

Gator
Posts: 16
Joined: Thu Jul 26, 2007 2:16 pm
Location: Gainesville, FL - USA

Post by Gator » Sun Aug 24, 2008 3:11 am

MikeC wrote:
mariosimas wrote:interesting review from XBit Labs:

http://www.xbitlabs.com/articles/cooler ... -c12p.html
It's a well done review, but their focus is quite different from ours. They urge Noctua to make a high performance HS w/o regard to noise.

Also the results they obtained with the Thermalright SI-128 is way better (at all fan speeds) than what we got with ours, which didn't even come close to the Noctua's performance. Makes me wonder whether it's their sample, our sample, their testing or ours?
I'd really like to see a three-way comparison between the Noctua NH-C12P, the Thermalright SI-128 SE, and the Scythe Mini-Ninja - as these are probably the heatsinks most likely to find their way into an HTPC case like the Antec Fusion / NSK-2400 / NSK-2480. It seems to me that the NH-C12P might provide better cooling to the motherboard VRM circuitry than either of the other two?

Although it was theorized earlier that the NH-C12P would be a good fit for the NSK2480 / Fusion cases, does anyone reading SPCR actually have such a config? I was particularly concerned with the possibility that the NH-C12P might actually overhang the right hand end of the motherboard enough to touch the right-hand 120mm case fans in the Fusion series cases? What sort of clearances exist on that side?

nici
Posts: 3011
Joined: Thu Dec 16, 2004 8:49 am
Location: Suomi Finland Perkele

Post by nici » Sun Aug 24, 2008 3:27 am

That most likely depends on the motherboard used as the CPU socket placement varies slightly. I'll post my setup in the gallery once i put it together, OrigenAE S16V, noctua C12P and P5Q mobo. But it really depends on where the socket is. Id does not go over the mobo on the P5Q.

Modo
Posts: 486
Joined: Wed Apr 16, 2008 3:32 am
Location: Poland

Post by Modo » Sun Aug 24, 2008 5:13 am

Gator wrote:What sort of clearances exist on that side?
The clearance in a NSK 2480 is enough for a Ninja Mini, and its fin layout provides very good airflow. The top clearance in this case doesn't allow enough room for a fan on a SI-128 SE, so that heatsink can't be used in its optimal configuration (fins are close, so it needs a fan). I believe you can turn the Noctua in any direction on its mounting frame, and it is lower than the SI-128 SE, so clearance should not be a problem. It would be interesting to find out whether the good airflow provided by a Ninja Mini in the Fusion will work better than the overall efficiency of the Noctua.

Gator
Posts: 16
Joined: Thu Jul 26, 2007 2:16 pm
Location: Gainesville, FL - USA

Post by Gator » Sun Aug 24, 2008 5:16 am

nici wrote:That most likely depends on the motherboard used as the CPU socket placement varies slightly. I'll post my setup in the gallery once i put it together, OrigenAE S16V, noctua C12P and P5Q mobo. But it really depends on where the socket is. Id does not go over the mobo on the P5Q.
Thanks, glad to hear that. Hopefully this holds true for most of the ASUS P5Q-xxx line, (I am considering a P5Q-EM at present)

Another question comes to mind about the NH-C12P, ...it looks like one can choose to have the side with the heatpipes face in any direction... so one may have a choice of what side faces towards the motherboard's DIMM slots.... from first glance it looks like the heatpipe side might interfere with taller DIMM's (like Corsair's "DHX" aka "Dominator" line), and the opposite side has that huge "overhang" of fins.... but how much vertical space is there "under" that overhang? Enough for the "DHX" type DIMMs?

(successful builds without heartburn are all about researching "interference" issues up front) :D

nici
Posts: 3011
Joined: Thu Dec 16, 2004 8:49 am
Location: Suomi Finland Perkele

Post by nici » Sun Aug 24, 2008 6:26 am

I very much doubt it has room for "Dominator" type memory if the slots are under the heatpipes. Maybe on the side of the heatsink. Also, on Intel boards it is indeed possible to choose the orientation of the heatsink. Not on AMD boards though, only two ways, and which way that is depends on the orientation of the socket.

Gator
Posts: 16
Joined: Thu Jul 26, 2007 2:16 pm
Location: Gainesville, FL - USA

Post by Gator » Thu Aug 28, 2008 6:39 pm

Some followup to my earlier post, I sent an email query to Noctua and received the following diagram in response to my query (Thank you Noctua!) :)

Image

Which implies that if the NH-C12P is mounted with the curved heatpipe side towards the DIMM slots, it may conflict with nearby DIMM's with only 30mm clearance. However, it seems likely that if mounted the other way around with the curved heatpipes towards the I/O panel cutout and the taller overhang above the DIMM slots, the 53mm clearance should be enough to clear most DIMM's, albeit maybe not all brands.

When used with a 120mm fan in cases like the Antec NSK-24xx/Fusion-series this "downdraft" cooler might have significant advantages with regards to cooling the motherboard circuits and memory DIMM's versus horizontal-flow tower-types like the Scythe "Ninja" family.

nici
Posts: 3011
Joined: Thu Dec 16, 2004 8:49 am
Location: Suomi Finland Perkele

Post by nici » Fri Aug 29, 2008 5:29 am

The memory should also fit on the side of the heatsink, at least on the P5Q there´s a few mm of space between the edge of the heatsink and the memory slots. If it fits probably depends on how wide the heatsinks on the memory are.

Rez
Posts: 8
Joined: Mon Jan 05, 2009 9:13 pm
Location: Europe

Will it fit? Will it cool?

Post by Rez » Mon Jan 05, 2009 9:26 pm

Hi, I'm new here, and sorry for bumping a relatively old thread, but I urgently need a new pc and am looking at a cpu cooler to order with it.

Would this cooler be of any use in a tower case with no side openings? I'm looking at a cooler for the Lian Li PC-8 (120mm intake and exhaust fan; 21 cm wide vs 11,4 cm for the cooler + fan).

It seems weird to put a cooler there with its airflow and fin orientation perpendicular to the case airflow... on the other hand the clearance should be enough, I think? I just wonder where the fan would have to get air from.

Thanks!

Post Reply