Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!


Discussion on Article:
Noctua NH-D14 CPU Cooler: One Step Ahead

Started by: halfwaythere | Date 11/29/09 03:25:23 AM
Comments: 89 | Last Comment:  11/29/16 08:56:16 AM

Expand all threads | Collapse all threads


I'm gonna be honest with you guys, the test is alright. The only reason its not great is because your sample of the IFX is unusually good and the fact that you considered it the best heatsink until now.

Any user replacing a regular IFX, because yours is pretty special, with a Noctua will get greater differences between the two with similar fan setups.

I don't think your conclusion is fair either. The Noctua is better than the IFX in every respect aside from sheer size: the mounting kit is almost perfect, the fans are top notch, the build quality is very good and the base is perfect. Your comment about the quality of the base is unjustified because that small bump is there for a reason: to assure that the best pressure is where it matters, above the cpu cores. This was confirmed by the Noctua staff and its not an assumption.
1 0 [Posted by: halfwaythere  | Date: 11/29/09 03:25:23 AM]
- collapse thread

No, it's because our IFX-14's are lapped (or with "evened-out bases" ) . After I lapped mine, I saw a 6-7 degree improvement when running LinX (in exactly same ambient temp). I also increased the backplate mounting pressure, which probably did the 4 degrees of difference.. by the way, the LGA-1366 mounting bracket is pretty good because once installed, one does not have to remove the motherboard to change CPU's.

Some have reported a 3-5 degree improvement after lapping their TRUE's or IFX-14'es.

They are just great heatsinks with rather thin fins (rather dense, but not too dense array). It's manufacturing at its excellence. Yes, Thermalright does make the base a little convex for a reason, too, but it's for smaller cores. It's always better to be slightly convex than slightly concave. Needless to say, a slightly concave base will fail for obvious reasons.

Also, the IFX-14 has four 8mm heatpipes, which are better than six 6mm heatpipes due to greater internal volume for absorbing and transferring heat. Less heatpipes also allow for better airflow, which cools the back of the fins.

The fins on the IFX-14 still look thinner, therefore, each cm^2 of surface area is doing a better job. The NH-D14 is just bigger with more surface area anyways. So, it makes sense to me that the NH-D14 does 2 degrees better than an IFX-14. 120mm to 140mm is still better, but not exponentially better, especially with those longer heatpipes that become less efficient outwards.

Asus Triton 88 is the one that just lost to the NH-D14, after having recently taken the heavyweight belt from the IFX-14.
0 0 [Posted by: Bo_Fox  | Date: 12/01/09 12:11:34 AM]

Why use a lapped heatsink then?
That is not going to give a good representation for most users.
If you are going to use one lapped then lap the other but better yet, use them both stock.
If thermalright sells then at such high prices and do not care to deliver a flat surface then that is THEIR problem.
Just a thought.
I am getting the D14 next week...any ideas of what type of performance increase over my current Mugen 2 I should expect?
I am thinking 5-7C
0 0 [Posted by: shaolin95  | Date: 12/01/09 10:28:39 AM]
- collapse thread

I am thinking 5-7C

Yes, I think soo too.
0 0 [Posted by: Jordan  | Date: 12/01/09 12:46:11 PM]

great aircooler ! thermelright got beaten and thats impressive!
0 0 [Posted by: 3Dkiller  | Date: 12/01/09 10:59:34 AM]

I too have an IFX-14 CPU cooler which is lapped and I use Coollaboratory Liquid Pro as thermal compound. When overclocking my i7 920 C0 to 4.1 GHz (BCLK 195 with turbo enabled and HT) - similar to yours - with same voltage in BIOS, my temperatures are even better. I use 3 fans in an Antec 1200 case, so I do not think that your IFX-14 is better than most, but in fact as you state, because its lapped. My core temperatures (Everest Ultima read)are when running Prime95 69, 65, 66 and 67 degrees celcius. It just so happens the IFX-14 is an awesome cooler, but so it that Noctua and an much better deal. Picture of temperature linked http://i342.photobucket.c...ums/o405/Grnkjr0/xbit.jpg
0 0 [Posted by: Grnkjr0  | Date: 12/03/09 01:17:36 PM]
- collapse thread

(Everest Ultima read)are when running Prime95 69, 65, 66 and 67 degrees celcius.

Try Linpack 64-bit
0 0 [Posted by: Jordan  | Date: 12/03/09 09:18:33 PM]
Next time maybe you can post a watchable screenshot. Testing inside a case in not relevant because its not something you can standardize. Also the fans you are using, the ambient temperature are other important factors. Liquid pro is also a thermal compound not very common but a few degrees better than a regular paste. I can guarantee you that the Noctua will be at least 4 degrees better than your IFX with the same fan and thermal compound setup.
0 0 [Posted by: halfwaythere  | Date: 12/05/09 04:20:56 AM]

I am interested to know all the Bios changes to make the 920 stable @ 4Ghz in this test.
0 0 [Posted by: mar1  | Date: 12/07/09 04:40:53 PM]

Really only one spec matters.


When will you put this in your tests?

Otherwise it is a waste of time.
0 0 [Posted by: Stuart  | Date: 12/11/09 08:52:56 PM]


Back to the Article

Add your Comment