Test Bed:
Processor: Intel C2D E6700 L628A303 @ 3.2GHz @ 1.4 Vcore
Mobo: DFI Dark Infinity P965s
Memory: 2 x 1GB Team Xtreem Micron D9GMH DDR2 Special OPB Edition
Graphic Card: XFX 6600 GT
PSU: FSP Fortron Epsilon FX700-GLN PSU
HDD: Maxtor 20GB IDE 5400 RPM Hard Drive

Contenders:
1. Noctua NH-U12F
2. TR Ultra-120 eXtreme
3. Xigmatek HDT S1283

Software Used:
Coretemp 0.94
Orthos Beta
Windows XP SP2

Test Method:
1. This time I have gotten myself a temperature probe to measure the ambient temperature and we found out that the ambient actually fluctuating a lot. The result will include the ambient temperature as well for better judgment. The result might not be 100% accurate as it might be affected by the mounting method, temperature fluctuations and user error.

2. The Fan speed is kept at the speed of 1600 rpm as the main objective of this test is to find out the best heatsink that can be operate in silent mode.

3. The Idle Temperature is measured after the system left open for 5 minutes and the Load Temperature is measured after 10 minutes of Orthos Stress. Coretemp 0.94 was used to get the value of the temperature for each of the criteria, idle and load.

4. This test will be a open bench like what we did on the first roundup, TIM that is used is Arctic Ceramique also as it doesn't need any curing duration. This time the amount of TIM that I applied will be slightly more as many forumer mention that the amount that I applied is too little.

5. I have slightly modded the mouunting mechanism, its the same as the water cooling mounting mechanism.



This post has been edited by sup3rfly: Aug 16 2007, 10:52 AM

Noctua NH-U12F

The U12 is built from 38 nickel plated aluminum fins and 4 pieces of U-shaped 6mm heatpipes, the copper base is machine lapped by the manufacturer for better contact with the processor. Check out the pic below to see how well is the pressure applied



This post has been edited by sup3rfly: Aug 16 2007, 11:30 AM

TR Ultra-120 eXtreme

The well known TR Ultra 120 eXtreme made a come back with a slightly different mounting mechanism and added more heatpipes into the heatsink. With the extra heatpipes the heatsink became very heavy and I really like it a lot



This post has been edited by sup3rfly: Aug 16 2007, 11:34 AM

Xigmatek HDT S1283

THe Xigmatek consist of 3 pieces of 8mm heatpipes and 55 aluminum fins. As you guys know the Xigmatek is using the latest approach called as Heatpipe Direct Touch, short for HDT to provide a direct between the heatpipes and the ihs. Well check the pic below on how the TIM spreaded
The method I apply the TIM is a lil different from the previous 2 heatsinks as I am applying the TIM directly on the base of the heatsink itself



This post has been edited by sup3rfly: Aug 16 2007, 08:41 PM

Dimensions of the Heatsinks









Results




Conclusion

Noctua NH-U12F
Noctua NH-U12F no doubt is a good heatsink as well, the structure of the heatsink is very sturdy. The only disadvantage I can see from this great heatsink is that the amount of fins, compared to the other 2 heatsinks, it has the least fins, the amount of the heatpipes might be the culprit of this heatsink for not being perform as well as the other 2.

TR Ultra-120 eXtreme
This new revision of the heatsink made a come back with additional heatpipes and revised mounting mechanism. the build quality of this mighty heatsink is the best compared to the other 2. It has 52 heatsink which do help in dissipating the heat from the heatpipe. You guys ever imagine Ultra-120 eXtreme with HDT technology??? I think it will kick ass

Xigmatek HDT S1283
The new king from the test. with the HDT approach, it became the new king for air cooling. This heatsink is rather light in weight. 600g without fan mounted on it, the build quality of this heatsink yet need to be improved.

Tuniq Tower 120
This heatsink was not featured in the test due to the inteference with the Noctua U6 n/b heatsink that is install on the mobo. I might do a simple test between Tuniq vs Xigmatek on Socket 939 system.

This post has been edited by sup3rfly: Aug 17 2007, 08:57 AM

reserve plz

maybe u can get a cardbox to cover/enclose the setup, to simulate casing temps. Cause open bench with fan the ambient temp sure fluctuate

well the temp prob is not in between the hsf fin, it was fix at 1 place to determine the overall room temp

xigmatek hdt s1283 the new champion??? wow...

wow hold ur horses 2C better than Ultra 120 EX?this baby gonna be super ultra el cheapo value for money go buy now cooler liao.xigmatek cost u rm179 and TR 120EX cost u rm250(no fan included) also need to buy good fan rm50.

bro wanna ask>?XIGMATEK has any clearance problem with NC-U6 on DARK?

well depends on how u install the hsf... i dont have any issue mouting it with U6... initially i wanted to put tuniq in for the challenge as well but it cant fit in coz of U6.....

lol the ximantec is the new campayon...

later i will get this new gem

No way!! The xigmatek is better than the TR Ultra 120 EX????

Nice review! Thx! Is this the end of the review? Any more test?

Actually, available reviews so far indicated that the TR Ultra 120 EX would beat the xigmatek in any test, although the xigmatek would be quite close behind. This is no surprise as the TR as 6 heatpipes as compared to the xigmatek, which only has 3.

From what I know, the xigmatek would be quite close in performance as compared to the TR in a closed case system. However, it is less efficient in an opened case system as compared to the TR.

This review indicates that maybe the xigmatek is afterall the best cooler!!!

Another issues that I would like to point out is that the TIM is not spread out to all the 3 heatpipes. I believe that the side heatpipes would perform better had the TIM covered the whole heatpipe. What do you guys think?

I actually applied TIM to all 3 heatpipes on my xigmatek to make sure all heatpipes are in good contact with the IHS.

One last thing, I still think the TR is slightly more superior to the xigmatek. However, in terms of pricing (+performance), I think we now know who is the king of air cooler - The XIGMATEK!!!!

Note : Of course, need to wait for more reviews too.

well for me both are on par.... and my testing is not 100% accurate one...its just a reference for u guys only

thanks bro fuiyhhhh gimme heart attack due to my ignorance...btw how did u installed it vertically or horizontally?

some pics wud help a newbie like me

i post pic again later can??? didnt take any pic
u wait for a while plz

Noctua NH-U12F lose behind a lot . good job xigmatek ..

value + performance = king of cooler xigmatek~

Would you mind taking a picture of the 3 heatsinks side-by-side? From frontal, side and top view. This will really help us get an idea of their relative size against each other. Thx!

This post has been edited by kmarc: Aug 16 2007, 04:16 PM

Good review, how bout whacking the speed higher to c how it copes with the heatload

i think i'm gonna buy the Xigmatek for testing purposes too la for my FX-60 (old, but i guess, still acceptable aight...?), wondering if it on par with my belove HR-01

ok picture uploaded

aiyak cant already.... ultra extreme not with me anymore

i suggest u not to get it 1st.... having some issue with the mounting on 939 mobos.... still need sometime to do the testing for amd
will post some results later the seller is liasing with the manfacturer regarding this matter as well

, owh really... well, actually, i do wanted to test it out on S-939 mobo, or maybe can loan some of Xigmatek

Thx for the pictures!!!

Look at the base. The noctua has the thickest base and the heatpipes are high up. The TR's heatpipes are quite near the base but nothing can beat the HDT's base!!!

The noctua is also relatively bulkier as compared to the other two. Which one is wider? I think the TR is wider than the Xigmatek, right?

Imagine the efficiency of the HDT, 3 heatpipes can be on par with conventional 6-heatpipe designs....

This post has been edited by kmarc: Aug 16 2007, 05:05 PM

sup3rfly , may I know that before u mount the cooler, u applied the TIM on the IHS of the processor or on the base of the cooler?

for noctua and TR i apply on IHS but xigmatek i apply on the base as instruct by the manual

The Xigmatek heatpipes have a bigger radius.

haizz..nice temp for newer hsf..put it tuniq oso la bro..gonna c the result for those

bro u havent answered my question

can save me the hassle when putting the HSF and NC-U6 later

nice quick comparisons, bro

but u forget to test the result for OC environment, which most of us usually do, OC OC OC hehe

never thought xigmatek can be on par with ultra120Ex, let alone better it...

btw, i apply 3 stripes of TIM on the HSF base and another "spiral" shape on my cpu. wonder it perform worst or better as my AS5 pretty sticky. expired already? lol

now i'm really happy as my e6600 really cooled down way lower than my ex-athlon, at 32C idle! using push-pull config at silent mode!

its OCed system la bro... not stock speed at least
dont want to push it so high need time to make it stable la

haha..gud jobs bro..planning to sell the hsf after testing??..if yes then pls let me know ya..mayb will change my current hsf..

Nice comparisons..
i always thought Ultra120Ex is the king..
Xigmatek looks likes the new champ..

As usual from you, bro... High quality and detail review.

i dont have any of them bro....i borrowed them to test oni

well my test is not 100% correct but both cooler performs well

thx bro for the compliment

nice review superfly, enjoyed the poisons from ur old review too btw, and was a comfort that my tt bt vx was the winner. but the new poisons from this review is hard to resist... omg xigmatek >_<

what retension bracket they using btw? the old 775 pushpins? if they using something different then i may straight away go for it and sell off my tt bt vx
/hate 775 pushpins

and thanks again for the help last time along with shawty hehe, manage to settle down my temp issues and i have a stable 3.25 ghz qx6700 running at 40c pc probe and 60c coretemp vs the horrendous 50-60 pc probe and 70-80+c on core temp. still need a lot of work though, mainly from cleaning the leftover residues on the heatsink/cpu heatspreader. but no arctic cleaner or any alternatives available

well its pushpin as well...but during the testing i was using water cooling kind of mounting
as for the arctic cleaner u may wanna try minyak cap kapak.... its a lil oily but it does work.... just apply a lil on the ihs...dun too much

same exact 775 push pin? and yea really wanna get those mounting someday, from sunway right? ><

for the cleaner, not to doubt you but you sure or not, i thought minyak cap kapak is well, oil, and it would have left over oil residues?

that i m not sure...didnt check the temp after use hahaha
well its the same as intel stock push pin about the mounting u better get the from the link i show u the same as the dtek fuzion mouting that will be better.... the reason is because we cant find those plastic washer anywhere here....

heh k k thanks again

wow... i can't believe that Xigmatek wins...
what fan did u use in this test?

why do ppl suprise at the Xigmatek's performance?

issit because it's no as expensive as TRUE?

The Xigmatek is good... I just hate pushpins

yes. i thought more expensive TR120 eXtreme was going to win.

The price for this 3 heatsinks?

This post has been edited by Liuteva: Nov 28 2007, 06:18 AM

Please apply the thermal paste evenly cause the thermal paste will affect the result too. For the pic u can see the Xigmatech have more even thermal paste surface than the other

and this r some results i get from web.

Overcloker




Attached thumbnail(s)

IMHO this test is too simple. If you refer to a proper test like the ones they do on Anandtech, they not only test the coolers' performance at idle and load, but also how high each cooler is able to overclock, i.e., the cooling ceiling of each cooler. For example, they tested the CM Hyper 212 and found it to be very close to TR-120EX at lower overclocks, but once you go higher the Hyper 212 will saturate faster, meaning at a certain point it will not be able to cool any more, while the TR continues further.

Another thing to consider, some reviews have pointed to the Xigmatek being noisier than others, especially when the fans are maxed out (which they needed to do in order to release the full potential of the 1283). TS should have tested both min and max fan speeds. The Noctua and TR-120EX have more fins and are deeper, thus needing higher airflow but are designed such that noise is minimised.

That said, at the current settings for this test, and assuming one doesn't need to OC higher, the Xigmatek in this review does indeed seem to be performing really well.

Cheers

Wah, abang honda city modded so active in forum now.
So free now ah?

No more spending time on your car leh?

i find wat u say true bout the cm hyper 212... when i pump high voltage, high overclocks, the cm hyper has a tendency to wane away.. maybe due to the structure of the cooler itself.

I wonder if orientation of the HSFs affect performance for multi core-processors, as some are direct heatpipe contact with the processor heatspreader. In some configurations the heatpipes may actually rest over a larger area of the core's immediate heat dissipation through the heat spreader in multi core processors than others.

say, i mount my tuniq tower vertically(in casing), will it have difference performance compared to horizontal mounting? coz the pressure from horizontal is much better

Haha! Just like you retired from OCing, I retire from car modding loh Spend more time modding pc plak

---

Added on March 23, 2008, 6:03 amIINM I've read somewhere that for processors with integrated heat spreaders like the Core series, the most important part is the middle where the cores come in contact with the heat spreader. So it's best to mount in a way that all three direct heatpipes get in contact with the HS. Mounting it the other way, where only one heatpipe is in full contact while the others aren't, means that only the middle heatpipe is doing most of the work

However, you also need to consider the orientation of your heatsink in relation to airflow. No point having the best contact if that means having to position the fins in contra-flow of your casing's ventilation

---

Added on March 23, 2008, 6:18 amFrom what I gather, the number and size/length of heatpipes play a crucial role in determining how much, and how fast the heatsink is able to transport heat away from the proc. By adding two additional heatpipes to the original TR-120 improved the performance of the Extreme by a fair bit.

The next important piece of the puzzle is in the number of fins it has, to dissipate the heat from the pipes. How the fins are designed and placed, will determine the type of fan required in terms of CFM and static pressure. Fin design also affects noise levels.

Some fans may have high CFM in free-to-air situations (no restrictions), but place it in front of a restriction like a radiator or heat sink fins and it will not perform as well. Case in point are the Silverstones. The FM-121 has higher CFM than the FM-123, but the latter has a better static pressure rating so it's better to use the FM-123 for heat sink or radiator applications while the former is better for case ventilation.

As for the Hyper 212, have you tried to attach another fan? That heat sink is well known to perform much better with 2 fans. It even comes with all the necessary attachments for a 2nd fan

This post has been edited by gregy: Mar 23 2008, 06:26 AM

yeah i did, and its awesome.. i mean for what you pay for, its definitely the best bang for the buck, just for an addition of an extra fan..