Introduction

Close on the heels of my last roundup I'm happy to present you with the first for 2006. In the last roundup of 2005 I started using an Antec Sonata II for my test bed; and I’ll continue this method of testing in this roundup.

New heatsinks tested at Madshrimps:

AMD’s new Heat pipe Stock cooling

Thermalright HR-01

Thermalright SI-120

Thermaltake Big Typhoon

New heatsinks previously not tested inside a case:

Scythe NCU-2000 *

Scythe Ninja *

Thermalright XP-90C *

Thermalright XP-90 *

And heatsinks included for comparison which were already tested in the Sonata II:

AMD Stock Cooling *

Arctic Cooling Freezer 64 *

Arctic Cooling Freezer 64 Pro *

Asus X-Mars *

Coolermaster Susurro *

Noctua NH-U 12 *

Noctua NH-U 9 *

Sharkoon Red Shock *

Silverstone NT02 *

Scythe Samurai Z *

Scythe Shogun *

Thermalright XP-120 *

Zalman CNPS9500CU *

If you want to know more about the heatsinks starred (*) in the lists above please visit my heatsink review overview

Test Setups and Methodology

JMke's Test Setup
CPU	A64 3200+
Mainboard	DFI NF3 Lanparty
Memory	1 * 256Mb PC3700 OCZ
Other	ATI R9000 Passive Cooling Silverstone EFN-300 300W Passive Cooled PSU Maxtor 120GB IDE HDD

in-take temperature was measured at 22°C for all tests, but temp fluctuations, different mounting and user error can account up to 1-3°C of inaccuracy in the obtained results. Please keep this in mind when looking at the results. Each heatsink was tested repeatedly; if we got questionable results the test was restarted.

Noise level of each HSF combo was recorded with SmartSensor SL4001A, the sensor was placed ~50cm away from the case. The lowest dBA reading in the test room was 32.5dBA with everything turned off!

System was stressed by running K7 CPU Burn for 30min (after Thermal Compound’s burn-in); this application pushes the temperature higher then any other application or game we’ve yet encountered. Speedfan was used to log maximum obtained temperatures.

Arctic Silver kindly send us their “Lumière” thermal testing compound which has the same colour as Ceramique, but only a break in time of 30min!

Arctic Silver’s ArctiClean was used to clean off thermal paste of the CPU and heatsink between tests

Fans used for comparison

To eliminate as much variables in the tests I test each heatsink with a “reference” fan if I can mount them. If the HSF comes with its own fan, I will compare the performance of that fan to the reference one I use.

Delta NFB0912L 92mm: 42CFM

Papst 120mm 4412 F/2GLL: 40CFM


The Case

Since I’m only using an Athlon 3200+ for my tests, it would be interesting to overclock the CPU so its maximum heat output increases and it can simulate a higher clocked Athlon 64.

I recently purchased a power meter similar to this. Doing a few basic measurements with the test system gave these results for full system wattage usage.

Athlon S754 3200+ @ 2200Mhz - 1.5v: idle: 67Watt / Load: 125Watt

Athlon S754 3200+ @ 2420Mhz - 1.7v: idle: 78Watt / Load: 165Watt

In my days of Athlon XP HSF testing an increase of 0.1v vcore would result in 4-6°C higher CPU temps, so without much surprise the temperature results in this roundup with the 1.7v Athlon 64 will be much higher.

Noise was recorded approx. 50cm away from the case at an angle, here’s a (very bad) drawing of how the dBA meter was position opposite the case and the test-room.


Green box = Sonata II – White Dot = dBA meter

3 different test scenarios were configured as such:

Test Setup 1: Overclocked and Silent

- Athlon 64 3200+ @ 2400 – 1.7v vcore
- Antec Sonata II + Silverstone Passive PSU
- AcoustiFan DustPROOF 120mm @ 5v in the rear as outtake (mounted with soft-mounts)
- nVidia TNT2 Passive cooled video card
- Noise produced with system running without HSF fan: 34dBA @ 50cm

Test Setup 2: Stock speeds and Silent

- Athlon 64 3200+ @ 2200 – 1.5v vcore
- Antec Sonata II + Silverstone Passive PSU
- AcoustiFan DustPROOF 120mm @ 5v in the rear as outtake (mounted with soft-mounts)
- nVidia TNT2 Passive cooled video card
- Noise produced with system running without HSF fan: 34dBA @ 50cm

Test Setup 3: Stock speeds and Silent – Passive CPU cooling

- Athlon 64 3200+ @ 2200 – 1.5v vcore
- Antec Sonata II + Silverstone Passive PSU
- AcoustiFan DustPROOF 120mm @ 12v/7.5v/5v in the rear as outtake (mounted with soft-mounts)
- nVidia TNT2 Passive cooled video card
- Noise produced with system running without HSF fan: 34dBA @ 50cm

What was measured?

The CPU temperature was measured with SpeedFan and highest value recorded

Temperature of air coming into to the case at the front

PWM temperature through SpeedFan, this represent the area around the CPU socket, the power management caps which you see on a motherboard, they are there to make sure the power which is fed into the motherboard coming from the PSU is filtered and delivered the CPU and other components. Too high temperature will cause Vcore fluctuations which in turn causes system instability.

Let's get started with our first new contestant ->

Thermalright SI-120:
Supplied by: Bacata

When Thermalright released the first 120mm powered CPU cooling in the world; they traded in compatibility for performance. The XP-120 was at the top of the performance/noise balance list for a long while. Other companies jumped on the large CPU fan bandwagon and the XP-120 started loosing ground. Currently it’s still an excellent heatsink but it doesn’t fit all motherboards out there and newer heatsinks are offering better performance.

So Thermalright went back into their labs and came up with the SI-120, it touts compatibility with all motherboards and while the basic design hasn’t been overhauled; there might be some performance differences between the two versions? That’s what I’m going to find out.

It’s been on the market for a while now and can be found priced at ~€40 / ~$45

Shipped in a plain cardboard box with the company logo at the top - nothing fancy.

Specifications :

Compatibility:

AMD: Athlon64/64+ Socket754/940/939

Intel S478 and S775 (through a custom mounting bracket)

Fan included: None (supports 120mm)

Heatsink Dimensions: L135xW125xH90mm
Weight: 400gram

In the Box :

A large sticker

2 clips to mount a fan (not compatible with bridged fans like this one)

A manual

A rather large tube of thermal compound

2 adhesive rubber strips used for vibration reduction (between heatsink and fan)

A K8 mounting bracket and screws

Construction :

The SI-120 resemblances the XP-120 strongly but you’ll notice the changes immediately.

- The unit is quite a bit higher than the XP, so this means longer heat pipes. They also ditched the small heatsink from the base.



- The total surface of the base is larger and quite smooth.




- Seen from the top the difference between the two is less visible.




Installation next ->

Installation :

The SI-120’s installation method does not differ from the XP-120. On the AMD Athlon 65 platform you swap out the default bracket, with the one included from Thermalright. This brackets mimics the default Intel S478 bracket and it also allows you to use Intel only heatsinks (which use the Intel bracket) on your Athlon 64 system.

The SI-120 is clipped on easily there’s enough clearance from all the motherboard’s components. The only thing to watch out for is the uninstall of this heatsink, the best method I found was to remove the 120mm fan, push down on the heatsink and use a small flat screwdriver to remove the clips on one side. Then you slowly twist and tilt the heatsink so it comes off without taking your CPU with it. If you don’t pay attention you’ll have your CPU still attached to the base when you remove the heatsink. There’s a very real risk of bending the pins, even S754 offers more than enough pins to bend.

Thermalright’s XP-90(c) and XP-120 also have this same problem, and many other heatsinks with large flat bases; taking your time when installing and uninstalling can save you money.

With that all said, let’s see how much clearance the SI-120 provides compared to the XP:

Clears the capicators around the socket

XP-120:


SI-120:


XP-120:


SI-120:


Performance and Noise :

Compared to the XP-90, XP-90C and the new/old AMD stock cooler. First at stock speeds:



The SI-120 takes the overall lead! the PWM area only saw a 1° drop compared to the XP, I had expected more; the CPU temps take a bigger plunge downwards: 3°C!

Time to increase the heat; At overclocked and overvolted settings the XP-90 is overwhelmed and the system crashes. It must be said that the Delta 92mm fan I use for testing is geared toward SILENCE, not performance; if you don’t mind extra noise, a faster fan will let you overclock your system easily with XP-90. So here are the results of the SI-120:



While a 3°C difference can still be called a “close call”, a 5°C difference between high end heatsinks can be called significant. The SI-120 has a comfortable lead; the PWM temperature remains in the safe zone; while the XP-90C with 74°C for PWM is worrying.


The Verdict:

When the SI-120 was first released many respected review websites got a chance to test its performance and compare it to the XP-120; almost all of them found the difference to be very small, in the 1-2°C region. Fact is though that they performed their tests on an open test bench, where there is plenty of fresh cool air around. Inside a case however it tells a different story. The SI-120’s increased height and larger base give it a clear performance advantage over the XP-120, together with better compatibility it’s the best unit from Thermalright for silent computing.

PRO
Excellent performance
High compatibility
Easy Clip-On Installation
Fair pricing

CON
be careful during uninstall of this HS – CPU might pop out of its socket.
Default fan clips only work with none-bridged fan. (Others can be ordered for ~$1)

Where to buy?


Next up is Thermalright’s latest offering, the HR-01 ->

Thermalright HR-01:
Supplied by: Bacata

Thermalright first tower heatsink is upon us, the HR-01 is aimed at silent computing; it doesn’t come with a fan nor does it provide an easy method of attaching one. Some versions of the HR-01 (especially those for S775) are shipped with a plastic fan duct.

It’s not been on the market for long so its price may still drop: ~€50 / ~$50

The unit is shipped in a plain cardboard box, the optional fan duct is shipped separately.

Specifications :

Compatibility:

AMD: Athlon64/64+ Socket754/940/939

Intel S478 (needs separate retention kit)

Intel S775 (another version altogether - tested on page 6)

Fan included: None and None supported

From TR: Currently there are no adapters or fan wire clips available for fan installation on HR-01. But if you are the type that likes to do a little DIY, here's what you can do

Heatsink Dimensions: L110xW60xH159.5mm
Weight: 525gram

In the Box :

A large sticker

A manual (A4 sheet)

A rather large tube of thermal compound

Mounting clip for use with standard A64 bracket

Construction :

The HR-01 cools the CPU through the use of 4 U formed heat pipes joined at the base; a large series of thin aluminum fins dissipate the heat.



The base has visible machine markings, the SI-120’s base was smoother




Compared to the SI-120 the design is quite different; it’ll be interesting to see how the unit will perform.





Installation & test on AMD Athlon 64 next ->

Installation on AMD Athlon 64:

The clip is easy to use and makes this one of the easiest tower heatsinks to install on Athlon 64. When the heatsink is in place there seems to be room for rotating it clock and counter-clock wise. However if you look closely again at the base of the heatsink you will see that at the sides there are 2 small ridges.



The HR-01’s clip on system also has a drawback, you can not change the orientation, so if the CPU socket is facing the wrong way, you are out of options. On the DFI NF3 board that is the case; the heatsink is facing the wrong way… so at first I was enthusiast about the fact that I could turn the HR-01 a bit; however in practice I noticed a large increase in CPU temperatures (close to overheating) when doing this, meaning that the small ridges at the side of the heatsink were the cause of bad contact when trying to turn the HR-01. If Thermalright is able to lower these ridges below the base, you would be able to point the heatsink almost every which way you want.

I must admit I’m not much of a DIY person; the only casemods I’ve done turned out horrible - they didn't looked bad but performed good (thermal wise with all the blowholes;)); and while my “hardware killed” list is not very extensive compared to amount that has passed through my hands, I do tend to catch any “newbie” prone errors which could be made, mostly because I can get clumsy. With that said, I have sufficiently prepared you for the photo below; I mounted a fan using an old mouse cord.


Thermalright HR-01 with fan “mounted”

Beside the ghetto fan mounting you can also see that the HR-01 is clearly pointing the wrong way, this will have a negative effect on performance.

I compared the HR-01 to Scythe Ninja (which has fan mounting clips) and the Scythe NCU-2000 (does not have fan mounting clips – NCU-2005 is version which does have these). So I repeated my ghetto fan mounting mod:


Scythe NCU-2000 with fan “mounted”

The NCU-2000 can be rotated 90°C so the fan blows towards the rear of the case.


Performance and Noise :

The temperature results below are taken with active cooling, the passive cooling tests can be found at the overall performance comparison pages.

First at stock speeds:



The NCU and HR-01 are very closely matched, the higher PWM temperature with HR-01 can be attributed to the fact it’s not blowing air towards the back. CPU temperatures with HR-01 would likely also be lower if you could orientate the HR-01 correctly. The Scythe Ninja takes comfortable lead in this test, and stays ahead even with Papst at 50% fan speed.

Increase the heat:



The stock AMD cooling solutions are close to 4 times louder as the tower heatsinks with the silent Papst 120mm fan, the NCU-2000 barely keeps the system running stable, the HR-01 also comes close to dangerous values, PWM is very high. The Shogun and Ninja are evenly matched and have a comfortable 5-10°C lead over the competition.

Thermalright did not design the HR-01 to be used with active cooling and it shows, the performance numbers show it trailing the latest tower heatsinks. The fact that the unit is orientated wrong doesn’t help either.

Let’s see how the HR-01 does on a S775 system ->

Test on Intel S775 / LGA775
Tested by Liquid3D: Sample supplied by Thermalright.

The Intel Socket 775 version of the HR-01 is shipped with the fan duct.

Liquid3D’s Intel Test System
CPU	Pentium 630 Retail (SL7Z9 3.0GHz 2MB L2 1.25V ~ 1.388Vcore) Socket-775
Mainboards	1.) Asus P5AD2-E Premium (BIOS 1005) 2.) Asus P5ND2-SLI Deluxe
Memory	Crucial Tracer Ballistix 5300 (2x512MB DC CL4-4-4-12)
Graphics	1.) AOpen Aeolus 7800GTX-DVD256
Power Supply	PCPower&Cooling TurboCool 850 SSI
Cooling	Intel Stock LGA-775 HSF Thermaltake Blue ORB II Alphacool NexXxos XP with Cora 642 passive radiator Thermalright HR-01
Case	Thermaltake Kandalf: - 1x120mm in-take front - 1x120mm exhaust back - 1x92mm exhaust back

The room temperature during test was 20°C , but temp fluctuations, different mounting and user error can account up to 1-3°C of inaccuracy in the obtained results. Please keep this in mind when looking at the results. Each heatsink was tested repeatedly; if we got questionable results the test was restarted.

the system was stressed with S&M tool

The LGA775 version of the HR-01 has its mounting system integrated in the unit; installation is straight forward through the use of push pins, using the same method as the stock Intel cooling.





However, while the mounting system is perfect for the stock Intel cooling, with the HR-01 and its 500+gr weight it’s on the verge. While installing the plastic duct I had the pins pop out when trying to squeeze the duct between the rear 120mm fan and the HR-01. The leverage created by the height of the unit already causes some strain on the pins so being careful when installing the fan duct is advised.

The rear fan in the Thermaltake Kandalf is a blue LED fan: flashy result




Performance and Noise :

I tested the HR-01 in 3 ways :

Without the fan duct

With the fan duct and the rear fan blowing hot air out the case

With the fan duct and the rear fan drawing cool air in the case

This is the outcome:



In a well ventilated case the HR-01 proves to be an excellent performer when the fan duct is used. When the rear fan is flipped and set to draw in cool air the temperatures take another plunge. It’s also obvious that the HR-01 benefits enormously from the included fan duct, without it the performance is almost identical to the stock Intel cooling (but without the extra noise).

With the CPU running at default speed the HR-01 trails the passively cooled water cooling setup by 6°C, however when the heat is turned up the HR-01 is able to cope with the extra heat much better, closing in on the Alphacool kit.

Although I did recently obtain a dBA meter I’m waiting for it to be calibrated; however I can give you the details on the 120mm fans used inside the system. They are identical to the one supplied with the Thermaltake Big Typhoon, JMke tests this same fan on the next pages, he compared the fan to his Papst 120mm 4412 F/2GLL and found out they were evenly matched.


The Verdict:

Thermalright’s first tower heatsink showed excellent performance on the Socket 775 Intel setup, the fan duct really helps to boost its performance. The Athlon 64 version has some drawbacks due to its mounting system not allowing the unit to be rotated to face the correction direction which hurts its performance, the Athlon 64 version sadly doesn’t ship with the fan duct. Neither version has support for a fan to be mounted, something which could have been added easily if they wanted to.

While the HR-01 does deliver passive CPU cooling, in terms of not having to strap a fan to the unit to make it work, it does depend heavily on case airflow to cool your CPU. Good case cooling means that you need fans present, which are all small noise makers, so having a silent system with the HR-01 means you might need to think about ghetto style ducting to get the most out of this product.

PRO
Excellent performance with fan duct
High compatibility
Can be rotated freely on Intel S775.
Easy Clip-On Installation (both A64 and Intel version)
Competitive pricing

CON
Depends heavily on case airflow
Installation of fan duct on Intel S775 can be tricky
Can not be rotated freely on Athlon 64 which hurts performance with some motherboard/case combo's

Where to buy?

Update - new bracket for free 90°C angle:

You’ve heard of HR-01, but with most AMD motherboards, there exists a problem…the way the AMD socket is positioned and oriented, the HR-01–K8 couldn’t be turned to face the 12cm exhaust fan in order to work with the fan duct that we developed. Hence, the solution…S-type HS clip. Now you can rotate the HR-01 90 degrees to face the rear fan and connect the fan duct for maximum performance and true silence! The way HR-01 is meant to work!! Passively and quietly with performance level that no other heatsinks can match! And that includes active cooling heatsinks, too!

Next up is Thermaltake’s high end heatsink, the Big Typhoon ->

Thermaltake Big Typhoon:
Supplied by: Thermaltake

Thermaltake was absent from my last roundup and many readers suggested I did something about that. I contacted Thermaltake and they were happy to provide their latest and best heatsink for review, the Big Typhoon.

It’s on of their big boys alright; equipped 6 heat pipes and having a 120mm fan for cooling it towers over your motherboard. First impressions were good.

The Big Typhoon is not extremely new, priced very competitive: ~€35 / ~$35

It comes in a stylish retail package:



Specifications :

Compatibility:

AMD: Athlon64/64+ Socket754/940/939

Intel S478 and S775

Fan included: 120x120x25mm (54.4CFM)

Heatsink Dimensions: L122xW122xH103mm
Weight: 813gram

In the Box :

A large collection of mounting gear for the different platforms

A detailed picture manual

A small bag of generic thermal compound

Construction :

The Big Typhoon is colossal, at first sight when you see it installed you might think of a Thermalright XP-120 on steroids. It larger and heavier than any other heatsink I’ve seen yet which has its fan installed to blow air onto the motherboard.



A profile shot reveals that there is quite a bit of room left below the aluminum fins ; allowing the air to escape.



The base is made out of copper and joins the 6 copper heat pipes. Finishing is "ok" but lapping the base might improve performance by a degree or two.



Looking closer at the Big Typhoon with the fan removed the layout of the heat pipes and the aluminum fins does ring a bell. If you were to enlarge the base of the Sonic Tower and then bend the heat pipes into a different shape you basically get:




The 120mm has a custom paint job, sleeved 3-pin cable



This fan was an excellent choice by Thermaltake, more on the next page ->

Installation :

The Big Typhoon’s installation bracket is designed to be used with the default motherboard back plate; however some motherboards use a custom back plate which is incompatible with this installation method. In that case you are advised to use the back plate included with the Big Typhoon. The manual included outlines this clearly and explains with pictures how to check if the Typhoon will fit straight away or not.

Motherboard removal was necessary in my case, as I could not reach the screws otherwise (the PSU was blocking the path) and the DFI NF3 back plate is one of those on the not compatible list.



Placed back inside the case, the BT takes up quite a bit of room, but clears the DDR slots as well as the caps around the CPU socket; its small base and height make the BT compatible with practically all motherboards out there.



Performance and Noise :

I’ve seen plenty of topics in different enthusiast forums regarding which heatsink somebody should purchase to get the best performance/noise ratio. I’ve compared the Big Tyhpoon in the graphs below to the Scythe Ninja and Thermalright SI-120 in hopes to provide some answers.

First at stocks speeds:



The included fan is really a winner in my book; it offers performance on par with the Papst 120mm. As you can see with the Papst @ 100% the difference between the Ninja, SI-120 and BT is negligible. However when the fan’s speed is halved the BT falls back and trails the Ninja by a good 5-6°C.

Let’s turn up the heat:



The Big Typhoon is right up there with the best heatsinks currently available; the 2-3°C difference is well within my margin of error.

The Verdict:

Thermaltake has made a winner, the Big Typhoon really impressed me, for a lower price than the competition you get a high quality heatsink and excellent 120mm fan – used together they deliver performance on par with the best out there.

The BT fits all current CPU sockets and is fairly easy to install, although you do need to remove the motherboard. I advise to use the included metal mounting plate if your Athlon 64 motherboard has a plastic one, weighing in at 813gr it’s a wise thing to do.

PRO
Excellent performance
High compatibility
Included 120mm fan offers good performance/noise ratio
Very competitively priced

CON
Weighs over 800 gram – use of metal mounting plate advised (comes included)
At VERY low fan speed performance is less stellar

Where to buy?

Extra note: An XtremeSystems forum member has done a simple yet nifty modification of his Big Typhoon using a piece of foam. It forces all the air to go through the heatsink's fins and increases performance a bit. Not a bad find!


Photo by davexl from XS.org

Last newcomer in this roundup is the AMD retail heatsink equipped with heat pipes ->

AMD Retail heatsink with Heat Pipes:
Kindly provided by forum member Rutar from Swiss, thanks!

My fellow reviewer Lazyman tested this heatsink not so long ago and found performance to be on par with the Thermalright XP-90 when both use the same 80mm fan.

At the beginning of January 2006 Hexus.net published their review of AMD’s new Athlon 64 FX-60, in that review they dedicated a section to the new stock cooling which came with the CPU. Their photos reveal that it is the same unit I have here today for testing. However their comments are most interesting:


This would mean that Coolermaster has partnered with AMD to provide their new high end retail coolers. They state each heat pipe is capable of 30W heat capacity, so in total a 120W device.

So do these new heatsinks come with all of AMD’s processors? The answer is unfortunately: no. So far from what I can tell from user experience shared at forums around the web these new heatsinks come with Athlon X2 above 3800+ and Opteron Dual Core CPU’s.

While they can’t be bought in shops, they can be found for $15-25 on eBay.



Specifications :

Compatibility:

AMD: Athlon64/64+ Socket754/940/939

Fan included: Delta 80x80x25mm (temp controlled – max ~5200RPM)

Heatsink Dimensions Approx: L95xW90xH65mm
Weight: ~400gram

In the Box :

In the retail box you most likely also find a CPU and an AMD case badge :-)
The heatsink comes pre-applied with a high quality Shin-Etsu compound.

Construction :

A copper base which joins 4 copper heat pipes




A series of small aluminum fins for heat dissipation



Installation :

It installs like any other retail AMD Athlon 64 heatsink, of course!



The 80mm fan has a small thermal sensor placed near the center of the fan



In the performance tests below I let the sensor decide how fast the fan should spin, mainly because at full (100%) speed the noise level is deafening.

Performance and Noise & Verdict:

The new retail cooling is directly compared to the old full aluminum one, Coolermaster’s own Susurro heatsink and the Asus X-Mars. The 92mm fan on the Susurro has a 80mm frame and I tested it also on the stock cooling; The Asus X-Mars is equipped with 2 heat pipes, while the Susurro has none. Let’s see how they all do at stock A64 3200+ speeds:


Compared to the old one you get a 3°C drop in temperatures and 3dBA noise the Delta fan is running at 3300RPM. With the 92mm fan equipped another 2°C reduction is seen but the noise can cause a head-ache real soon, the tight fitted aluminum fins create even more turbulence – result is 65dBA at 50cm.

Let’s turn up the heat:



The thermal sensor of the Delta has noticed the increased heat and is now spinning at 4000rpm, producing 52dBA, which not really silent, it’s actually louder than the old A64 heatsink, but delivers better temperatures – 5°C temp drop CPU and PWM.

With the 92mm fan the performance is on par with the best out there, but your hearing is paying the price.

PRO
Free with higher end AMD CPU’s – can be found cheap on eBay.
Auto sensing fan works well
Okay performance

CON
Not really a silent solution if your CPU is running at full load.


Silent computing with passive CPU cooling compared more in-depth on the next page ->

Passive CPU cooling:

In this roundup I’ve tested the Thermalright HR-01 with active cooling, but how does it do without a fan strapped to it?

The rear Acoustifan was configured to run at 12/7.5/5v which generated respectively 43/37/34dBA at 50cm. As an addional test I made a small fan duct out of the card board box of the HR-01 and installed in on the NCU-2000 and HR-01.

This is what they look like with some Ghetto ducting




Running the CPU overclocked was too overwhelming for this passive solution, only when running the rear fan at 12v and an additional 120mm fan in the front would make it possible without the CPU overheating, but that would be defeating the silent purpose of this test.

Even now with the rear fan at 12v the PC is generation 43dBA which is more than with each heatsink strapped with Papst @ 12v and rear fan at 5v.



The NCU-2000 and HR-01 are very evenly matched, even with the HR-01 orientated the wrong way; the ghetto fan duct allowed the HR-01 to be run with the rear fan at 7.5v and offer better performance than with the rear fan at 12v without the duct. Temperatures dropped ~10°C by simply adding two cardboard panels. Not bad!

The Ninja does not need a fan duct to make it take the performance crown in this test, at 12v it has a comfortable lead and even with the fan at 5v it’s able to keep the CPU running stable.

Anandtech Forum member corsa accurately noticed that I left out PWM values for the passive tests, you can find them in graph below, this time the results are ordered by PWM value:



The fan duct does a good job at keeping the PWM values in safe regions, when the rear fan is at 5V the temperatures really go up into 60-70°C range.

On to the active CPU cooling tests, a global comparison of all 21 coolers tested so far inside the Sonata II case ->

Features and Compatibility List:

	Fan Support	Fan Included?	Weight (gram)	Motherboard removal?	AMD	Intel
AMD Stock Cooling	70mm	Yes	325	No	S754/S939	no
AMD Stock Heatpipe	80mm	Yes	400	No	S754/S939	no
AC Freezer 64	Custom 92mm	Yes	460	No	S754/S939	no
AC Freezer 64 Pro	Custom 100mm	Yes	528	No	S754/S939	no
Asus X-Mars	70mm	Yes	385	No	S754/S939	no
Coolermaster Susurro	Custom 92mm	Yes	760	No	S754/S939/AM2	no
Noctua NH-U 12	2*120mm	No	700	Yes	S754/S939	S478/S775
Noctua NH-U 9	80/92mm	No	570	Yes	S754/S939	S478/S775
Scythe Samurai Z	92mm	Yes	360	No	S754/S939	S478/S775
Scythe NCU-2000	-	-	505	Yes	S754/S939	S478
Scythe Ninja	120mm	No	665	No	S754/S939	S478/S775
Scythe Shogun	120mm	Yes	790	Yes	S754/S939	S478/S775
Sharkoon Red Shock	80/92/120mm	No	820	Yes	S754/S939	S478/S775
Silverstone NT02	80mm	Yes	880	Yes	S754/S939	S478/S775
Thermalright HR-01	-	-	525	No	S754/S939	S478/S775
Thermalright SI-120	120mm	No	400	No	S754/S939	S478/S775
Thermalright XP-90	80/92mm	No	360	No	S754/S939	S478/S775
Thermalright XP-90C	80/92mm	No	690	No	S754/S939	S478/S775
Thermalright XP-120	120mm	No	370	No	S754/S939	S478/S775
Thermaltake Big Typhoon	120mm	Yes	813	Yes	S754/S939	S478/S775
Zalman CNPS9500LED	Custom Size	Yes	530	No	S754/S939	S478/S775

On to the performance results ->