Smallish LGA775 Heatsink Roundup – Part 2

In our continuing quest for compact, effective and silent CPU coolers, we ran a gladiatorial roundup with competitors from Silverstone, Nexus and Cooler Master seeking to wrest the top spot from the Scythe Big Shuriken.

September 21, 2009 by Lawrence Lee

Our low profile LGA775 CPU cooler roundup
from two months ago revealed only one clear-cut recommendation: the Scythe Big
Shuriken, a short but capable cooler with an excellent fan at a reasonable price.
However, it is far from the ideal compact cooler — its low clearance over the pushpin mounting system makes it excruciating to install, and taller motherboard
components like Northbridge heatsinks can impinge on its use. As small form
factor systems become more popular, the demand for better low profile
coolers grows.

Today we have a new set of contenders, a varied group, most of which may be
too big to be considered “low profile” by many. However, they are
smaller than the typical tower giants that dominate best-of cooler ranks and there are many cases with just enough
room to accommodate them. As they wouldn’t fare well against the
behemoths we test on a regular basis, we threw them together in this gladiatorial romp.

Heatsinks Tested: Physical Dimensions (measured)
Heatsink	Height	Fan Diameter	Fin Thickness	Fin Spacing	Weight
Silverstone NT06-E (Nexus fan)	107 mm	111 mm	0.34 mm	1.4 mm	550 g
Cooler Master Geminii S	86 mm	112 mm	0.28 mm	1.6 mm	500 g
Arctic Cooling Alpine 7 Pro	86 mm	85 mm	0.93 mm	1.8 mm	450 g
Nexus LOW-7000	70 mm	111 mm	0.33 mm	2.0 mm	480 g
Intel Q6600 HSF (Nexus fan)	67 mm	111 mm	varies	varies	540 g
Arctic Cooling Alpine 7 GT	64 mm	74 mm	0.44 mm	2.6 mm	280 g
Intel Aluminum HSF	63 mm	77 mm	varies	varies	330 g
Intel Q6600 HSF	62 mm	82 mm	varies	varies	440 g
Thermolab Micro Silencer	58 mm	85 mm	0.43 mm	1.4 mm	270 g
Scythe Big Shuriken	57 mm	113 mm	0.32 mm	1.3 mm	420 g
Intel Q9550 HSF	45 mm	79 mm	varies	varies	260 g
Thermolab Nano Silencer	39 mm	76 mm	0.38 mm	1.5 mm	210 g
Silverstone NT07-775	36 mm	80 mm	varies	varies	300 g
Coolers tested in our last roundup in green.

TESTING

Testing was done according to our
unique heatsink testing methodology, A quick summary of the components,
tools, and procedures follows below.

Key Components in Heatsink Test Platform

• Intel
  Pentium D 950 Presler core, C1 stepping. TDP of 95W; under our test
  load, it measures 78W including losses in the VRMs.
• Asus P5Q-EM motherboard.
  A microATX board with integrated graphics and short solid-state capacitors
  around the CPU socket, and a diminutive northbridge heatsink for maximum compatibility.
• Intel
  X25-M 80GB 2.5″ solid-state drive.
• 1GB
  of Corsair XMS2 DDR2 memory. 2 x 512MB PC2-8500.
• FSP Zen 300W
  fanless power supply.
• Arctic Silver
  Lumière: Special fast-curing thermal interface material, designed
  specifically for test labs.

Measurement and Analysis Tools

• Seasonic
  Power Angel for measuring AC power at the wall to ensure that the
  heat output remains consistent.
• Custom-built, four-channel variable DC power supply, used to regulate
  the fan speed during the test.
• PC-based spectrum analyzer:
  SpectraPlus with ACO Pacific mic and M-Audio digital
  audio interfaces.
• Anechoic chamber
  with ambient level of 11 dBA or lower
• Various other tools for testing fans, as documented in our
  standard fan testing methodology.
• SpeedFan,
  used to monitor the on-chip thermal sensor. This sensor is not calibrated,
  so results are not universally applicable.
• CPUBurn
  P6, used to stress the CPU heavily, generating more heat than most
  real applications. Two instances are used to ensure that both cores are stressed.
• Throttlewatch
  2.01, used to monitor the throttling feature of the CPU to determine
  when overheating occurs.

Load testing was accomplished using CPUBurn to stress the processor, and the
graph function in SpeedFan was used to make sure that the load temperature was
stable for at least ten minutes. The stock fan was tested at various voltages
to represent a good cross-section of its airflow and noise performance.

Silverstone
NT07-775 – $20USD

The Silverstone NT07-775 is the smallest heatsink we’ve ever tested. It designed
for ultra thin cases like the Silverstone LC19
and ML02.

The NT07-775 stands only 36 mm high and weighs 300 grams.

 

The heatsink shares the same radial fin orientation as the Intel stock cooler, but it is 9 mm shorter. The fan has thinner, more numerous blades. Unfortunately the support struts are almost parallel to the trailing edges of the fan blades, increasing the potential for tonal turbulence noise.

 

Like the heatsinks that ship with Intel’s quad core CPUs, the NT07-775 has a copper core.

 

Installed on our test platform.

 

Test Results
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1880 RPM	24 dBA	40°C	0.51
10V	1660 RPM	20 dBA	44°C	0.56
9V	1490 RPM	17 dBA	49°C	0.63
8V	1320 RPM	15 dBA	55°C+*	0.71+
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better. *Testing stopped after thermal rise exceeded 55°C.

The NT07-775 was fairly quiet overall compared to most CPU coolers, though
you would expect the fan to spin faster to compensate for its small size. At
12V, it had an annoying whine with plenty of buzzing generated by the motor.
The noise level was tolerable at 10V and below, but it did develop a bit of
a drone. At 8V it was very quiet, barely audible in most environments, but the cooling performance would not be good enough if your system is pushed hard regularly, except perhaps with the coolest CPUs.

Intel Q6600 HSF (Nexus fan)

We have always appreciated the core design of the Intel stock cooler, but
of course could not get past the terrible sounding fan that usually accompanies
it. The copper-core version is a prime candidate for a simple mod: A fan swap with a 120 mm fan. The end result is a relatively low profile heatsink with a big airflow footprint that can help keep motherboard components cool.

A pair of pliers made short work of the stock fan. Mounting a 120mm fan securely to the heatsink body requires some creativity. As a demonstration, we used four wire ties to bind the heatsink fins to the fan struts of a Scythe Slipstream fan.

 

Installed inside a Silverstone SG06. As you can see clearance was ample.

 

For our standard thermal test, we used our reference Nexus 120 mm fan. This combination should improve overall cooling and acoustics over the stock Intel fan. The entire package weighs 540 grams and is 67 mm tall.

 

Test Results
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1100 RPM	16 dBA	39°C	0.50
9V	890 RPM	13 dBA	45°C	0.58
7V	720 RPM	12 dBA	52°C	0.67
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better.

Arctic
Cooling Alpine 7 Pro – $15USD

A revisit of the Alpine 7 Pro is warranted as it was tested long before
we installed our anechoic chamber, and we use it time
and time again in the lab. Its manageable size and excellent acoustics makes
it our defacto substitute for an Intel stock cooler.

The Alpine 7 Pro stands 86 mm tall and weighs 450g. The heatsink has a simple design with broad thick 0.93 mm fins spaced a healthy 1.8 mm apart.

 

Installed in our test platform.

 

Test Results
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1930 RPM	28 dBA	24°C	0.31
9V	1600 RPM	23 dBA	28°C	0.36
7V	1430 RPM	17 dBA	33°C	0.42
5V	950 RPM	12 dBA	42°C	0.54
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better.

The soft-mounted fan measures 85 mm in diameter and is one of the better models
in our roundup. At 12V it is very loud with some whine, while the rest of the
noise is simply soft-sounding air turbulence. At 9V it can be considered quiet,
running very smooth with a SPL of 23 dBA@1m. At 7V it is audible but the acoustic
profile is completely benign with just a soft ‘whoosh.’ At 5V, it’s completely
inaudible at one meter.

Silverstone
NT06-E – $55USD

The NT06-E is the premier compact cooler from Silverstone carrying a $55 price-tag.
It does not ship with a fan, but according to Silverstone, if there is a 120
mm power supply fan directly above it (as is the case with their Sugo
SG05/06), it can be run passively depending on the processor and fan
speed.

The NT06-E has a five heatpipe design and a bolt-thru/backplate mounting mechanism.

 

The NT06-E is a “C” shaped cooler like the Shuriken, but it’s a fair bit taller and doesn’t rely on pushpins for mounting. It weighs in at 430 grams and sits 82 mm high. Unfortunately the addition of a standard 120 mm case fan pushes it to 107 mm and 550 grams, which simply won’t work in most compact cases.

 

The fins are very thin at 0.34 mm, and they are tightly packed with an average separation of 1.4 mm. It takes a standard 120 mm fan.

 

The NT06-E is easy to install on the LGA775 socket. A backplate with screws goes on the back and thumbscrews finish the job topside.

 

As the NT06-E does not ship with a fan, we tested it with our reference Nexus 120 mm fan. It is secured using 4 long black screws.

 

Test Results
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1100 RPM	16 dBA	29°C	0.37
9V	890 RPM	13 dBA	30°C	0.38
7V	720 RPM	12 dBA	33°C	0.42
5V	530 RPM	11 dBA	43°C	0.55
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better.

Nexus
LOW-7000 – $55USD

The Nexus LOW-7000 looks like a smaller, less refined cousin of
the NT06-E. Its physical dimensions puts it closer to the Big Shuriken.

The LOW-7000 has four heatpipes and a Nexus branded fan.

 

The sample we received was bent like an old man with a bad hip, but luckily it was easy enough to bend back to its proper upright position. The LOW-7000 is 70 mm tall and weighs 480 grams.

 

The included fan is similar to the Yate Loon D12 series with almost straight trailing edges and sharp axe-like leading edges.

 

There is a small secondary heatsink attached above the base plate, a feature common on Scythe heatsinks. The main body is comprised of 0.33 mm thick fins spaced 2.0 mm apart.

 

Like the NT06-E, the LOW-7000 also has a hard-mounting system, but in this case the screws go in through the back side of the motherboard.

 

Installed on out test platform.

 

Test Results
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1900 RPM	31 dBA	18°C	0.23
9V	1500 RPM	24 dBA	21°C	0.27
7V	1200 RPM	18 dBA	23°C	0.29
5V	840 RPM	12 dBA	28°C	0.36
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better.

At 12V, the fan was very turbulent with a mild droning type character
and far too loud overall at 31 dBA. At 9V it generated 24 dBA, and was much
more tolerable with a smooth but buzzy profile. At 7V it was fairly faint with
some clicking generated by the bearings. The same can be said at 5V, though
it was inaudible at this level from one meter’s distance.

Cooler
Master Geminii S – $35USD

We wouldn’t call the Geminii S a compact cooler, but it is smaller
than the NT06-E with a fan. It is also short enough to fit in some cases that
most third party heatsinks will not (such as the Luxa2
LM100).

The Geminii S and included accessories.

 

At 86 mm tall and weighing in at 500 grams, it isn’t exactly a light weight. Its shape is reminiscent of the formidable Noctua NH-12CP. It sports 5 heatpipes and ships with a 120 mm fan. The fins are extremely thin at 0.28 mm, and are spaced 1.6 mm apart on average.

 

To mount the Geminii S, bolts are screwed tight into the mounting arms before they are installed. The threads on the bolts are reversed, that is they are tightened by turning counterclockwise.

 

The bolts are secured to a backplate on the other side with nuts. The nuts tighten in the traditional way (clockwise), so the nuts and bolts will never get stuck together and rotate together if one side becomes loose.

 

Mounted on our test platform.

 

Test Results (stock fan)
Fan Voltage	Fan Speed	SPL@1m	Thermal Rise	°C/W
12V	1930 RPM	30 dBA	16°C	0.21
9V	1530 RPM	24 dBA	18°C	0.23
7V	1200 RPM	20 dBA	21°C	0.27
5V	800 RPM	14 dBA	29°C	0.37
Test Results (reference fan)
12V	1100 RPM	16 dBA	21°C	0.27
9V	890 RPM	13 dBA	25°C	0.32
7V	720 RPM	12 dBA	32°C	0.40
Thermal Rise: Temperature rise above ambient at load. °C/W: based on the amount of heat dissipated by the CPU (measured 78W); lower is better.

The fan included with the Geminii S was surprisingly good, rivaling the acoustic
qualities of Scythe and Nexus fans. At 12V and 9V, it was turbulent but smooth.
At 7V it developed a bit of a buzz that was only audible close-up, but sounded
completely benign at 1m. At 5V it was inaudible.

COMPARISONS

Note: All the comparison tables below are aligned to the closest SPL levels.

Midget Battle Royale: Silverstone NT07-775 vs. Intel Q9550 HSF vs. Thermolab Nano Silencer
Silverstone NT07-775		Intel Q9550 HSF		Thermolab Nano Silencer
SPL@1m	Thermal Rise	SPL@1m	Thermal Rise	SPL@1m	Thermal Rise
24 dBA	40°C	26 dBA	26°C
20 dBA	44°C			22 dBA	37°C
17 dBA	49°C	18 dBA	40°C	18 dBA	42°C
15 dBA	55°C+*	15 dBA	51°C	15 dBA	49°C
*Testing stopped after thermal rise exceeded 55°C.

The smallest heatsink in our roundup, the Silverstone NT07-775, is also the
worst. Compared to two other two miniature coolers, the Q9550 stock heatsink
and the Thermolab Nano Silencer, it came in last. When each cooler’s fan was
set to generate 17~18 dBA, the NT07-775 lost by a significant 8°C. At 15
dBA, the difference lessened somewhat to 5°C. However the NT07-775 is in
a league of its own due to its extremly slim profile — if you consider
its application, none of the heatsinks we’ve tested are fair competitors.

Comparison: Intel Stock Coolers
Intel Aluminum HSF		Intel Q6600 HSF		Nexus-fan modded Intel cooler
SPL@1m	Thermal Rise	SPL@1m	Thermal Rise	SPL@1m	Thermal Rise
19 dBA	37°C	21 dBA	36°C
14~15 dBA	46°C	16 dBA	48°C	16 dBA	39°C
13 dBA	50°C			13 dBA	45°C
				12 dBA	52°C

The harsh sounding fans equipped on Intel stock coolers may tempt users to
dial down the fan speed to tolerable levels, but unfortunately doing so pays
a steep thermal price. Replacing the fan with a 120 mm Nexus results in much
improvement. At 12V/16 dBA the Nexus fan cooled the CPU 9°C better compared
to the same heatsink with the stock fan speed lowered to an equivalent measured
noise level (though the Nexus actually sounds a lot better). At 9V and 13 dBA,
it posted a 5°C improvement over a standard all-aluminum Intel stock heatsink
running at 6V at the same SPL. While not the most elegant solution, it certainly
is worth consideration if your cooling needs are not great.

Comparison: Arctic Cooling Alpine 7 Pro vs. Intel Stock Coolers
Intel Aluminum HSF		Intel Q6600 HSF (Nexus fan)		Arctic Cooling Alpine 7 Pro
SPL@1m	Thermal Rise	SPL@1m	Thermal Rise	SPL@1m	Thermal Rise
25 dBA	30°C			24 dBA	28°C
19 dBA	37°C			18 dBA	33°C
13 dBA	50°C	13 dBA	45°C
		12 dBA	52°C	12 dBA	42°C

The Alpine 7 Pro performs more or less the same as the Intel stock coolers
when their respective fans at generating sound levels above 20 dBA@1m. At quieter
levels, the Alpine is far superior, boasting 4°C and 1 dBA advantage at
7V compared to the Intel all-aluminum cooler at 9V. At the 12 dBA level, it
delivered a 10°C improvement over the Intel/Nexus combination The wide fin
spacing makes the Alpine an ideal low airflow cooler as long as it is used with
a modest processor.

COMPARISONS (Continued)

Note: All the comparison tables below are aligned to the closest SPL levels.

Comparison: Big Shuriken vs. Nexus LOW-7000 vs. Silverstone NT06-E
Scythe Big Shuriken		Nexus LOW-7000		Silverstone NT06-E (Nexus fan)
SPL@1m	Thermal Rise	SPL@1m	Thermal Rise	SPL@1m	Thermal Rise
23 dBA	21°C	24 dBA	21°C
18 dBA	22°C	18 dBA	23°C	16 dBA	29°C
13~14 dBA	27°C			13 dBA	30°C
		12 dBA	28°C	12 dBA	33°C
11 dBA	34°C			11 dBA	43°C

The Nexus LOW-7000 performed very similarly to the Scythe Shuriken, posting
almost identical numbers at equivalent noise levels. The Silverstone NT06-E’s
performance was more difficult to judge as there were few direct points of comparison
with our reference Nexus fan. At 13 dBA, the CPU ran 3°C warmer than the
Shuriken at 13~14 dBA. At 12 dBA the NT06-E was at a 5°C disadvantage to
the LOW-7000. At 11 dBA, performance was extremely disappointing — it got
beat down by the Shuriken by a sizable 7°C margin.

It would seem that the NT06-E is a slightly less efficient than the Shuriken
and LOW-7000 at higher fan speeds. At low fan speeds, the difference increases
dramatically. Looking at the NT06-E’s design, we would say the combination of
its height and its tight fin spacing is to blame for its poor low airflow performance.
Narrow fin spacing makes it more difficult for the fan to push air through them
— the LOW-7000 doesn’t have that problem. The Shuriken’s fins are also tightly
grouped, but much more of the heatpipe surface area is directly beneath the
fan, and as the Shuriken is almost one inch shorter than the NT06-E, the air
blowing from the fan does not have to travel as far.

Comparison: Big Shuriken vs. Cooler Master Geminii S
Scythe Big Shuriken		CM Geminii S		CM Geminii S (Nexus fan)
SPL@1m	Thermal Rise	SPL@1m	Thermal Rise	SPL@1m	Thermal Rise
23 dBA	21°C	24 dBA	18°C
18 dBA	22°C	20 dBA	21°C
				16 dBA	21 °C
13~14 dBA	27°C	14 dBA	29°C	13 dBA	25°C
				12 dBA	32°C

The Cooler Master Geminii S wasn’t able to clearly defeat the Shuriken, at
least not with the stock fan. At 9V, it generated 24 dBA, and operated 3°C
cooler than the Shuriken producing 23 dBA. At 5V/14 dBA however, the roles reversed,
with the Shuriken posting a 2°C lead despite generating half a dB less at
9V. Pairing the Geminii S with our reference fan greatly improved its low airflow
performance. At 9V/13 dBA it was 1 dB quieter than the stock fan at 5V, yet
managed to cool the CPU by an extra 4°C.

COMPARISONS (Continued)

Comparison at 20 dBA (+/- 2 dB)
Cooler	SPL	Thermal Rise	Height	Weight
Cooler Master Geminii S	20 dBA	21°C	86 mm	500 g
Scythe Big Shuriken	18 dBA	22°C	57 mm	420 g
Nexus LOW-7000	18 dBA	23°C	70 mm	480 g
Intel Q6600 HSF	21 dBA	34°C	62 mm	440 g
Thermolab Micro Silencer	20 dBA	36°C	58 mm	270 g
Intel Aluminum HSF	19 dBA	37°C	63 mm	330 g
Thermolab Nano Silencer (9V)	22 dBA	37°C	39 mm	210 g
Arctic Cooling Alpine 7 GT	20~21 dBA	39°C	64 mm	280 g
Intel Q9550 HSF	18 dBA	40°C	45 mm	260 g
Thermolab Nano Silencer (8V)	18 dBA	42°C	39 mm	210 g
Silverstone NT07-775	20 dBA	44°C	36 mm	300 g
Coolers tested in our last roundup in green.

When dialed down so that their respective fans generate 20 dBA@1m (+/- 2 dB),
it is clear that there are two classes of coolers — those that perform
the same or worse than the stock coolers, and those that are a cut above. The
Shuriken, LOW-3000, and Geminii S easily outpace the competition by more than
10°C. On the other side of the spectrum, the NT07-775 is hopelessly outmatched
due to its size.

Comparison at 15 dBA (+/- 2 dB)
Cooler	SPL	Thermal Rise	Height	Weight
Cooler Master Geminii S (Nexus fan)	16 dBA	21°C	86 mm	500 g
Scythe Big Shuriken	13~14 dBA	27°C	57 mm	420 g
Cooler Master Geminii S	14 dBA	29°C	86 mm	500 g
Silverstone NT06-E (Nexus fan)	16 dBA	29°C	107 mm	550 g
Arctic Cooling Alpine 7 Pro	17 dBA	33°C	86 mm	450 g
Nexus-fan modded Intel cooler	16 dBA	39°C	67 mm	540 g
Intel Q6600 HSF	16 dBA	42°C	62 mm	440 g
Intel Aluminum HSF	14~15 dBA	46°C	63 mm	330 g
Arctic Cooling Alpine 7 GT	15 dBA	46°C	64 mm	280 g
Thermolab Micro Silencer	15 dBA	48°C	58 mm	270 g
Thermolab Nano Silencer	15 dBA	49°C	39 mm	210 g
Intel Q9550 HSF	15 dBA	51°C	45 mm	260 g
Silverstone NT07-775	15 dBA	56°C	36 mm	300 g
Coolers tested in our last roundup in green.

At 15 dBA@1m (+/- 2 dBA), the NT07-775 remains in the basement. The Intel copper
heatsink with Nexus fan takes a small lead amongst the smaller coolers, and
it has unrivaled acoustics. The Alpine 7 Pro impresses, coming within a few
degrees of the more expensive NT06-E, Shuriken, and Geminii S — at $15,
it’s clearly the field’s best budget cooler. The Shuriken and Geminii S prove
to be better than the NT06-E at the extremely limited airflow of the 15 dBA@m test.

MP3 SOUND RECORDINGS

These recordings were made with a high
resolution, lab quality, digital recording system inside SPCR’s
own 11 dBA ambient anechoic chamber, then converted to LAME 128kbps
encoded MP3s. We’ve listened long and hard to ensure there is no audible degradation
from the original WAV files to these MP3s. They represent a quick snapshot of
what we heard during the review.

These recordings are intended to give you an idea of how the product sounds
in actual use — one meter is a reasonable typical distance between a computer
or computer component and your ear. The recording contains stretches of ambient
noise that you can use to judge the relative loudness of the subject. Be aware
that very quiet subjects may not be audible — if we couldn’t hear it from
one meter, chances are we couldn’t record it either!

The recording starts with 10 second segments of room ambiance, then the fan
at various levels. For the most realistic results, set the volume so that
the starting ambient level is just barely audible, then don’t change the volume
setting again.

• Silverstone
  NT07-775 at one meter
  — 8V (15 dBA@1m)
  — 9V (17 dBA@1m)
  — 10V (20 dBA@1m)
  — 12V (24 dBA@1m)

• Arctic
  Cooling Alpine 7 Pro at one meter
  — 5V (12 dBA@1m)
  — 7V (17 dBA@1m)
  — 9V (23 dBA@1m)
  — 12V (28 dBA@1m)

• Nexus
  LOW-7000 at one meter
  — 5V (12 dBA@1m)
  — 7V (18 dBA@1m)
  — 9V (24 dBA@1m)
  — 12V (31 dBA@1m)

• Cooler
  Master Geminii S at one meter
  — 5V (14 dBA@1m)
  — 7V (20 dBA@1m)
  — 9V (24 BA@1m)
  — 12V (30 dBA@1m)

• Nexus
  120mm Real Silent Case fan at one meter
  — 5V (11 dBA@1m)
  — 7V (12 dBA@1m)
  — 9V (13 dBA@1m)
  — 12V (16 dBA@1m)

FINAL THOUGHTS

Our roundup can’t pinpoint a single “best” model, because that depends on just how tight a space you have and how hot a CPU you’re trying to cool. But it should help you choose one that works best for you. Here’s our summary:

Silverstone NT07-775: The smallest CPU heatsink we’ve tested performs
as expected. The heatsink itself is pretty much the same design as Intel’s stock
coolers, but it is even smaller than the stock low profile versions shipping
with Intel’s 45 nm retail box processors. It does generate slightly less noise
at full speed, but the acoustic character is not that much better. It is a specialty
product designed for super slim cases like the LC19
and ML02
— if your case has clearance for something bigger, no reason to consider it.

Nexus-fan modded Intel heatsink: If you’d rather not replace the Intel stock
cooler, pairing it with a quality fan (if you have one handy) is a worthwhile
endeavor. Placing a Nexus 120 mm fan atop a full-sized Intel cooler with a copper
core resulted in an improvement in both CPU temperature and especially acoustics.
With a combined height of 67 mm, it won’t exceed the limit in most cases, and
will provide better cooling for the components around the socket as well. [Editor’s Note: In a Silverstone NT05 case, a stock Intel heatsink modded with a Scythe Slipstream 120mm fan kept an Intel E7200 CPU at 43°C in idle and 78°C after an hour of Prime95 stress testing, while generating a constant 13 dBA@1m, spinning at ~600rpm. This was with 23°C room ambient.]

Arctic Cooling Alpine Alpine 7 Pro: A re-test of the Alpine 7 Pro, this
time in our anechoic chamber, confirms it is a suitable quiet replacement for
a full-sized Intel stock cooler. They perform similarly at high speed, but the
Alpine 7 Pro exerts a substantial lead at lower fan speeds to its wide fin placement.
The character of the noise it generates is also far superior, very smooth and
innocuous. At 86 mm tall, it is nowhere close to low profile, but it will fit
in a lot of cases where taller heatsinks with 92/120 mm fans will not. $15 is
a very fair price for the product.

Nexus LOW-7000: The Nexus LOW-7000 acquitted itself very nicely, performing
on par with the Scythe Big Shuriken. The Nexus is taller, but much easier to
install since it doesn’t rely on pushpins and the extra height insures nothing
on the PCB will interfere with it. Unfortunately at $55, it caries a $25 price
premium over the Shuriken. Furthermore, the included fan, while much better
acoustically than many stock models, isn’t nearly as smooth as the Slipstreams
that Scythe ships with its heatsinks.

Silverstone NT06-E: The other Silverstone in our roundup was also a
disappointment. It has a solid mounting system, but its performance was underwhelming
considering its cost and physical dimensions. Even using our reference Nexus
fan (which typically performs better than most fans) it failed to keep pace
with the Shuriken and LOW-7000. When the fan speed was reduced, it fell even
further behind. It retails for $55, which is the same as the LOW-7000, but you
also need to purchase a fan as Silverstone does not include one. It seems that
they designed it specifically to be used fanlessly in a Sugo
SG05/06 where the power supply fan is directly above it — if you add
a fan on top, it is too big to fit in most low profile and cube-style cases.

Cooler Master Geminii S: The Geminii S performed fairly well, keeping
pace with the Shuriken and LOW-7000 until fan speeds were decreased drastically.
It’s affordable, has a good solid mounting system and ships with a surprisingly
good fan with excellent acoustics. If you have enough space to accommodate the
Geminii S, it is a worthwhile investment.

The big winner from the first part of our roundup, the Scythe Big Shuriken,
edged out our honorable mentions, the Nexus LOW-7000 due to its lower price,
and the Geminii S due to its much shorter profile. However it is nice to have
alternatives which take aim at the Shuriken’s shortcomings, namely its frustrating
mounting system and lack of clearance. Still, the Shuriken undoubtedly delivers
the best value of the bunch, providing excellent performance, a smooth quiet
fan, and a price-tag that won’t make you cringe.

Our thanks to Silverstone,
Arctic Cooling,
Nexus, and Cooler
Master for the heatsinks used in today’s roundup.

* * *

Articles of Related Interest

Scythe Top-Down Coolers:
Kabuto vs. Zipang 2
LGA775 Low Profile Heatsink
Roundup
Scythe Mugen-2 CPU Cooler
Scythe Katana 3: Same slant, new version
CNPS10X Extreme: Zalman’s Extreme Makeover
ZEROtherm CORE92 Direct-Touch CPU
Cooler

* * *

Discuss this article in the
SPCR forums.