FrostyTech Mk.II
Synthetic Thermal Heatsink Test Results
Intel Pentium D / Extreme
Edition & Quad Core class heatsinks tested at 150W Intel Core
2 Duo / Pentium 4 / Celeron class heatsinks tested at
85W Results based on readings
from a 30x30mm copper interface die with 150W and 85W heat loads
applied on the LGA775 version of the Mk.II Test Platform.
Temperature results listed for each heatsink were calculated by
determining the rise over ambient temperatures at time of test.
Information on Frostytech's test methodology is available
here. |
 |
| Socket 775
compatible CPU heatsinks | |
| Manufacturer: |
Model No.: |
Fan Speed: |
150W Thermal Test*
(°C) |
85W Thermal Test
(°C) |
Noise Level (dBA) |
|
| Thermalright |
Ultra 120 |
|
17.8 |
10.6 |
53.7 |
| Zalman |
CNPS9700 LED |
high |
19.0 |
10.4 |
56.5 |
| Thermalright |
Ultra-90 |
high |
19.6 |
13.0 |
56.2 |
| Enzotech |
Ultra-X |
high |
20.0 |
8.5 |
59.1 |
| Apack |
Zerotherm BTF90 |
|
21.8 |
12.7 |
52.1 |
| Apack |
Zerotherm BTF80 |
|
22.8 |
13.9 |
51.9 |
| Cooljag |
Falcon 92-AL |
high |
23.2 |
14.2 |
52.1 |
| Rosewill |
RCX-Z3 |
high |
23.3 |
14.0 |
58.6 |
| Scythe |
Mine Rev. B |
|
23.6 |
15.0 |
36.8 |
| Cooljag |
Falcon 92-Cu |
high |
23.6 |
14.8 |
52.8 |
| Akasa |
AK-920 |
high |
24.3 |
14.1 |
52.9 |
| Asus |
Arctic Square |
high |
24.9 |
14.9 |
44.9 |
| Akasa |
AK-922 (EVObue) |
high |
25.5 |
15.4 |
60.3 |
| Thermalfly |
F1-IACSHA |
high |
29.0 |
18.3 |
60.4 |
| Thermalfly |
I75DUB |
high |
29.5 |
16.8 |
68.5 |
| Zalman |
CNPS9500 AT |
|
29.6 |
17.3 |
53.3 |
| Alpha Novatech |
PRE9060M92P |
high |
29.9 |
18.5 |
51.5 |
| Scythe |
Kama Cross |
|
30.0 |
18.8 |
40.0 |
| Akasa |
AK-961 |
|
30.0 |
18.7 |
60.7 |
| Rosewill |
RCX-Z2-EX |
high |
30.1 |
18.7 |
45.9 |
| nPowertek |
NPH-775-1 |
|
30.1 |
18.8 |
50.1 |
| Sunon |
Waturbo |
|
30.6 |
18.3 |
57.5 |
| Apack |
CF800 |
high |
31.4 |
20.7 |
48.4 |
| Thermaltake |
Jungle512 |
high |
31.6 |
20.4 |
60.8 |
| Xigmatek |
AIO-S80DP |
high |
32.4 |
21.4 |
55.0 |
| Scythe |
Katana II |
|
32.5 |
18.6 |
37.9 |
| Rosewill |
RCX-Z2 |
|
32.6 |
19.3 |
59.9 |
| Zalman |
CNPS7700-Cu |
high |
32.9 |
19.7 |
50.7 |
| Apack |
CF900 |
high |
32.9 |
20.3 |
48.3 |
| Evercool |
PT02-9232CP |
|
33.3 |
19.9 |
63.3 |
| Zalman |
CNPS7700-Cu LED |
high |
33.5 |
20.6 |
49.8 |
| Intel |
Stock Pentium D |
high |
33.8 |
20.7 |
47.5 |
| nPowerTek |
SF775-2 |
high |
34.7 |
21.1 |
64.6 |
| Akasa |
AK-960E |
|
36.1 |
21.4 |
56.3 |
| Thermalfly |
I75DUH |
|
37.8 |
22.8 |
48.7 |
| Cooljag |
OAK-H |
high |
38.5 |
22.9 |
55.7 |
| Cooljag |
OAK-H2 |
high |
38.6 |
24.1 |
55.6 |
| LS Cable |
SHS-T700 |
|
38.7 |
22.8 |
48.5 |
| Rosewill |
RCX-Z3 |
low |
42.4 |
25.4 |
45.6 |
| Cooljag |
OAK-H |
low |
95.3 |
54.5 |
29.5 |
|
|
|
150W Thermal Test
(°C) |
85W Thermal Test
(°C) |
Noise Level (dBA) |
| *Heatsinks are ranked according to
the 150W thermal test results column (rise over ambient temp.). Low
temperatures with low noise levels are considered best. For reference
heatsinks with variable-speed fans, only the high speed (12V) fan test
result is included in the comparison sheet; more detailed results reside
in each specific heatsink review. |
Intel processors vary widely in heat output between single and multi-core processors, so Frostytech relies
upon on two different thermal values to represent the upper limits of existing Intel single and
multi-core CPU families. First off, a 150W heat load was applied by the
Intel LGA775 version of FrostyTech's synthetic
thermal test platform to the Rosewill
RCX-Z3 heatsink. This yielded a rise above ambient temperature of 23.3°C,
with 58.6dBA noise. That's a good result, but not low enough to be considered
top of the line... and nearly 60 dBA is somewhat loud.
When the 92mm PWM fan speed
was dropped down to about 1200RPM,
temperatures rose to 42.4°C over ambient, while noise levels fell down to a
much more reasonable 45.6 dBA. When faced with a 150W heat load,
Rosewill's RCX-Z3 heatsink handles it well, so long as air flow is sufficient.
The Rosewill RCX-Z3 heatsink was next
tested with an 85W heat load. In this case the heatsink maintained a good
14.0°C temperature over ambient at full fan speed. As fan speed was reduced to low
noise mode, the heatsink reacted in a predictable manner - the synthetic CPU die
temperatures rose to 25.4°C.
Clearly as
we can see from the synthetic thermal test results here, the Rosewill RCX-Z3 heatsink
is only really able to shine when its fan is loudly spinning away at full RPM.
As the volume of airflow passing through its many aluminum fins
decreases, the heatsink quickly looses effectiveness and the plastic fan shroud starts trapping
the heat in. In the end, we're left wondering if it would be better ditching
the blue LEDs and plastic fan shroud entirely so the heatsink could potentially
be operated with a quieter fan. As it stands, the Rosewill RCX-Z3 is only
worthwhile when going at full tilt, and then it is just too loud for modern day
computer users to accept.
Related
Articles:
For more reviews on the latest heatsinks and
cooling solutions, rely on FrostyTech's inventory of 350+ heatsink reviews. Here are a
few other articles that you might enjoy as well...
- Xigmatek
AIO-S80DP Self-Contained Watercooling Heatsink Review
- TTIC nPowerTek
NPH-K8 Big Heatsink Review
- LS Cable SHS-T700 Heatsink Review
- TTIC nPowerTek
NPH-K8 Big Heatsink Review
- Enzotech Ultra-X Heatsink Review
- Akasa EVO 120
AK-920 Low Noise Heatsink Review
- Akasa AK-960E
Socket 775 Heatsink Review
- Apack Zerotherm
CF900 Socket 775 Heatsink Review
- Alpha Novatech
PRE9060M92P Socket 775 Core 2 Duo Heatsink Review
- Zalman
CNPS7700-Cu LED Low Noise Heatsink Review
- Zalman CNPS9700
LED Low Noise Heatsink Review
- Thermalfly I75DUB Bifurcated Fin Socket 775 Heatsink Review