Articles: Cooling
 

Bookmark and Share

(4) 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 ]

Cooling Efficiency Tests

The results of our cooling efficiency tests are given on the diagram and in the table below:

Working with their default fans (the first block in the diagram), the Zalman CNPS11X Extreme is 1°C inferior to the CNPS10X Performa at 800 RPM but 3°C better than the latter at 1000 RPM. The 200RPM increase in speed resulted in a temperature reduction of 8°C with the CNPS11X Extreme and 4°C with the CNPS10X Performa. The difference is due to the denser pack of fins in the V-shaped heatsink. The same goes for the higher speeds: the CNPS11X Extreme is 5°C ahead of the CNPS10X Performa at 1200 RPM. The latter cooler can only catch up with the former at 1600 RPM.

The performance of the CNPS11X Extreme does not improve much at 1400 RPM and 1600 RPM: switching to the higher speed leads to a 1°C decrease in CPU temperature. And we can lower the temperature by 2°C more by switching to 1800 RPM and then to the max speed.

The new cooler is overall more effective than the Performa (by 3 to 5°) in the speed range of 1000 to 1600 RPM. On the other hand, the CNPS11X Extreme can't match the Thermalright Archon with its default fan and at a comparable level of noise. Can this be changed by installing an alternative fan on its V-shaped heatsink?

Well, as you can see from the next block of results in the diagram above, replacing the CNPS11X Extreme's default fan with a Thermalright TR-FDB didn’t improve its performance. On the contrary, the cooler’s performance worsened a little. The temperature remains the same at the min and max speeds of the fan (800 and 2000 RPM) but the default fan is 1-2°C better than the TR-FDB at the medium speeds. The V-shaped heatsink of the CNPS11X Extreme does not allow to install a second fan whereas the CNPS10X Performa supports two fans simultaneously and easily catches up with and even overtakes the CNPS11X Extreme at 1600 and 2000 RPM despite the latter’s costing twice more money.

However, I’ve got another fan to test the CNPS11X Extreme with. The Panaflo H1A is a top-performance fan with a thickness of 38 millimeters. When equipped with it, the cooler lowers the CPU temperature by 4°C at 800 RPM compared to its own performance with the default fan. The advantage is 3°C at 1000 RPM, 1°C at 1200 RPM, and 2°C at 1400, 1600 and 2000 RPM. The Panaflo H1A lowers the temperature by only 1°C at the maximum 2430 RPM, yet it has already done its job well at the lower speeds. Thus, if you equip your Zalman CNPS11X Extreme with a high-performance 38mm fan, you can make it about as effective as a Thermalright Archon with two TY-140 fans!

Next I tested the Zalman CNPS11X Extreme for its ability to help in overclocking my six-core CPU. I could overclock the CPU to 4400 MHz at a voltage of 1.45 volts at the cooler’s fan speed of 1200 RPM as well as 1980 RPM (the peak temperature of the hottest CPU core was 8°C lower at 1980 than at 1200 RPM).


Zalman CNPS11X Extreme at 1200 RPM

Zalman CNPS11X Extreme at 1980 RPM

The CNPS11X Extreme couldn’t make the CPU stable at a higher clock rate and voltage with its default fan. When equipped with a Panaflo H1A working at 1200 RPM, it could keep the CPU running at 4450 MHz with a voltage of 1.46875 volts. The peak CPU temperature was 90°:


Zalman CNPS11X Extreme at 1200 RPM (Panaflo H1A fan)

At the maximum speed of the default fan the Zalman CNPS11X Extreme helped overclock the CPU to 4475 MHz at a voltage of 1.46875 volts. This is close to the best result my CPU can show with air cooling. The CPU temperature was 81°C. With a Panaflo H1A at 2430 RPM the CNPS11X Extreme reached the same result in this test.


Zalman CNPS11X Extreme at 1980 RPM

Zalman CNPS11X Extreme at 2430 RPM
(Panaflo H1A fan)

This is the limit of the CNPS11X Extreme, yet it is quite a high limit. For example, I achieved the same results with a Thermalright Archon with two 140mm TY-140 fans working at 1280 RPM.


Thermalright Archon with 2 fans at 1280 RPM

Of course, the CNPS11X Extreme is going to be noisier. Let's see how much noisier, exactly.

 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 ]

Discussion

Comments currently: 4
Discussion started: 05/15/11 10:04:21 AM
Latest comment: 01/10/14 07:19:03 AM

View comments

Add your Comment