NoiseControl 1888

Written by Stein Magnus Jodal
Published 26 August 2001

Cost: NOK 480, USD 50
Where From: Microplex
Manufacturer: NoiseControl

Today I'm going to take a closer look at the NoiseControl 1888 Slot A fan. It's German, silent, made out of metal and - it's supposed to cool down CPU's. The whole thing looks pretty much like a rectangulary metal thingy with a nice black fan on. The big question is: does it help my poor little classic Athlon 800 MHz?

Benchmarks

I've compared the NC1888 to my previous fan, the well known Thermaltake Golden Orb. The benchmark rig was an Addtronics 7896A with five extra fans, three 80mm, one 92mm and one 120mm. I gave the fans, which are supposed to run at 12 volts, 7-9 volts with katmai's fancontroller.

The CPU, an Athlon 800 MHz, has been the victim of some of my previous experiments, and has thus no plastic cover. It's just the coolingplate and the PCB, which probably makes it a little bit cooler than the normal pack. The tempprobe, standard with Abit mobos, was attached to one of the two clips keeping the PCB and the coolingplate together. This is both closer to the core than normal, and does not decrease the heat transfer from the CPU to the coolingplate, like the normal placement between the coolingplate and the fan does.

Since the simple way often is the best I did everything plain simple. Both fans were tested at 800 MHz (original setting), 900 MHz (100x9.0) and 1 GHz (100x10.0). The CPU was overclocked with a GFD. I used standard volt settings at 800 MHz, 1.70 volt to the core and the I/O at 3.30. At 900 MHz and 1 GHz I used 1.80/3.30 because I've always had to set the core volt a bit higher when overclocking earlier. After I finished the benchmarking I started my computer at 1 GHz and forgot to set the core to 1.80 again, and it has been running at 1.70 without problems since. I can't see much difference in the temperatures with different volt settings on the same speed, but it's always a good thing to keep the volt down.

The idlepart is the temperature after a restart and five minutes of nothing, while 100% is the temperature after five minutes of heavy load using CPU Stability Test 6.0.