Bookmark and Share

Articles: Mobile

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 ]

To check up this cooler we carried out a small experiment. We started the assembled system up having disabled the 80mm fan. If there’s really bad contact between the cooler’s sole and the CPU, the CPU temperature should jump suddenly. Or the system wouldn’t start up at all due to overheat. The barebone did start up, however, and we ran the monitoring program from the SiS SuperUtility Pro bundle (we’ll talk about it shortly) and – a miracle! – the temperature of the idle CPU didn’t change even by a single degree. It remained a mystery if the thermal paste included with the barebone or anything else was the reason.

Without the fan the cooler behaved like many other passive coolers on heat pipes do. Our launching S’n’M led to a sudden increase of the CPU temperature till it got too high and the system shut down automatically. On the other hand, a majority of quality passive coolers can keep the temperature just below the highest limit, though. So, that’s another indication that the quality of the E-bot’s CPU cooler is below average.

Another device that often suffers from lack of cool air in mini-systems is the hard disk drive. In the E-bot, the hard drive is placed at the bottom of the case so that the cool air coming through the vent holes in the bottom cooled it the first thing.

You can see the results of our temperature tests below, but we want first to say a few words about the noise parameters of this barebone. You can reduce the noise of an air-based cooling system by reducing the speed of the fans or by decreasing the friction in the fans’ engines. Friction is a parameter that can hardly be improved, so low speeds is the only way to building a quiet computer. The E-bot offers you two options: to set the fan speeds at their maximums, or to set them to some average value which is not changed irrespective of the load on the system. You can do this in the BIOS. In the first case the E-bot sounds like a taking-off jetliner, yet the high speeds don’t practically affect the cooling efficiency. In the second case the noise is more tolerable. The external power supply brings more noise when its fan reaches its max speed from time to time, and you cannot control that. But overall the noise from the system is quite acceptable.

Here are the temperatures of the system in different modes:

FSB frequency

Idle

S’n’M

3Dmark 2001SE

HDD temperature

CPU

SYS

CPU

SYS

CPU

SYS

200MHz

25

34

51

36

38

35

37

212MHz

25

34

52

36

40

35

37

All temperatures are given in Co

The testbed was configured as follows:

  • Intel Pentium 4 3.2GHz (Northwood)
  • 2x512MB OCZ PC4000
  • E-bot barebone
  • The cooler enclosed with the E-bot

The speeds of the fans remained constant during the tests: 2860rpm of the CPU fan and 5444 of the North Bridge fan. The increase of the temperature by 20°C means the fan is not fast enough to cool the CPU normally. That’s no good news for the cooling system of the E-bot.

Now we want to add a few words on the monitoring program you receive with the E-bot in the SuperUtility Pro bundle.

The program is called SuperStep and does the following: it keeps track of the system’s two thermal diodes and the speeds of the fans, allows setting the FSB frequency, displays the effective frequencies of the CPU as well as the AGP and PCI busses, and outputs all the basic voltages. It can also warn the user when some parameter is out of the norm. That’s an extensive list of functions of course. The only things SuperStep cannot do are keeping log files or drawing statistical graphs.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 ]

Discussion

Comments currently: 8
Discussion started: 05/05/05 05:12:07 PM
Latest comment: 11/23/07 11:25:30 AM

View comments

You must log in to add comments.

Forgot password? Registration

remember me