Asus Eee Box B202: An Atom-based mini PC

Complete|Mobile Systems
Viewing page 5 of 6 pages. Previous 1 2 3 4 5 6 Next


The Eee Box idled at just 16W. It drew 6W more when running two instances of Prime95 or playing hi def video clips. Power consumption increased to 25W when the IGP was stressed with ATITool while Prime95 was running.

These numbers are very modest, especially in comparison to those obtained with the Intel D945GCLF mini-ITX board reviewed last week. That system drew 27W in idle, 31~33W playing video, and 38W at full tilt; all this with a PSU that's one of the most efficient we've tested at low loads. That the Eee Box lowers power consumption by over 10W at all states is excellent. What these numbers suggest is that the 945GSE Express chipset used in the Eee Box is considerably more efficient than the 945GC of the Intel mini-ITX board.

A summary page at Intel of the 945GSE Mobile Express chipset states that the maximum TDP of the chipset itself is just 6W. On the same page, under the heading "Valid Processor Combinations", one combination is listed: The Atom N270 and 945GSE Express Chipset with 82801GBM I/O Controller Hub (ICH7M), with a "System TDP" of just 11.8W. This explains why the Eee Box is so much more energy efficient than the Intel D945GCLF.

It even suggests that there may be room for further reduction of power with a more efficient AC/DC adapter: If the total TDP of the system is only 11.8W, and the AC power measured is 25W, then the power supply is working at less than 50% efficiency. Alas, it may not be quite so simple. The PSU is also supplying power for the wifi and the hard drive, and perhaps a bit to the USB keyboard and mouse as well.

As it stands, with a typical 19~20" LCD monitor set to medium brightness, the total power consumption of an Eee Box system would be no more than ~50W on average, and possibly as low as 40W depending on your LCD monitor. Those are compelling low power consumption figures.

Test Results: Asus Eee Box B202
Test State
CPU Usage
Power (AC)
Standby (S3)
Dark Knight
Rush Hour
Coral Reef
Prime95 + ATITool
Grey boxes indicate test failure.

As with the D945GCLF test system, we weren't able to run our typical video test suite. The system is inadequate for playing any 1080p content. Not only did the audio skip during playback, but there were frequent pauses and massive frame-loss. The most demanding clip the system could play flawlessly was a 720p H.264 trailer for "Dark Knight." It generally did fine with most 720p clips as well.


Everyday tasks such as web browsing, checking emails, and uploading photos from an SD card felt little different from most other PC systems in the lab. Standard desktop functions like opening windows, searching for files, accessing other PC folders through the wireless network — all these felt perfectly normal. Sure, it's not as speedy as the quickest quad-core, 4GB RAM, VelociRaptor driven PC that's under my desk these days, but neither did it feel like a slug.

Trying to do any "serious" image editing with large photos in Photoshop was quite slow, however; nothing beyond a quick "optimize for web" process is recommended here. Ditto with video editing; the Eee Box lacks the horsepower to make such work smooth or pleasant.


When the Eee Box is turned on, a small fan (presumably a blower of some type) starts with a short 1-2 second burst of high speed, then slows right down to very low speed. The noise of this fan and of the 2.5" hard drive is audible from a couple feet away, but it's pretty quiet. Asus states that it's "barely audible — 26 decibels", but no other details about the noise measurement are given.

In a quiet, carpeted home office (10' x 20' x 8') with an ambient level of 19 dBA, the SPL when the Eee Box was powered up (on a desk) measured 22 dBA. Under a wide variety of loads with the room temperature at 22~25°C, the measured SPL of the Eee Box hardly varied, rising perhaps 1 dBA under the worst conditions. The subjective impression did change under such load, as it was caused mostly by the fan spinning faster, and a tonal peak was noticed at ~420Hz when the fan was spinning at its highest speed. Under all the various normal load conditions, the fan never reached the speed or noise level noted when the automatic fan control was turned off in the BIOS.

Mounted on the back of the monitor, atop table on our hemi-anechoic chamber.

Measured in the 11 dBA ambient of our hemi-anechoic chamber, the SPL at both idle and peak was far lower than the 26 dB specified by Asus. It's possible that this specification is in unweighted decibels; i.e., without the A-weighting normally applied to simulate human perception of SPL.

Measured / Perceived Noise
in anechoic chamber
Sleep n/a Silent
Idle 18 [email protected] Extremely quiet; audible from within a couple of feet but benign, with no tonal aspects.
Full load 19~20 [email protected] Still quiet, but with a bit of tonality, centered at around 420Hz.
Idle, on monitor 14 [email protected] - from front of monitor The monitor blocks much of the noise. If the wall behind is some distance away or covered with damping material such as a curtain, the subjective noise is even lower than measured.
Full load, on monitor 15~16 [email protected] - from front of monitor See comments above. The tonality at 420Hz is still there, but substantially muted by the monitor.
Q-fan off; fan on full speed 37 [email protected] Mostly broadband noise but loud, with some tonality in the mid-band. Fan runs at full speed. There's no reason to ever turn Q-fan off.


SpeedFan gave us some numbers to look at, despite the absence of any hardware monitoring in the Eee Box BIOS. The changes between idle load and maximum load were small, like the changed in the fan noise. SpeedFan indicated 2900 RPM at idle and 3200~3300 RPM at maximum load, with temperature rises in the CPU and "system" of just 2~4°C. The hard drive temperature ranged 45~48°C. All this was in an ambient room temperature of 23~26°C. Asus has done a fine job of achieving good cooling with minimal noise in the Eee Box.

SpeedFan screen capture with Eee Box at full load.

Previous 1 2 3 4 5 6 Next

Complete|Mobile Systems - Article Index
Help support this site, buy from one of our affiliate retailers!