Viewing page 3 of 3 pages.
If you have read SPCR's review of the VIA EPIA-M10000 Mini-ITX board, then the performance of the Hush Mini-ITX will be no surprise. The performance is entirely dictated by the VIA mainboard, which has the critical CPU and both video & audio subsystems embedded. Connectivity is on par with any modern PC, with 10/100 LAN, multiple USB 2.0 and firewire ports all built right into the board.
The VIA EPIA-M10000 does not fare well with conventional benchmarks such as Sisoftware Sandra, PCMark and others. They show the unit to be roughly equivalent in performance to an Intel Celeron or P3, 500~600 MHz.
In actual use with web browsing, e-mail, ordinary office work, creating web pages or watching DVD movies, the Hush Mini-ITX performs perfectly well, without any sense of being underpowered even in comparison with >2.5 GHz Intel and AMD processor systems. It's only when one tries to perform complex image editing work with big files in Photoshop, serious video-editing or playing 3D games that the lack of horsepower is noticeable.
When the system was stressed for over an hour using CPUBurn software, the CPU thermal diode reported a temperature of 68°C in a room of 27°C. This may seem a bit high, but it is not really for the C3 processor core in this board. The stand-alone C3 has shown amazing survivability in high temperatures, including a successful 24-hour gaming run without any heatsink. (Look for this video in VIA's web site.)
The cover was removed towards the end of this test. A finger on the CPU cooling block was instictively yanked back with alacrity. Yes, it was very hot. The heatpipe was not quite as hot, and the heatsink was cooler still, even at the back corner where the heatpipe is clamped. The temperature of the heatsink actually dropped very close to ambient at the front end of its long length. A couple of different conclusions could be reached:
- The heatpipe is not transferring as much heat as it could to the heatsink.
- The cooling system is just loafing under the mere 15W load of the CPU; the heatsink is capable of far greater cooling.
In any case, the system remained perfectly stable throughout the stress testing, and nothing on the outside of the case became uncomfortably hot.
1) Anechoic Chamber Measurements
Hush Technologies have gone a little further with their claims about low noise than most PC makers. They subjected the Hush Mini-ITX PC to serious acoustic testing. In a rather unusual move, they've made the complete acoustic report available as a PDF file for download from their website.
We are intimately familiar with the contents of that report, having been present during the acoustic measurements of the Hush Mini-ITX in the anechoic chamber. In fact, most of the photos shown here are of the sample actually tested for the acoustic report.
SPCR was instrumental in bringing the Acoustics Lab at the University of British Columbia (UBC) and Hush Technologies together. My involvement with a fan noise research project at UBC was taking me to their labs frequently when this Hush Mini-ITX sample arrived. It was the first completely fanless production PC I'd seen. One thing led to another, and Hush ended up contracting the UBC Acoustic Lab to conduct a complete sound power test in accordance with standard ISO 7779. I participated in the testing and helped to draft the report.
The gist of the report results:
|Mode of Operation
SPL at Operator Position**
|Hard drive seeking
* For a discussion of sound power, please see page 2 of Noise in Computing: A Primer
As per ISO 7779: The unit was placed on a table 0.75M tall, with the microphone positioned 0.5M in front of the unit and 1.2M above the floor.
In case there is any doubt, this is outstandingly quiet measured acoustic performance!
2) Subjective Perceptions
The noise of the Hush Mini-ITX PC in normal use is extremely low. In general usage, it is quiet enough to be inaudible for anyone not seated directly in front of it. Any noise beyond the residual ambient even in a quiet home -- normal conversation, music in the background, noise from the street, people in the kitchen -- is enough to mask the noise of the Hush.
What noise it does produce comes entirely from its drives -- the hard drive and the optical drive. The latter is not particularly quiet, In comparison with desktop form factor DVD drives from Samsung, LG and Toshiba, it is audibly noisier, perhaps by as much as 5~6 dBA/1meter during DVD movie playback. (For CD access, it is at around the same loudness level.) Keep in mind that during DVD movie playback, the volume of the move sound track would be at a far higher level than 4.2 bels, more than enough to obscure the DVD drive at virtually any distance.
Because optical drives make noise only when accessed, the noise they make, even when loud, is not usually as disturbing as other PC noises. The user is more in control of the noise. The user knows it is temporary (during software installation, data transfer, a burn, gameplay) and will stop as soon as the optical disc is removed.
Despite its vanishingly low noise level, the Hush suffers the disadvantage of its intended placement: On the desk rather than under, perhaps acting as a rest or base for a LCD monitor, which it would match stylistically. On the desk, it is much closer to the operator's ears than a tower style PC placed on the floor under the desk. The Seagate Barracuda IV 40G hard drive in this Hush is arguably one of the quietest hard drive ever made. Yet, in the Hush Mini-ITX PC, when situated on top of the desk as intended, it is clearly if softly audible almost all the time in a quiet home office. It is normally a soft "whirring" noise.
The measured increase in noise when the drive is seeking is about 2.5 dBA at the ISO 7779 defined "operator position", which is essentially ~0.6 meter from the front of the PC. Although a 2.5 dBA change is small and the measured <22 dBA at .6 meter noise level is extremely low, it is still plainly audible. Few would fail to hear the distinctive clickety-click noise when the drive seeks and writes.
Ironically, part of the problem is the absence of any wind noise from cooling fans. The white/pink broadband noise (shhhhhhhhhhhh or phhhhhhhhhhhh) of fans in ordinary PCs helps to mask hard drive noise. In the Hush, any drive noise and any change in noise stands in stark relief against the background of zero fan noise, and so is psychoacoustically easy to perceive.
Improving the HDD Damping
The main PC in my home office is equipped with not one but two Seagate Barracuda IV 40G hard drives. This is a carefully custom-built mid-tower system that is extremely quiet. It sits under the desk that the Hush sat upon while being user tested. Despite there being two drives in this PC compared to the one in the Hush, no drive noise is audible from where I sit, and seek noise is also inaudible. Why?
- It is positioned at least twice the distance from my ears, compared to the Hush on the desk
- These 2 drives are suspended in elastic webbing, which allows NO drive vibrations to be transmitted into the case. (Drive suspension is detailed in the article Hard Drive Silencing: Sandwiches & Suspensions)
Examining the mounting system of the HDD in the Hush, it seemed that the damping effect of those gray mushy pads was being mechanically short circuited, so to speak, by the direct contact of the aluminum cover piece held down with screws and wedged into the side slot. (See discussion above.)
In an effort to reduce the HDD noise, the aluminum cover piece was removed altogether, the two top-side pads put on top of the bottom ones for double thickness, and the drive set down on top of the now double-thick pads. Care was taken to ensure that no part of the drive made contact with anything in the case (even cables) other than the soft pads before the cover was closed. Given the sticky nature of the pads, there was virtually no risk whatsoever of the drive being dislodged or moved even with an accidental knock.
The reduction in HDD noise was immediately apparent, at idle as well as in seek. A slight higher pitch aspect to the whirring that had been there before was noticeably absent, and the overall level was lower. (How much lower I cannot say, as no sound level meter was available at the time.) Seek could still be distinguished when it occurred, but much subdued, much like when the string damping pedal is engaged on a piano.
Please Note: The above comments about hard drive noise are not meant to suggest that most consumers would find the Hush excessive in this regard. The Hush is extremely quiet by almost any standard. However, the experiment with HDD mounting shows it can be improved still further.