AOpen EY855-II Pentium M SFF barebones system

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

TESTING & MEASUREMENTS

No performance benchmark tests were run on this system. The results cannot vary by more than a percent or two from the tests already documented in our review of 855-chipset P-M desktop boards, or in reviews of 855 Pentium M platforms by other hardware web sites. The choices of processor, memory, video card and hard drive will affect the outcomes far more than any variances in motherboard performance.

We encountered a serious problem in testing. None of the standard temperature monitoring tools such as SpeedFan, Motherboard Monitor or even AOpen's own hardware monitor utility would provide any accurate readings off the sensors in the EY855-II. The only temperature we could monitor at all was that of the hard drive, which stayed at 31~33°C throughout the testing. The good thing is that temperature monitoring is probably completely unnecessary because the Pentium M -- and this system as configured -- runs so cool. Still, this is not comforting to most advanced users, who prefer to know more about the thermal conditions in their PC.

The fan control in the BIOS was set to Smart Control throughout the testing. The CPU alarm was set to the minimum temperature of 70°C.

Power and Acoustics

The total AC power draw was measured using a Seasonic Power Angel AC power meter. The ambient temperature was 23°C during testing. Because the CPU alarm never went off, we know that the CPU always stayed below 70°C.

Activity
AC Power
SPL*
Idle in Windows XP
34W
25 dBA@1m
Max Load (CPUBurn, 30 min)
45W
25 dBA@1m
PSU fan manually stopped
-
20 dBA@1m
* Sound Pressure Level measured 1 meter from front panel in 17 dBA ambient room.
The ISO 7779 "Seated Operator Position" SPL reading puts the mic 0.5m in front of and 0.45m above the table top. Our 1m SPL measurement is 2.5~3 dBA higher.

The noise level of the CPU cooling fan and the PSU fan remained constant at any load during testing. This is reflected in the 25 dBA@1m reading obtained for both idle and maximum CPU load in the table above. The noise level never changed in a couple of weeks of general use.

The overall noise can be characterized as quiet, but with a trace of humming or buzzing quality that is related to case panel vibrations, caused mostly by the PSU fan. Most of the noise in our test system came from the power supply fan, whose minimum noise level is considerably higher than that of the CPU heatsink fan or the Samsung HDD we used. The effect was easily heard by stopping the PSU fan with a plastic zap strap (cable tie) jammed into the fan blades; the level dropped by 5 dBA@1m. The PSU fan makes turbulence noise, bearing noise, and also causes some case vibration noise.

Given the low power draw of the Pentium M processors and the very modest maximum power demand of this system even with the most power hungry of VGA cards, AOpen could have done more to optimize the PSU for quieter operation. There is no need for a 275W PSU. In fact, a 120W PSU would probably suffice. They could have focused more on improving PSU efficiency, tailoring the efficiency curve of the PSU to the low power demand of the system, and replaced the PSU fan with a lower speed unit with higher quality bearings. They could have done these things, but they did not.

Audio Recording of the Test System

High resolution audio recordings were made of the system running normally and also with the PSU fan temporarily stopped. These recordings were combined into a single 20-second MP3 file. The first 10 seconds is the noise of the system running normally; the last 10 seconds is with the PSU fan stopped. The difference should be very clearly audible on any audio playback system.

MP3: AOpen EY855-II w/ P-M 1.6 GHz & suspended Samsung notebook drive; 10 seconds normal, 10 seconds w/PSU fan stopped

Comparisons

MP3: Arctic Cooling Silentium T2 with test system (3.5" HDD suspended) at idle - 23 dBA/1m

MP3: Arctic Cooling Silentium T2 (3.5" HDD suspended) at max load - 34 dBA/1m

MP3: Shuttle XPC SB86i with Samsung Notebook Drive, Idle: 29 dBA/1m

MP3: Shuttle XPC SB86i with Samsung Notebook Drive, Load: 33 dBA/1m

MP3: Soltek QBic EQ3901-300P, Blower Low, HSF Ramping from Low (33 dBA/1m) to High (39 dBA/1m)

SPCR MP3s: HOW TO LISTEN & COMPARE

The recordings above were made with a high resolution studio quality digital recording system. The microphone is 3" from the edge of the fan frame at a 45° angle, facing the intake side of the fan to avoid direct wind noise. The ambient noise during all recordings is 20 dBA or lower.

A quick and simple way to use these recordings for valid listening comparisons is to play the quietest recording on only one speaker (or a pair of headphones) and set the volume so it is just barely audible a meter away. You must turn off any special sound effects, and set equalizer / tone controls to neutral or flat. Don't touch the volume setting afterwards, and use the same one speaker when you listen to any of the other files. The end result will be reasonably close to the actual recorded sound levels.

Here is a recording of a very quiet sound that is barely audible from 1 meter away even in a super quiet room.

For full details on how to calibrate your sound system to get the most valid listening comparison, please see the yellow text box entitled Listen to the Fans on page 3 of the article SPCR's Test / Sound Lab: A Short Tour.

CrystalCPUID

This sophisticated utility, which Jan Kivar covered in a recent feature article, allows direct user control over dynamic CPU clock speed and voltage changes for processors that support such functions. This includes AMD processors with Cool 'n' Quiet as well as Intel processors with Enhanced Intel Speedstep Technology. As the Pentium M features EIST, we decided to give CrystalCPUID a try. (For full details about CrystalCPUID, please see the above linked article by Jan Kivar.)

CrystalCPUID is indeed functional with this board and CPU. We did not explore its functionality fully, but can report that manual underclocking and undervolting of the CPU is most definitely possible. At 600 MHz and 0.988V, the lowest settings recommended by Intel for this CPU, the AC system idle power dropped to 28W, down about 6W from the standard 1.6 GHz and 1.34V setting.


CrystalCPUID appears to function fine on the EY855-II.

Admittedly, with the CPU already running so cool, and with the system already drawing so little power, there may be little incentive to utilize CrystalCPUID. But it's nice to know that even lower power consumption and heat are possible.

CONCLUSIONS

AOpen has taken a leadership role in desktop Pentium M systems. This is their second Pentium M desktop product, and they have already released another Pentium M desktop mATX motherboard, the i915GMm-HFS, based on the new updated Intel chipset for the P-M The desktop P-M market has got to be small, but it is probably a good niche market with users willing to pay a premium for high performance without the heat. noise and power consumption penalties of conventional desktop platforms. The EY855-II is unique; as far as we know, it is the only SFF barebones system build around the Pentium M platform.

Aside from the wide-gap fin heatsink, AOpen has not made any significant revisions to the XC Cube chassis they have been using for some 18 months. The EY855-II runs quietly mostly because the Pentium M processor generates so little heat that the thermally controlled fans of the PSU and the CPU heatsink never have to ramp up, at least in the mild temperature of our test environment with the modest test components. Even with the hottest P-M model, we doubt this would change. A hot VGA card would probably trigger fan ramp up, however.

We credit AOpen for making a basically quiet chassis, but at the same time, they could have done better. A bit of work on improved acoustics for the PSU and a built-in notebook drive soft mounting feature would have made this a truly quiet product, as the 20 dBA@1m reading without the PSU fan suggests. A natural modification for the experienced modder would be to remove the stock PSU fan, cut away the PSU fan grill, and mount a quieter 80mm fan with soft grommets on the outside of the back panel, with manual control over its speed. This would reduce the PSU fan noise and probably eliminate vibration induced noise from the case.

In stock form, the EY855-II is no quieter than Shuttle's Zen SFF system with a much hotter P4 CPU (at least at idle). At maximum load, this AOpen does best the Zen, but credit must go to the Pentium M's low power dissipation rather than AOpen's design. (If the Shuttle Zen's overall case / cooling design and external PSU was applied to a Pentium M platform, it would definitely be quieter than the EY855-II.) Admittedly, the EY855-II has a couple of big advantages over the Shuttle Zen: An AGP slot and the reduced power consumption and generally better performance of the P-M over much hotter P4s that run at much higher clock speeds. Still, there is the higher CPU price factor, and we don't know how hot a video card the EY855-II will take before its thermally controlled fans will start ramping up.

The AOpen EY855-II SFF is attractive, unique and easy to assemble. It is now the quietest breadbox-style SFF system we've reviewed so far and the only P-M option in a SFF barebones system.

Much thanks to AOpen for the EY855-II sample.

* * *

Discuss this article in the Silent PC Review Forums.



Previous 1 2 3 4 5 6

Complete|Mobile Systems - Article Index
Help support this site, buy the AOpen EY855-II from one of our affiliate retailers!