Hiper Media Center Barebones PC

Complete|Mobile Systems
Viewing page 5 of 8 pages. Previous 1 2 3 4 5 6 7 8 Next


  • The Hiper was tested in various states: Off, Sleep, Idle (with and without Cool 'n' Quiet), video playback, and full load using two instances of CPUBurn K7.
  • System power consumption was measured at the AC outlet using a Seasonic Power Angel.
  • We also examined CPU usage during video playback to see how well the integrated graphics handled different clips. Clips were played with Windows Media Player 11 and CPU usage was measured using the Windows Task Manager for the duration.
  • SpeedFan utility software was used to monitor all the temperature feeds in the system. There are some 7-8 sensors reporting temperatures, including the one in the hard drive.
  • A standard 19" 1280x1024 LCD BenQ monitor as well as a Sony Bravia 40" 720p LCD TV were used for display. The latter's native resolution of 1366x768 was not supported by the graphics driver, but 1280x768 was more than acceptable.
  • The lab's ambient noise level was around 18~20 dBA, and the ambient temperature was 20°C.
Hiper Media Center PC w/ Windows XP: CPU load, AC Power & Noise
CPU Usage
AC Power
Core 0
Core 1
Sleep (S3)
31 dBA
Idle w/ Cool 'n' Quiet
31 dBA
Rush Hour 3
31 dBA
Coral Reef Adventure (1080p-WMV3)
31 dBA
35 dBA
*SPL is the sound pressure level, measured in "A" weighted decibels with a high sensitivity B&K sound level meter at a distance of one meter.

The Rush Hour 3 trailer is a 1080p clip encoded with H.264. It has a good mixture of light and dark scenes, interspersed with fast-motion action and cutaways.
(H.264, 1920x816, 24fps, ~10100kbps)

The Coral Reef Adventure trailer is a 1080p clip encoded in VC-1 using the WMV3 codec (commonly recognized by the moniker, "HD WMV"). It features multiple outdoor landscape and dark underwater scenes.
(VC-1 (WMV3), 1440x1080, 24fps, ~7700kbps)


The video performance was perfectly capable at the 1280x1024 native resolution of the 19" LCD monitor as well as at 1280x768 on the 40" Sony Bravia LCD TV for all imaging programs, video editing, etc. Both Rush Hour 3 and Coral Reef Adventure clips also played fine, although with multiple applications operating, the 1080p clip could sometimes stutter a bit. The built-in WiFi card worked as it should, detecting and allowing easy access at 54 mbps to the wireless network in the lab. The WiFi appeared to add about 2W to the overall power demand when enabled; it is possible to manage it so that it turns off automatically when not needed.

A riser card was unfortunately not supplied with the sample system, which meant the system could not be checked for TV functions with a PCI tuner card. The main question here is not TV functionality, which would be determined mostly by the chosen tuner card, but the card's effect on overall cooling and noise. The card would block the extra hole on the back panel, and occupy enough space in the center of the case that cooling might be affected.

One minor complaint is the exposed location of the reset power button just below the DVD slot: It's too easily accessed by accident.


The minimum noise of the system after Windows stabilized at idle was 31 [email protected], moderately high by SPCR standards. The primary source of noise was the power supply, whose two fans could be easily heard from more than 10 feet away in our quiet test environment. The small size of these fans means that tonal whine was a key aspect of the sonic signature, which made it more annoying than if it had been a broader bandwidth noise, like wind noise. Surprisingly, the tonal peak was centered around 1,000 Hz, not that high a frequency. The CPU cooler fan did speed up a bit under high load, but its overall noise signature was fairly smooth and unobtrusive. The small fan in the "bump" under the drive cage contributed almost no noise at all, as it ran at a constant, very low speed. At extreme CPU load with CPUBurn, increases in fan speeds caused noise to go up to 35 [email protected]

This is not particularly quiet performance, considering that for desktop systems, SPCR's upper limit of "quiet" is defined to be 30 [email protected] Any higher than that, and it's not quiet by our standards.

However, we are not quite as stringent for a media PC, which usually works in a considerably noisier environment: When a media PC being used, we can expect the sound of the movie or TV program, or music to have a significant masking effect on the PC's noise. About the lowest usable volume for such programming is 30 dBA average as perceived by the listener. At normal viewing/listening levels, peaks easily jump into the 50 and 60 [email protected] range. Naturally a PC that emits 30 dBA @1m may not be very intrusive or audible under such conditions. The noise of the PC becomes a more serious issue when the programming is paused, however. These considerations are discussed in detail in the section ACOUSTICS AROUND A MEDIA PC on page two of Cases: Basics & Recommendations.


Seven temperature sensors were detected by SpeedFan, excluding the one in the hard drive. It's difficult to identify which component each sensor is reporting on, but there was no cause for any concern about any of the reported temperatures. At idle, all the temperatures always remained well under 50°C (typically near or under 40°C). At the highest load with two instances of CPUBurn, one sensor reached 54°C after nearly an hour; this was probably the CPU.

One sensor did report 80°C, but this was constant, regardless of load, and this data can safely be discarded as erroneous and insignificant. The system was perfectly well cooled for the installed components. We'd caution against sandwiching the Hiper Media Center system between two other components in a stac k, as the heat could be trapped and cause instability. Take note: We'd make the same caution about any PC.

Previous 1 2 3 4 5 6 7 8 Next

Complete|Mobile Systems - Article Index
Help support this site, buy the Hiper Media Center HMC2K53AA3 from one of our affiliate retailers!