Viewing page 5 of 5 pages.
Previous 1 2 3 4 5
TESTING
The following tools were used during testing:
Because the system noise was more or less constant — the fans did not change
speed, and drive seeks were not noticeable above the idle noise — most of our
testing consisted of an examination of power and thermal dissipation. The CPU
thermal diode was not calibrated, so thermal measurements should only be evaluated
relative to each other, not in absolute terms.
Ambient temperature during testing was 23°C. Ambient noise level was 16
dBA/1m.
|
Puget Custom Computer
|
|
Activity State
|
CPU Temperature
|
GPU Temperature
|
AC Power Draw
|
|
Idle (EIST)
|
33°C
|
52°C
|
54W
|
|
Idle (no EIST)
|
40°C
|
53°C
|
61W
|
|
CPU Load (2x CPUBurn)
|
56°C
|
53°C
|
82W
|
|
Full Load (ATI Tool + 2x CPUBurn)
|
54°C
|
82°C
|
92W
|
In spite of our worries about airflow around its heatsink, the CPU was well cooled. The maximum temperature of 56°C was low enough
to be perfectly safe, although we would not recommend pushing the system hard in 35°C ambient.
The CPU temperature fell slightly when the graphics card was engaged.
The graphics card also showed no signs of overheating, although the thermal
diode showed a much higher temperature than the CPU. It is widely accepted
(if somewhat unsubstantiated) that nVidia's graphics cards can withstand reported
temperatures of more than 100°C, so the 82°C peak that we saw seems
fine.
One small gripe was that Puget did not enable EIST (Enhanced Intel SpeedStep) by default. This is
as simple as selecting "Minimal Power Management" in the Control Panel,
so there is no reason why it could not be enabled. The power savings from doing
this were not large — only about 15% — but we had hoped that such
a simple step would be a part of Puget's standard build procedure.
At full load, the system drew less than 100W from the wall. That's less than
one fifth of the power supply's capacity (why so large?), and quite an impressive
number for a system equipped with an separate graphics card.
The graphics card did not help the power consumption at
idle. The 54W that the system drew at idle was modest, but it is higher than the <40W that can be achieved using integrated graphics.
SOUND TESTING
|
Puget Custom Computer: Noise Level
|
|
Activity State
|
SPL (dBA/1m)
|
|
Idle / Load
|
20
|
|
Hard Drive Spun Down
|
20
|
The noise measurements bear out our subjective impression of the system: It's
quiet... very quiet. It may have missed the 18 dBA@1m level set by the fanless
system that they sent us before, but it didn't miss by much. It required close
listening to pick the system noise out from the background.
As mentioned, the noise did not change under use. This is hardly a surprise
— the system fan is not thermally controlled, and the
Seasonic power supply did not get louder until DC output reached 200W on our test bench when
we tested a sample for our review. This is more than double the maximum load of the system reviewed here. Even
the noise from the hard drive did not change, since the operating system was
installed on the flash drive and did not produce seek noise.
This provoked a question: Why was the drive allowed to spin at all? It's the
main source of noise, and can easily be disabled using the same window used
to enable SpeedStep. We answered our own question by spinning down the drive
using SeaTools
to see what the system sounded like without it.
To our surprise, the measured noise level did not change, and the noise character
was significantly worse. Without the drive, the bearing noise from the fans
was more plainly audible and could be picked out from the background as a low buzz.
When the drive was turned back on, the noise returned to the hushed whoosh from
before, with no clear tones and no distinct character. Subjectively, it seemed
quieter, even though a new source of noise had been added. Such is the nature of human sonic perception, which is why we always emphasize the importance of listening over measurements alone.
MP3 RECORDINGS
MP3:
Puget Custom Computers Core Duo System: 20 dBA@1m
Comparables:
MP3:
Shuttle SD11G5 with Nexus 92 fan, 920 rpm (any load): 20 dBA@1m
MP3:
Zalman TNN-300 w/Seagate Momentus 7200.1 on Sorbothane: <20 dBA@1m
MP3:
Arctic Cooling Silentium T2 at Idle, 23 dBA@1m
MP3: P180
"Hot Potato" Configuration 4: 25 dBA@1m
MP3:
SilverStone Temjin TJ-07, Configuration 7 (Quiet Nexus Fans): 24 dBA@1m
|
HOW TO LISTEN & COMPARE
These recordings were made with a high resolution studio
quality digital recording system. The microphone was 3" from the
front bezel of the case at a 45° angle, facing the intake side of the
fan to avoid direct wind noise. It is best to download the sound files
to your computer before listening.
To set the volume to a realistic level (similar to
the original), try playing this Nexus
92mm case fan @ 5V (17 dBA/1m) recording and set
the volume so that it is barely audible. Then don't reset the volume and
play the other sound files. Of course, all tone controls and other effects
should be turned off or set to neutral. For full details on how to calibrate
your sound system playback level to get the most valid listening comparison,
please see the yellow text box entitled Listen to the Fans
on page four of the article
SPCR's Test / Sound Lab: A Short Tour. |
CONCLUSIONS
Once again, Puget Custom Computers has delivered an exceptionally quiet system
— this time at a more reasonable price. As shipped, the system is as quiet
as any system we've built on our own, and probably faster. There's not much
here to complain about, but since it is an independent review, we feel compelled to make an effort. The only major omission
that we found was that SpeedStep was disabled by default. Preinstalling and
configuring Nero DriveSpeed would have been nice too, but that costs extra money,
so it's not really a complaint.
Best of all, unlike the system we reviewed previously, you
can actually buy this one — or close enough. Puget offers the complete
system — without the solid state drive — for about US$1800. If you
ask nicely, they can add that as well... if you can afford an extra US$550.
Even if you don't want the exact system reviewed here, the system that Puget
built for us makes a statement: Puget is capable and willing to produce systems
that are as quiet as any custom-built system. If you're willing to work with
them (and the company prides itself on the level of interaction between its
customers and the people who actually build the systems), it is possible to
buy a system that is unique and quiet without getting your hands dirty. During
the buying process, Puget will review the options you've chosen, communicate
suggestions, and accept special requests. If you want a system like the one
in this review, Puget will be happy to make it for you. If you want to tweak
it for your own requirements, they will accommodate those just as happily.
Much thanks to Puget
Custom Computers for building us this custom review sample.
*
SPCR Articles of Related Interest
Puget Custom
Pentium M Rig: A Silent WC System
17" Apple iMac - The
Official SPCR Review
Fanless Ultra Powerhouse
PC by EndPCNoise
Shuttle SD11G5: Pentium
M SFF PC
* * *
Discuss
this article in the Silent PC Review Forums.
| Help support this site, buy from one of our affiliate retailers! |
|
