Intel D510M0 Motherboard: Atom 2.0
TEST RESULTS
System Power Draw
|
Test Results: System Power (AC)
|
|
Test State
|
HFX Micro S13
(D945GCLF2)
|
Asus EB1501
(ION)
|
D510M0 +
T3500
|
|
Idle
|
25W
|
20W
|
18W
|
|
CPU Load
|
33W
|
27W
|
24W
|
|
CPU + GPU
Load
|
39W
|
39W
|
29W
|
The D510M0 and Morex T3500 combination used a lot less power than the D945GCLF2
residing in the HFX Micro S13,
between 25% and 30% both idle and on load. The T3500's PSU and DC adapter are
very similar to that of the picoPSU used in the HFX system, and even if their
respective efficiencies are off by a couple of percentage points, the difference
in power draw would still be significant. The D510M0 + T3500 configuration also
edges out the ION based EeeBox EB1501, though it is a much smaller machine that
possibly uses mobile hardware.
The top exterior of the case got warm during testing, as did the large heatsink
on the motherboard. However, it wasn't hot enough to be a concern it
past the touch-test with flying colors at least.
Video Playback
|
Test Results: Video Playback
|
|
Test State
|
HFX Micro S13 (D945GCLF2)
|
D510M0 + T3500
|
|
Avg.
CPU
|
System Power
|
Avg. CPU
|
System Power
|
|
Rush Hour
(1080p H.264)
|
65%
|
33W
|
53%
|
23W
|
|
Coral Reef
(WMV-HD)
|
29%
|
31W
|
28%
|
22W
|
|
Undead Battle
(720p x264)
|
39%
|
28W
|
38%
|
22W
|
|
Spaceship
(1080p x264)
|
47%
|
30W
|
42%
|
22W
|
|
Iron Man
(720p Flash)
|
N/A
|
51%
|
22W
|
|
Grey boxes indicate test failure.
|
Without video hardware acceleration, video playback was similar to the Atom
330 and GMA 950 combination found in the D945GCLF2. In any event, most of the
clips in our test suite played fine through CPU power alone, with the exception
of the WMV-HD clip which so far has failed to render properly on any non-ION
Atom system. Smooth x264 playback, even in 1080p, can be achieved using CoreAVC
as a software decoder and 720p Flash video is no problem for the D510. There
was about a 25% improvement in power consumption during video playback compared
to the D945GCLF2.
Performance
|
Test Results: Benchmarks
|
|
Test
|
D945GCLF2
|
D510M0
|
|
NOD32
|
18:26
|
17:42
|
|
WinRAR
|
11:24
|
10:55
|
|
iTunes
|
20:19
|
19:51
|
|
TMPGEnc
|
15:05
|
14:33
|
|
PCMark05
|
2024
|
2100
|
|
3DMark05
|
256
|
359
|
On our timed benchmarks, the new Atom board only managed to improve upon its
predecessor by 133 seconds total for a 3.4% overall improvement. Incidentally,
the D510, thanks to its higher bus speed, runs at 1.66GHz versus the Atom 330's
1.60GHz that's an increase of about 4%. Unfortunately it doesn't look
like the changes in architecture have made performance any better, clock per
clock. There might be a bottleneck somewhere in the design that hasn't been
addressed.
GMA 3150 did push the D510MO's 3DMark05 score 40% higher than the D945GCLF2,
but the difference is only significant in relative, not absolute terms. It doesn't
come close to performing as well as ION, or even Intel's GMA 4500.
| Help support this site, buy from one of our affiliate retailers! | |
» 147342 reads
|