Viewing page 8 of 10 pages.
Previous 1 2 3 4 5 6 7 8 9 10 Next
MORE TEST RESULTS
|
Comparison: CPU Scaling
|
|
Test State
|
CPU @ 2.5Ghz
|
CPU @ 1.5Ghz
|
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
|
Idle (C&Q)
|
1%
|
2%
|
38W
|
1%
|
2%
|
38W*
|
|
H.264
|
2%
|
8%
|
~53W
|
5%
|
12%
|
~42W
|
|
WMV3
|
28%
|
39%
|
~65W
|
38%
|
52%
|
~48W
|
|
WVC1
|
47%
|
68%
|
~69W
|
70%
|
96%
|
~53W
|
|
Blu Ray
|
23%
|
31%
|
~70W
|
32%
|
44%
|
~56W
|
|
CPUBurn
|
100%
|
101W
|
100%
|
58W
|
|
CPUBurn + ATITool
|
100%
|
109W
|
100%
|
65W
|
|
1GB of system RAM and 128MB of VRAM were used during
testing
* Arrived at by manually underclocking and undervolting. Underclocking
disables C&Q by default.
|
Just how slow of a processor can one use? Gradually underclocking the CPU,
we found that the Blu Ray disc began to stutter at about 1.1Ghz, while audio
glitches were detected in the WVC1 clip at 1.4Ghz. 1.5Ghz was the lowest clock
speed that would smoothly play back all our clips. This was a fantastic result
as the lowest clocked X2 on the market is 2.0Ghz. We also undervolted the CPU
to 1.0V (the same voltage as Cool-N-Quiet) resulting in Blu Ray playback only
requiring 56W from the wall — we wonder how much juice a commercial Blu Ray
player uses.
|
Comparison: Assigned Video Memory
|
|
Test State
|
128MB VRAM
|
256MB VRAM
|
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
|
Idle
|
1%
|
2%
|
38W
|
1%
|
2%
|
38W
|
|
H.264
|
2%
|
8%
|
~53W
|
2%
|
8%
|
~53W
|
|
WMV3
|
28%
|
39%
|
~65W
|
28%
|
41%
|
~66W
|
|
WVC1
|
47%
|
68%
|
~69W
|
43%
|
73%
|
~70W
|
|
Blu Ray
|
23%
|
31%
|
~70W
|
17%
|
28%
|
~70W
|
|
CPUBurn
|
100%
|
101W
|
100%
|
101W
|
|
CPUBurn + ATITool
|
100%
|
109W
|
100%
|
109W
|
|
1GB of system RAM was used during testing
|
AMD recommended that we use the 256MB of video memory, 2GB of system memory,
and to disable Cool-N-Quiet to get optimal Blu Ray playback. Increasing the
amount of video memory garnered some slight improvement during Blu Ray playback,
but it wasn't really necessary. Alos, despite Cool-N-Quiet being enabled, even
when CPU usage was low as during the H.264 clip, we noticed that the CPU throttled
up regardless.
|
Comparison: System Memory
|
|
Test State
|
1GB RAM
|
2GB RAM
|
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
Mean CPU Use
|
Peak CPU Use
|
System Power
|
|
Idle
|
1%
|
2%
|
38W
|
1%
|
2%
|
41W
|
|
H.264
|
2%
|
8%
|
~53W
|
2%
|
8%
|
~54W
|
|
WMV3
|
28%
|
41%
|
~66W
|
24%
|
39%
|
~68W
|
|
WVC1
|
43%
|
73%
|
~70W
|
43%
|
61%
|
~72W
|
|
Blu Ray
|
17%
|
28%
|
~70W
|
17%
|
21%
|
~73W
|
|
CPUBurn
|
100%
|
101W
|
100%
|
103W
|
|
CPUBurn + ATITool
|
100%
|
109W
|
100%
|
115W
|
|
256MB of VRAM was used during testing
|
Increasing the amount of system memory also did very little. We saw some minor
improvement in Blu Ray and WVC1 playback but at the cost of higher overall power
consumption due to the extra stick of DDR2. Even with Vista being an infamous
memory hog, using 1GB of RAM with 256MB assigned to the IGP didn't cause us
any issues.
| Help support this site, buy the Gigabyte GA-MA78GM-S2H motherboard from one of our affiliate retailers! |
|
