Do you have a question? Post it now! No Registration Necessary. Now with pictures!
- Posted on
- Puzzling performance issue
- Barry Watzman
May 28, 2006, 6:51 pm
rate this thread
#1 2004 Toshiba A45-S121, Celeron M 2.8GHz, 128k Cache, 400MHz FSB, 256
megs single channel DDR memory.
#2 2006 Toshiba A105-S2712, Centrino Pentium M 740 (1.73GHz), 2MB Cache,
533MHz FSB, 1GB of dual channel DDR2 memory
Both systems running Windows XP SP2 Home edition.
I ran a benchmark performance test on both systems, the test was
PassMark Version 4 (it was an optional component of Norton SystemWorks
I expected the A105 to do a lot better, but much to my surprise, the A45
did better on the performance test in almost all CPU and memory tests
(disk I/O on the A105 (which has SATA 5400 rpm drive) was faster, and
the video tests were a lot faster, but on all of the memory and CPU
tests the A45 pretty much killed the A105).
Anyone have any idea why? As far as I can tell, the A105, which has
Speedstep and other power savers, was running at full speed (it was
plugged into the wall for the tests).
Re: Puzzling performance issue
Barry Watzman wrote:
Not being familiar with exactly what the benchmarking tasks are, my
guess would be that the tests are simple instruction executions, in
which case the Celeron running at the higher clock speed will complete
The Celeron is a great processor for casual users, as casual users are
usually not doing anything too intensive and only one major task at a
time. In this scenario, the key variables that would determine
execution speed are clock speed and available resources (e.g. RAM,
video RAM), and not the type of processor. If the user is doing more
intensive things like gaming, video, and developing, then the
performance difference will begin to be demonstrated.