Turion MK-36 has slow clock speed....I think.

[Andy]

I used to do a lot more hardware type stuff with PC's, but I've
started to become more of a network/web guy so my knowledge of
hardware has pretty much gone downhill steadily over the last few
years. However, this does not seem right to me.

I have an Acer Aspire 5100 Laptop with an AMD Turion 64 MK-36 CPU.
Lately it seems to be running slower than before. This usually isn't
cause for alarm, as I install so many programs on this thing and
usually wipe it out every 3 months or so. For kicks I downloaded CPU-
Z and it reports the CPU as running at 800 MHZ Core Speed with a x4
multiplier on a 200 mhz bus. I was a little shocked so I went to
AMD's site and found their clock utility. It also says 800 MHZ. This
CPU is supposedly 2000 MHZ. The memory is DDR2-5300 (266 MHZ).
Shouldn't I be reading 2000 Core clock speed with a 7.5x multipler?
If so, what could be wrong?

Andy

 

[Paul]

I used to do a lot more hardware type stuff with PC's, but I've
started to become more of a network/web guy so my knowledge of
hardware has pretty much gone downhill steadily over the last few
years. However, this does not seem right to me.

I have an Acer Aspire 5100 Laptop with an AMD Turion 64 MK-36 CPU.
Lately it seems to be running slower than before. This usually isn't
cause for alarm, as I install so many programs on this thing and
usually wipe it out every 3 months or so. For kicks I downloaded CPU-
Z and it reports the CPU as running at 800 MHZ Core Speed with a x4
multiplier on a 200 mhz bus. I was a little shocked so I went to
AMD's site and found their clock utility. It also says 800 MHZ. This
CPU is supposedly 2000 MHZ. The memory is DDR2-5300 (266 MHZ).
Shouldn't I be reading 2000 Core clock speed with a 7.5x multipler?
If so, what could be wrong?

Andy

The base clock is 200MHz. The FID and VID vary, according to the
power management state.

On a desktop, the feature is called Cool N' Quiet. When the CPU is
at 100% load (a program using as many cycles as it can), the
multiplier would be set to 10x, giving 2000MHz core. The VID would
rise to the value necessary to support operation at 2000MHz. If the
program was then terminated, and the desktop was idle, power management
would make a correction. The multiplier would drop to the low level
(say 4x), the VID would drop a small amount, and the processor would
use less power.

So the question would be, whether you have selected a Power Management
scheme that forces the clock to stay low, or you just haven't tried
loading it up. Get a copy of Orthos, as it will run two copies of
Prime95 Torture Test. Prime95 itself is another candidate, but
it runs just one copy. These will give the machine some loading.

http://sp2004.fre3.com/beta/beta2.htm (Orthos - Stress Prime 2004)
http://www.mersenne.org/freesoft.htm (Prime95)

While either one of those programs is running its "Torture Test",
again fire up CPUZ. See if it reports 2000MHz or not.

Another program with a graphical display, is RMClock.

http://cpu.rightmark.org/download.shtml
http://cpu.rightmark.org/download/rmclock_225_bin.exe

There is a display on RMClock, that shows the current FID and
VID value. Maybe you'll catch the thing running at x10 with
RMClock. Since my FID is a constant, my graph is a flat line.
The one in this example, seems to be varying a lot

http://cpu.rightmark.org/pic/rmclock2_2.png

Some laptops change Power Management policy, based on whether
the unit is plugged to the wall, or is running on battery.
But the control panel may allow you to change the policy.
For example, in my OS, I have a "Power Options" icon in
my control panel folder, but yours is likely different.
Check out your Power Management policy, before installing
RMClock, because RMClock will register itself as available
to change the Power Management settings for you. I'd want
to check it out, before doing anything else to it.

Paul