CPU Clock Freq

[jason]

Nikolas Britton wrote:

> Can anyone explain why the clock is off by 17Mhz? This is non critical 
> btw I was just playing with the diff command an wasn't expecting to 
> see this, the system is FreeBSD 5.2.1 running as a guest OS in VMWare 
> (Win2k host).....my guess is its just vmware playing tricks on freebsd...
>
> #diff dmesg.today dmesg.yesterday
> 8c8
> < CPU: Intel(R) Celeron(R) CPU 1.70GHz (1733.85-MHz 686-class CPU)
> ---
> > CPU: Intel(R) Celeron(R) CPU 1.70GHz (1716.78-MHz 686-class CPU)
> 79c79
> < Timecounter "TSC" frequency 1733846104 Hz quality 800
> ---
> > Timecounter "TSC" frequency 1716778304 Hz quality 800
> 85a86,91
> > WARNING: / was not properly dismounted
> > WARNING: /tmp was not properly dismounted
> > /tmp: mount pending error: blocks 4 files 3
> > WARNING: /usr was not properly dismounted
> > WARNING: /var was not properly dismounted
> > cd9660: RockRidge Extension
>
> _______________________________________________
> freebsd-newbies at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-newbies
> To unsubscribe, send any mail to 
> "freebsd-newbies-unsubscribe at freebsd.org"
>
If no else will take this one, because of percent error.  The clock 
generator make a reference clock much lower that you cpu.  The cpu uses 
multipliers of buses that are multiplies of this reference clock.  If 
the quartz crystal is off by 1%, then multiply by 10, 100, or 10,000 you 
can get 17 or more mhz off.  Also the temp of the crystal plays a role 
in the frequency at which it vibrates.  So a cold bootup vs a warm 
reboot will cause variance.  I am going from memory so this might not be 
perfect info.  Opps, I did not see the vmware part.  Well this info 
should still apply.  With a good motherboard monitor program you should 
see the cpu fluxuation a little too.  By the way are you  shutting down 
freebsd properly?

Jason