> When using TimGetTicks on am omap device 100 ticks will actually be
slightly
> less than one second, to be more exact 100 ticks is 327/327.68 = 0.9979248
> seconds.
>
> 10 minutes equals 600 seconds which on the omap devices is 600 * 327 /
> 327.68 = 598.75488 'real' seconds. This explains your missing second, or
> rather 1.25 seconds. :)

In my last post I did not take into account that the 32 kHz oscillator also
advances one clock period when the hardware register have reached zero and
resets to 327, and with this in mind the 32 kHz clock actually advances 328
clock periods for each OS tick.

Thus,100 ticks on the omap devices is actually 328/327.68 = 1.0009765625
seconds and 600 seconds on omap devices is 600 * 328/327.68 = 600.5859375
real seconds. To sum up, the omap devices loses .59 seconds per 10 minutes,
not gaining 1.25 seconds which the calculations showed in the last post.

I did some measurements on my Z71 and T|E today and they both shows this
exact behaviour, losing .59 seconds per 10 minutes.

Jonatan
www.clievideo.com



-- 
For information on using the Palm Developer Forums, or to unsubscribe, please see 
http://www.palmos.com/dev/support/forums/

Reply via email to