Re: Calculate 1MHz

[kadickey@alumni.princeton.edu (Kent Dickey)]

In article <lIWFb.81404$HH.69258@fe1.texas.rr.com>,
Bryan Parkoff <bryan.nospam.parkoff@nospam.com> wrote:
>    Please let me know if you think that my formula is very accurate.  It
>appears to be the most correct Hz for Apple II+, //c, and //e, but Apple
>IIgs has 2.8MHz that may be very close to my numbers below, but it is not
>really exact.
>
>14,318,180 Hz
>1,022,727.143 Hz    = 14,318,180 Hz / 14 Periods
>894,886.25 Hz         = 14,318,180 Hz / 16 Periods
>1,020,484.320 Hz    = 14,318,180 Hz / (((64 Cycle * 14 Periods) + (1 Cycle *
>16 Periods)) / 65 Cycle)
>
>(((64 Cycle * 14 Periods) + (1 Cycle * 16 Periods)) / 65 Cycle) =
>14.030769230 Periods
>1,020,484.320 Hz   = 14,318,180 Hz / 14.030769230 Periods
>
>    I suspect that Europeans use 50Hz instead of 60Hz for video so they use
>14.25MHz.  It is little slower than standard 14.3Mhz.  How can't 6502 MPU be
>affected?  What about Disk II that the speed may not be the same?
>Curious...
>
>-- 
>Bryan Parkoff
>

The formula from KEGS, which I believe to be accurate, is:

IIgs "1MHz" cycles in one real second =
	(28636360/28)*(65*7/(65*7+1)) = 1020484.320

This matches your result.  This is for a US NTSC Apple IIgs, which is all
KEGS cares about.

The clock stretching of one 7MHz clock every 64 1MHz clocks accounts for
the special factor (which is needed for the Apple II NTSC color trickery).

BTW, average physical crystals give you about 100ppm (some better, some worse)
accuracy, so only the first 4 or 5 significant digits are really meaningful.

Many of your Apple II questions are answered in the KEGS source code...

Kent