Re: Multinorm TV woes - no NTSC color display

["Michael J. Mahon" <mjmahon@aol.com>]

Linards Ticmanis wrote:
> Michael J. Mahon schrieb:
>
> > I see two possibilities--1) the 14MHz crystal you installed is just
> > enough out of tolerance that the monitor does not recognize the burst
> > as a color burst, or 2) the monitor is applying some strict criterion
> > to the video signal that the Apple does not conform to.  (This is more
> > of a problem with monitors that use digital decoding.)
>
> By the way... what criteria are theoretically available for this detection?
>
> A little brainstorming reveals:
>
> 1.) Number of lines per frame or per field
> 2.) Horizontal frequency
> 3.) Number of color carrier pulses per line
> 4.) Relative phase shift of color burst from line to line
> 5.) Number and arrangement of vertical sync pulses
>
> Most of these would need some tolerance but not too much, so that there
> are neither too many false positives, nor too many false negatives.
>
> Did I forget anything?
>
> I wonder, which of these would be easiest to implement in a TV set and
> to get right most of the time (i.e. the TV would be able to reliably
> tell apart all of NTSC-3.58, NTSC-4.43, PAL, PAL-60 and SECAM), but
> would fail for the Apple?

Sorry to be late responding--I'm travelling again...

All of the characteristics you mention and any other timing
relationship is a potential deal maker/breaker.  Once the
processing moves to the digital domain, it's a small matter
of programming to test for virtually anything you want to
think of as a "salient" characteristic of an NTSC video signal.

-michael