Re: How does the apple2 RGB card work?

["General_Failure" <tristan.mumford@gmail.com>]

Michael J. Mahon said:
>Since the video pulses for lores graphics and DHR are sent at 14MHz,
>each cycle of 3.58MHz covers 4 pulses.  Any "asymetrical" pattern
>will contain a 3.58MHz component (in a Fourier sense).

>So the patterns 1100, 0110, 0011, or 1001 are maximum amplitude
>(saturation) 3.58MHz signals, while the patterns 1000, 0100, 0010,
>0001, 1110, 0111, 1011, and 1101 are half-amplitude chroma signals.
>The patterns 1010, 0101, 1111, and 0000 do not contain any 3.58MHz
>component, and so are colorless.

You're right. It is an elegant exploit of a technology. And as realised
by many, this method does not work in PAL because the standard is
designed to neutralise a lot of abhorrent signals.
The standards of NTSC and PAL aren't really too different. However it
is the little differences that make all the difference. A good
explanation of how the video signal works exists on Rickard's PIC site
on this page I think <a
href="http://www.rickard.gunee.com/projects/video/pic/howto.php";>http://www.rickard.gunee.com/projects/video/pic/howto.php</a>
Velleman sells a kit that seems to be a direct rip or Rickard's PIC
Pong too. I bought one cheaply recently.
Anyway, back to topic.
The translation would either have to be done by a nice whack of logic,
or a massive lookup table, as Bryan seems to have implemented. That
must have taken a very long time.

At any rate, from all this information gained, many different methods
of implementation seem to be possible.