Re: Apple ][ to VGA monitor adapter board - looking for best color decoding method

[MdntTrain <js@cimmeri.com>]

On Apr 14, 12:45 pm, ferd...@gmx.de wrote:

> This is a nice idea, but it won't work in the way my card works.
> The 1-bit video stream of the ][ / //e is double-scanned, and the RGB
> conversion is done afterwards. This saves an ADC, some sort of clock
> regeneration, and lowers the data rate, so that normal RAM can be
> used. It also simplifies the scanning counters, because they only have
> to be 7 bits wide (912 pixels per line packed in 8-bit groups=114
> bytes).
> RAM size is not of any concern anymore.

I'm not at a high enough tech level to understand what you just wrote,
but I very much appreciate your attempt. :-)   I hope you've the time
for a detailed "circuit description" in your VGA manual, as that would
be a great vintage element in itself to have as well, as well as being
a "good novel" for some of us nuts.    Manuals these days -- if you
even get one -- are nearly all solid turd (I say "nearly" as I pray
there are exceptions -- though I've not yet seen one).

> Compatibility with the //gs is another problem, because the GS uses 2
> slightly different dot clocks. If the scan doubler works perfectly in
> one mode, it won't in the other.

So what about just having two clocks, with a switch to toggle them?
Given the gains, I certainly wouldn't mind manually toggling modes as
long as the switch was convenient.

We (my friend Roy and I) encountered this problem in trying to adapt
an Averlogix chip for doing low-cost IIgs RGB->VGA conversion.  We
actually inserted it prior to the IIgs video D/A converters.. so that
it would be D-D-A.  It worked great for the 7/14mhz modes, but we ran
out of smarts on how to get it to synch to the 8/16mhz modes, and
started to think about the filter method.  For best results, it seemed
to require a rather complex active filter that we didn't have the
experience and time to develop.

But remember, the IIgs RGB monitor itself provided the necessary
bandwidth rolloff ... -3db down at 8mhz.. and CRT-limited resolution
for dithering and such.  A super high bandwidth and resolution RGB
monitor won't produce the right IIgs visual effect -- aliasing issues
aside, it would have very discernable (and annoying) alternating bars
of color.   So either way, my belief is that analog bandwidth limiting/
filtering is necessary anyway to produce the correct visual effect.

> But using the RGB decoder without the scan doubler is possible, it
> just needs to be modified to run on half the clock (need to change two
> inductors, three caps and maybe two resistors). So it might be an
> option to make an YUV/RGB card w/o scan doubling.
> I just wouldn't know any uses for this card: there are little 15kHz
> analog RGB monitors available, and if, they usually also have a
> composite video input.

That's true.. all my RGB monitors have Composite in.  However, YUV
might be interesting.. especially to try with YUV-Digital DVI/HDMI
converters.


JS