Re: IIgs RGB to VGA experiment

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

js@cimmeri.com wrote:
> Ok.... more news.
>
> I managed to get the converter working fairly well with its J01 RGBS
> input by changing the power supply to a lab grade adjustable one, and
> adjusting the voltage very carefully.
>
> The picture is steady, 80char text is quite legible (but not 100%
> perfect), colors are fairly good though with some visible noise
> patterns on some shades, and contrast is almost as good as my IIgs RGB.
>  The results are not perfect, but are the best I've been able to get
> having only tested several composite-> vga converters in the past
> (which I found completely unacceptable).

Likewise.

> I have yet to compare with the Commodore 1084 monitor used in direct
> RGB mode, so will do that soon.

That should be essentially identical to your AppleColor RGB, except
for minor differences in phosphors and video drive levels.

> At 5V, the board did not put out a good picture on my Multisync...
> somewhat shaky as noted in my previous email.   But by lowerering the
> voltage to somewhere around 4.1v, there's a "magic spot" where suddenly
> the board gets very stable.

Now *that's* wierd!  The designer of the board would no doubt be
amazed to hear this.  ;-)

More seriously, I suspect that changing the voltage is causing some
variation in sampling oscillator frequency, and allowing it to lock
to some submultiple of the IIgs dot clock.  It would be much easier
to find this oscillator and just "tune" it, rather than changing the
operating voltage of the entire board--which, at 4.1v, has to be
making some parts quite marginal in operation.

> Since I replicated the shimmy with the lab adjustable supply, I think
> my first power supply (regulated linear 5v supply) is fine, but just
> not putting out the magic voltage number.   I suppose a more permanent
> install could tap 5V off the IIgs from somewhere (anyone know where a
> good spot is?), and since the wattage is fairly low (the board draws
> ..31A at ~4.1V, one could just rig up a variable resistor between the
> tap and board.

No, you'd need a voltage regulator--but that's not a problem.

As mentioned above, changing the operating voltage of the whole
board is not the best way to adjust sampling frequency.  ;-)

> Now, the problem still exists that the converter is going from
> 15.734khz -> 30.9khz... the Multisync reports 31khz x 60hz.   I think
> we would much rather have 31.5khz so as to be compatible with ALL VGA
> monitors -- not the select few that can sync down to 31khz.

Perhaps, but 2:1 is a "magic" ratio, and changing it slightly can
easily result in very visible artifacts.  (There is no good technical
reason why a monitor that works at 32kHz won't work at 31kHz except
that the designer didn't expect it.)

> My next test will be to try using the J02 input -- which is the RGBHV
> one -- because of it's advertised much greater flexibility (should do
> 15.75khz->31.5khz).  First, I will see if it will sync up only using
> the H line with the IIgs' composite sync... as many monitors do.  If
> that doesn't work, I'll have to build a gizmo using the LM3881N chip to
> give me the pseudo HV signals to feed the input.  I think this small
> chip can easily be piggybacked or flown over the converter board
> somewhere.

I'm a little amazed that this J02 input is anything more than a
slightly rewired version of the J01 input.  In other words, I would
expect both inputs to share exactly the same video processing, and
that the "difference" you see in the specs is just a typo.

> Thank you all -- Michael, David, Roger -- for your very supportive,
> ongoing comments.

Still hoping that you get a stable, high-resolution result!

Any chance you could post some pictures for comparison?

-michael

Music synthesis for 8-bit Apple II's!
Home page:  http://members.aol.com/MJMahon/

"The wastebasket is our most important design
tool--and it is seriously underused."