Re: 1084 woes

[Linards Ticmanis <ticmanis@coli.uni-sb.de>]

David Empson wrote:


> In mixed mode, the Apple II _does_ generate colour burst for the text
> lines.

OK... probably true. However it seems that the PAL card that I use doesn't.

It took me a while to find out that this card (made for a clone) expects 
an active-high TEXT instead of the Apple-provided active-low /TEXT on 
pin 23 of slot 7. So in an unmodified state I'd get colored text but B/W 
graphics. Exchanging one NPN transistor (2N3904), that shuts down the 
PAL quartz when active, with a PNP one (2N3906) that happened to be 
lying around fixed the problem - only in mixed mode I'd get the 
complementary colors (purple instead of green etc.) in a stripe covering 
about the second sixth of the screen, counting from the top, preceded 
and followed by a few lines in B/W. Also the text wasn't fringy, as it 
should be, derived from what you said. Only the top most line of text 
had a bit of a pinkish color to it. Totally removing the transistor gave 
color fringed text at all times, but with correct colors in the graphics.

It seems that the /TEXT signal on pin 23 of slot 7 _is_ indeed pulled 
low for the four lines of text in mixed mode. Unfortunately I don't have 
an Oscilloscope, but a quick multimeter check showed a somewhat 
decreased mean voltage on the pin in mixed mode, compared to 
graphics-only mode.

I guess either the quartz or the PAL chip TCA650 simply doesn't like to 
be switched on and off very quickly. I'll try inserting a 220uF 
capacitor between the transistor base and ground - combined with the 
base current limiting 4.7kOhm resistor that should delay the activation 
of the transistor for about 0.1s (if I calculated correctly), quite 
enough to get the quartz across those 4 lines of text alive.

--

Linards Ticmanis

 The Master said, "The business of laying on the colors follows the
preparation of the plain ground."