The story so far with my attemps to program the Sprite Card:
I'm able to program it using Orca/Pascal with various pointer tricks. I
have learned how to use the GS Toolbox to display hex values on screen
for debug purpose. After a three months break, I had time to wire an
RCA plug to it and get a display on an ordinary TV set (no RF
modulator). First disappointment, I only get black and white pictures
but I'm satisfied to see that old set of chips to actually work and
respond to wathever data I write in registers or in the VRAM.
I don't know why I don't get any colors. Perhaps I have an impedance
mismatch between TV input and card output or a defect in the Card's
video signal buffer amplifier (a pair of transistors - basically). Any
suggesstions about lack of colors?
Meanwhile, I'll write a set of VDP init and VRAM access routines to make
sprite/backdrop creation a little easier.
Yves
Michael J. Mahon worte
Yves McDonald wrote:
I think the paper about the sprite card was published in Byte issue
on LOGO language (circa 1982).
There are code samples in TI 9900 assembly language in TI books
indeed. From the chip count on the sprite card, the 9918A
interfaces nicely to the Apple bus.
Yes, it interfaces very easily, though the actual decoding on this card
is somewhat peculiar.
The card also contains a 2516 EPROM which contains 6502 firmware
used by Sprite Logo to control the card. I glanced at it, but have not
done much disassembly.
The TMS9918A was also used in the MSX computers, and that is a
good place to go for more information on the chip:
http://www.msxnet.org/tech/
After seeing how reticulated the 9918 is to control, I admit that my
interest flagged somewhat. But that can be remedied by writing a
set of low-level drivers to provide a more "natural" programming
interface.
I'd be very interested in any software that used this card.