At 14:19 19.03.2000 -0800, you wrote:
>It acts like I'm doing too many cycles during the
>Horizontal Fine Coarse (done during vblank)
>positioning and "bleeding" over to next scanline...if
>that makes sense.
If you hit RESxx during the horizontal blank, the object will be
positioned as if RESxx was hit at cycle 22. If you use a zero in
your delay loop, RESP0 will be hit at cycle 21. So there will be
a differnce of three pixels when the delay table values change
from 0 to 1, which you will have to make up for with the fine coarse
positioning.
Another problem with your demo is that the code, that determines if
the sprite should be drawn or not, takes too many cycles. If you
want to display a sprite from pixel 5, you have to change the
sprite's graphics and colour register within 24 cycles after the
WSYNC. In your demo the graphics register gets changed at cycle
29 and the colour register gets changed at cycle 36. So during
the leftmost part of the screen the sprite will display the graphics
and/or colour from the scanline before.
BTW you are doing the sprite positioning not only during the VBLANK
but also during the VSYNC. Doing anything during the three lines of
VSYNC might result in your game not syncronizing properly on badly
tuned TV sets.
And finally, if you want us to find a problem in your code, it would
be helpfull, if you could send the source code along with the binary,
so that we don't have to disassemble it first. ;-)
Ciao, Eckhard Stolberg
--
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
