>>I know, but I'm not sure how to handle this.
Now, I'm checking the switches at the beginning of overscan. I'm starting
VBLANK at the beginning of the decompression. Then the decompression
follows, which needs about half a frame/puzzle. After this, I'm simply
continueing with my vertical sync routine.
The problem seems so be, that the VSYNC comes to late and/or some VSYNCs are
missing, which I can't generate during decompression.<<
The new version still jumps when the level gets changed. You could probably
set up a TIM1024 timer that is long enough for the worst case in the
decompression routine. The routine that waits for the timer to expire can
then fill up the frame with the nessessary number of blank scanlines. That
would get you a stable screen and a constand blanking period for loading
new levels. Maybe you could set up a shorter timer period for the case
where the user is going through the levels one by one, as this probably
takes less time than decoding a random level.
BTW seem to be using adresses $fff8 and $fff9 in your game. This
prevents the program from being able to run on the supercharger.
Maybe you'd like to change that, if possible.
Ciao, Eckhard Stolberg
-
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
