Reading the paddles isn't really difficult, just the timing has to be really precise (like everything else on the 2600, really). I was looking at a similar problem in Quadracide and wound up doing my paddle reads in the overscan. However, I only needed a dozen positionings or so which let me off the hook to a certain extent. However, there's no reason why you couldn't start the paddle read in the overscan (or at the bottom of the visible screen) and the continue on into the next vblank.
Timer based interrupts would be really nice here, as it is you're probably going to need to find a way to slip polling the paddles into your regular code. Polling the paddle is quite quick, but it must be at regular intervals. The amount of time between polls determines the sensitivity of the paddle
Here is my code from Quadracide:
;Start Overscan
lda #2
STA VBLANK ; Make TIA output invisible, and charge paddles
lda #30
sta TIM64T
lda #23
sta Player0x
sta Player1x
sta Player2y
sta Player3y
; jmp skippaddles
ldy #22
nextpa0:
lda #128
bit INPT0
bne donepa0 ;2/3
dec Player0x ;5
jmp donepa0a ;3
donepa0:
nop
nop
sta $30 ;time waster
donepa0a:
lda #128
bit INPT1
bne donepa1
dec Player1x
jmp donepa1a ;3
donepa1:
nop
nop
sta $30 ;time waster
donepa1a:
lda #128
bit INPT2
bne donepa2
dec Player2y
jmp donepa2a ;3
donepa2:
nop
nop
sta $30 ;time waster
donepa2a:
lda #128
bit INPT3
bne donepa3
dec Player3y
jmp donepa3a ;3
donepa3:
nop
nop
sta $30 ;time waster
donepa3a:
dey
bpl nextpa0
skippaddles:
OverscanWait:
lda INTIM
bne OverscanWait
JMP MainLoop ;Continue forever.
----------------------------------------------------------------------------------------------
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
