At 21:14 20.06.2000 -0700, you wrote:
>How many scanlines can you spread out the reads in order to still achieve
>accurate results?
Depends on how you define acurate. A simple byte counter can have 256
differnt states, if you need to, so theoretically you would have to
read the paddles in every scanline during the entire frame to reach
this acuracy. If you only want to figure out if the paddle is in
left, right or neutral position, than it depends on big you want the
neutral space to be. I think 30 scanlines is a good value for this
case.
I also think that the paddle range was chosen, so that you can still
get enough values for a larger screen display like on PAL.
>There is another line in the documentation that reads something like "these
>registers are available to be written to at any time" although it does warn
>about doing successive HMOVEs too close to eachother.
That part of the documantation refers to triggerng the HMOVE register.
If you do that you start the fine movement process for all the objects.
But since that takes quite a lot of cycles to finish, the manual warns
you to change the fine movement registers to soon after a HMOVE, or
you might get more or less unpredictable results.
>With the latching, most 2600 games won't allow autofire over 30 times a
>second then, right? One on and one off state on alternating frames?
>
>What happens when you hold down the fire button close to the end of a frame
>and in the middle of the next frame you release it and then manage to press
>it down again by the end of the frame? What state will the register be in
>when it's read in vertical blank?
AFAIK the latches stay the same after a button was pressed, until you
manually reset them. If your program restes the latches between frames,
they will indicate pressed buttons for both frames in your example above.
Ciao, Eckhard Stolberg
--
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
