Posted By
TLC
on 2010-12-18
15:50:06
|  Re: Vertical splitting
Encyclopedia:I guess I'd leave that to someone else; I don't seem to have all cards in my hands.
Timing: if you inspect the first ten or so rows of my IRQ code, you'd notice that it delays some variable number of cycles, depending on a value which is derived from $ff1e. This part is going to accomodate for any random delays introduced by instructions executed in the moment the IRQ happens. There's no way to interfere with that code from the main program, ...well, unless you disable interrupts sometimes ie. do something nasty. So the answer is: the raster remains stable (ie. cycle exact) as long as you don't disable interrupts in your main program (...which is unlikely, as I guess). You can simply copy the logic of that routine -- only the number to be subtracted, and/or the number of LSRs would need to be changed.
Addendum to yesterday's chitchat: I watched some demos today, and found out that EDC's Imagine already contains pretty tough FLI effects, coded by MC. I didn't check these very closely, but as it appears, they lack the luma-chroma "mixing" effect of early FLI effects, thus, they're 2-line FLIs, or at least 1-line FLIs where only the luma data would be refreshed every lines. Either way, this shows that back in the summer (or early autumn) of 1992, MC did already know more about FLIs than anyone of us did. ...I couldn't track down the exact release date of Imagine, but it must have happened somewhere around September, 1992. ...Too bad, knowing more about that could have spent me writing the FLI Editor stuff the way I did. It's especially a pity, as I remember MC visiting me that summer and showing me some experimental FLI effects, when he came to Dombóvár to meet some of his friends.
|
