> Based on what I read in Video Demystified (excellently informative book,
> thanks to whoever it was that recommended it here), isn't the trick to get
> interlace to begin the second field (60 Hz) of each frame (30 Hz) in the
> _middle_ of a scanline? This means that all the scanlines of field 2 will
> be offset vertically by half a scanline's worth of time from the scanlines
> of field 1, generating the interlace.
Good point. I don't think I tried that.
>
> In 2600 terms, you would enable VSYNC in the center (cycle 38) of scanline
> #263 (counting from 1) after you enabled it previously. Then you'd turn
> off VSYNC in the middle of scanline #266, and turn it on again at the
> beginning of scanline 526 (scanline 1 of the next 30Hz frame.) RSYNC
> isn't involved - there's no interruption or resetting to the horizontal
> timing to get a vertical offset for interlace.
I think this would be a good thing to try. I would however add something
to that. If you don't add RSYNC, you will be screwed for trying to draw
anything during your kernal because you won't be able to use WSYNC. Who
here wants to do a kernal without WSYNC? I say this because if you used
WSYNC, it would release in the middle of the line instead of at the end.
To solve this, I would add the RSYNC just before/after the VSYNC that you
place in the middle of the line. This way the WSYNC would end at the same
time as the screen is(you've syncronized the TIA horizontal counter and
the tv's scan). This would actually make the scheme easier to code
because you would simply do this 1/2 line break each time(because each
time you do it you move the vcs's syncronization over by 1/2).
> Anybody tried this? My 2600 is far from set up at the moment. And this
> is assuming NTSC, although I'm fairly sure the principle is the same for
> PAL except for the scanline numbers.
Thanks for the suggestion, I may do this later on.
Mark
--
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
