Re: 264 Series and their chips

On 2011-08-14 21:14, Gerrit Heitsch wrote:

> Do you have a link giving some background on how that is done? I have
> seen full screen images on the C16, but so far was unable to find
> documentation that explains how to do this. Since TED doesn't have
> sprites you can't do things as you'd do with a VIC when you want to put
> graphics in the border area.

Border removal doesn't need making use of the NTSC-bit ie. the above
trick of enhancing horizontal resolution.

Either way, I don't know where they're documented.

In short, removing the vertical border is done by $ff1d tricks (setting
the raster register back and forth at certain positions, so that screen
redraw would be started earlier / stopped later than usual (taking care
of the number of lines per frame that should be kept unaffected), with
the additional work of switching screen memory at some position where
the screen memory page would normally run "out" ie. wrap around). On the
VIC, the possibility of using that trick is prevented by the absence of
a writeable rasterline counter. Here it's fully possible to "enlarge"
the screen area (without the chip stopping screen redraw after 25 rows),
so no sprites are needed either.

Removal of the horizontal border is rather tricky. $ff1e tricks are used
line-by-line (in a similar fashion to the removal of the vertical border
- tricking the TED so that checkpoints are triggered sooner, others
later than normal). Line sync (number of cycles per line) needs to be
kept unaffected. The resulting places are slightly hard to fill up with
valid screen data; it's either the idle-byte ($ffff) that is used, or
bitmap mode, which is displayed on the borders (if the effect is done
right), and it also does keep count of the number of displayed
positions. Either case, the ability to set colors for the newly inserted
places is limited.

       Message was sent through the cbm-hackers mailing list