Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!philmtl!philabs!ttidca!woodside
From: woodside@ttidca.TTI.COM (George Woodside)
Newsgroups: comp.sys.atari.st
Subject: Re: Thanks for help.
Message-ID: <17228@ttidca.TTI.COM>
Date: 7 Jun 90 14:16:48 GMT
References: <9006062131.AA12027@kaukau.comp.vuw.ac.nz>
Organization: Citicorp/TTI, Santa Monica
Lines: 26

In article <9006062131.AA12027@kaukau.comp.vuw.ac.nz> ST131BAP@ST1.VUW.AC.NZ writes:
...[edited]...
>.... One minor niggle though. The information
>I have recieved states that Spectrum stores three palettes per scan line.
>How are these palettes used? I can understand the idea of Display
>List Interrupts changing the palette data at the end of each scan line
>but dividing the scan lines in three produces an uneven no. of pixels
>for each palette set. Can anyone help me on this?

Spectrum displays use tight, and precisely synchronized timing loops to change
the color registers on the fly as the picutre is displayed. The first set of 
16 colors is loaded as the raster is in the overscan area, and is
set prior to the first pixel on the line. Then, as the raster traverses
the scan line, the registers are continuously being loaded. One register
changes each 10 pixels, as I recall. Any software attempting to deal with 
a spectrum file must, therefore, keep itself in synch with the current
values of the registers as it works through the scan line, and trrough
the picture.

The top line is always blank, and used to synch up the registers with
the timing loop. The remaining 199 lines get the sets of 64 pallette
entries each.
-- 
* George R. Woodside - Citicorp/TTI - Santa Monica, CA *
* Path:       woodside@ttidca                          *
*   or:       ..!{philabs|csun|psivax}!ttidca!woodside *