Original Publication Date: 1994-Feb-01
Included in the Prior Art Database: 2005-Mar-26
Disclosed is a method that allows for multiple color maps and multiple pixel types to coexist on the screen at the same time. The method also enables the expansion of 8- and 16-bit RGB pixels into 24-bit RGB pixels in a programmable manner.
a method that allows for multiple color maps and
multiple pixel types to coexist on the screen at the same time. The
method also enables the expansion of 8- and 16-bit RGB pixels into
24-bit RGB pixels in a programmable manner.
Given that a
graphics system allows for the support of multiple
pixel types, an addressing method is required that will efficiently
utilize the look-up table where the different pixel types are
expanded from 4, 8, or 16 bits into 8 bits of RED, 8 bits of GREEN,
and 8 bits of BLUE for digital-to-analog conversion and display on a
monitor. The 16-bit pixels that represent RGB values present an
additional problem, since there are 65000+ possible bit combinations.
system must have a way of identifying the various
pixel types. Given that, an addressing scheme can be constructed
that will allow the various pixel types to be displayed correctly.
Multiple pixel types can be displayed simultaneously using this
method. Three different look-up tables for 8-bit color indexed
pixels, 8-bit RGB pixels, and 16-bit RGB pixels can be supported
simultaneously with four 256 entry color look-up tables. The 8-bit
RGB pixels are assumed to be coded as 8 bit RGB pixels are assumed to
be coded as 3 bits of RED, 3 bits of GREEN, and 2 bits of BLUE. The
16-bit RGB pixels are assumed to be coded as 5 bits of RED, 6 bits of
Green, and 5 bits of Blue. The addressing scheme also assumes that
the RED, GREEN, and BLUE look-up tables can be addressed
tables are addressed in the following manner.
Two select bits are generated by the graphics system that can be used
as the high-order address bits for the look-up table. The pixel
value is concatenated with the select bits to generate the 10-bit
address into the look-up table.
LUT ADDRESS BITS 9 8 7 6 5 4 3 2 1 0
sel1 sel0 p7 p6 p5 p4 p3 p2 p1 p0
| PIXEL VALUE
|------ PIXEL TYPE INFORMATION
8-bit RGB in 332 mode and 16-bit RGB in 565 mode can
be done in a way that allows for programmability of the resulting RGB
values. Done in the usual fashion, the table for 8-bit RGB pixels
would require 256 entries, and one for 16-bit RGB would require over
65000 entries. The RGB information could also be hardwired to the
DAC (Digital-to-Analog Converter) inputs, but expanding the
information to 8 bits becomes a problem. Is the expansion done by
padding with 1s, 0s, or repeating the value of the pixel? ...