Pixel engine
AssigneeIntel Corporation
InventorsRalph Clayton Taylor, Michael Mantor, Vineet Goel, Val Gene Cook, Stuart Krupnik
Life of this patent
Estimating the term…
Loading patent…
AssigneeIntel Corporation
InventorsRalph Clayton Taylor, Michael Mantor, Vineet Goel, Val Gene Cook, Stuart Krupnik
Estimating the term…
There is provided a method for compressing texture values comprising: assigning texture values in a YUV format; packing the texture values into 32-bit words; and color promoting the texture values to 8-bit values. The YUV format has a Y component for every pixel sample, and U/V (they are also named Cr and Cb) components for every fourth sample. Every U/V sample coincides with four (2×2) Y samples. A single 32-bit word contains four packed Y values, one value each for U and V, and optionally four one-bit Alpha components as follows: YUV_0566-5-bits each of four Y values, 6-bits each for U and V; and YUV_1544-5-bits each of four Y values, 4-bits each for U and V, four 1-bit Alphas. The color promotion converts these components from 4-, 5-, or 6-bit values to 8-bit values. This method yields compression from 96 bits down to 32 bits, or 3:1 compression.
A method for determining the rate of change of texture address variables U and V as a function of address variables x and y of a pixel, wherein,
A method for compressing texture values comprising:
A method of performing motion compensation in a computer graphics engine having trilinear filtering hardware and a pallette RAM, comprising:
Dependencies are inferred from the claim text.