Method and apparatus for anti-aliasing using floating point subpixel color values and compression of same

• US 20060188163A1
• Filed: 01/31/2006
• Published: 08/24/2006
• Est. Priority Date: 02/13/2003
• Status: Active Grant

• Current Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Original Assignee: ATI Technologies Incorporated (Advanced Micro Devices, Inc.)
• Inventors: Elder, Gordon M.
• Granted Patent: US 7,764,833 B2
• Time in Patent Office: Days
• US Class Current: 382/232
• CPC Class Codes: G06T 11/40 Filling a planar surface by adding surface attributes, e.g. colour or texture
• CPC Class Codes: G06V 10/20 Image preprocessing
• CPC Class Codes: H04N 1/64 Systems for the transmission or the storage of the colour picture signal; Details therefor, e.g. coding or decoding means therefor H04N19/00 takes precedence
• CPC Class Codes: H04N 1/644 using a reduced set of representative colours, e.g. each representing a particular range in a colour space
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component
• CPC Class Codes: H04N 19/423 characterised by memory arrangements H04N19/433 takes precedence

Abstract

The present invention provides a method and apparatus for data compression that includes representing each sub-pixel of each pixel with a pointer corresponding to an attribute of the sub-pixel, the attribute being a floating point binary number. An overall attribute of each pixel is then determined. The determining of the overall attribute of each tile may include any one of assigning the attribute of the sub-pixels to the overall attribute of the pixel when the sub-pixels are represented by an identical pointer, and resolving the overall attribute of the tile by the attributes of the sub-pixels when the sub-pixels are represented by non-identical pointers of the pixel.

Claims

1. A method of data compression comprising:

representing each sub-pixel of each pixel of a display with a pointer corresponding to an attribute of the sub-pixel, the attribute being a floating point binary number; and

determining an overall attribute of each pixel by any one of;

assigning the attribute of the sub-pixels of the pixel to the attribute of the pixel when the sub-pixels of the pixel are represented by an identical pointer; and

resolving the attribute of the pixel by the attributes of the sub-pixels of the pixel when the sub-pixels of the pixel are represented by non-identical pointers.

2. The method of claim 1, wherein each attribute represents color information and further comprising storing the overall attributes of each pixel as pointer information in memory.

3. The method of claim 1, wherein the pointer represents at least two different floating point binary colors.

4. The method of claim 1, wherein the resolving comprises:

converting the attribute corresponding to each non-identical pointer to a binary fixed point attribute;

combining the binary fixed point attributes to determine a combined fixed point attribute;

converting the combined fixed point attribute to a combined binary floating point attribute; and

representing the overall attribute of the pixel by the combined binary floating point attribute.

5. The method of claim 4, wherein the combining comprises averaging the binary fixed point attributes based on the number of non-identical pointers of the pixel.

6. An apparatus comprising:

a logic operative to;

represent each sub-pixel of each pixel of a plurality of pixels of a display with a pointer corresponding to an attribute of the sub-pixel, the attribute being a floating point binary number; and

determine an overall attribute of each pixel by any one of;

assigning the attribute of the sub-pixels to the overall attribute of the pixel when the sub-pixels are represented by an identical pointer; and

resolving the overall attribute of the pixel by the attributes of the sub-pixels when the sub-pixels are represented by non-identical pointers of the pixel.

7. The apparatus of claim 6, wherein each attribute represents color information and further comprising the logic operative to store the overall attributes of each pixel as pointer information in memory.

8. The apparatus of claim 6, wherein the pointer represents at least two different floating point binary colors.

9. The apparatus of claim 6, wherein the resolving comprises the logic operative to:

convert the attribute corresponding to each non-identical pointer to a binary fixed point attribute;

combine the binary fixed point attributes to determine a combined fixed point attribute;

convert the combined fixed point attribute to a combined binary floating point attribute; and

represent the overall attribute of the pixel by the combined binary floating point attribute.

10. The apparatus of claim 9, wherein to combine comprises averaging the binary fixed point attributes based on the number of non-identical pointers of the pixel.

11. An apparatus comprising logic operative to represent each sub-pixel of each pixel of a display with a pointer corresponding to an attribute of the sub-pixel, the attribute being a floating point binary number.

12. The apparatus of claim 11, wherein each attribute represents color information and further comprising storing the overall attributes of each pixel as pointer information in memory.

13. The apparatus of claim 11, wherein the pointer represents at least two different floating point binary colors.

14. An apparatus comprising logic operative to:

receive pointers corresponding to sub-pixels of each pixels of a display, each pointer representing an attribute of the corresponding sub-pixel;

assign the attribute of the sub-pixels to an overall attribute of the pixel when the sub-pixels are represented by an identical pointer;

resolve the overall attribute of the pixel by the attributes of the sub-pixels when the sub-pixels are represented by non-identical pointers of the pixel.

15. The apparatus of claim 14, wherein to resolve the overall attribute of the pixel comprises:

converting the attribute corresponding to each non-identical pointer to a binary fixed point attribute;

combining the binary fixed point attributes to determine a combined fixed point attribute;

converting the combined fixed point attribute to a combined binary floating point attribute; and

representing the overall attribute of the pixel by the combined binary floating point attribute.

16. The apparatus of claim 15, wherein the combining comprises averaging the binary fixed point attributes based on the number of non-identical pointers of the pixel.

17. A method of data compression comprising:

storing sub-pixels values for a plurality of pixels; and

selecting from one of at least the following compression levels;

a clear state that utilizes a single floating point color value per tile;

an uncompressed state that utilizes a floating point color value stored for each sub-pixel of each pixel in a tile;

a partial compression state that utilize a pointer for each subpixel in a tile that selects a floating point color value for each subpixel; and

full compression that utilizes a single floating point color value for each pixel, to compress the plurality of sub-pixel values.

18. A method for anti-aliasing comprising:

storing floating point sub-pixels color values for a plurality of pixels; and

using at least some of the stored floating point sub-pixel color values to determine an anti-aliased color value for a pixel.