Coordinate computations for non-power of 2 texture maps

• US 20080273043A1
• Filed: 05/01/2007
• Published: 11/06/2008
• Est. Priority Date: 05/01/2007
• Status: Active Grant

• Current Assignee: Vivante Corporation (Verisilicon Microelectronics (Shanghai) Co. Ltd.)
• Original Assignee: Giquila Corp.
• Inventors: Allegrucci, Jean-Didier, Tai, Anthony Ya-Nai
• Granted Patent: US 8,207,980 B2
• Time in Patent Office: Days
• US Class Current: 345/582
• CPC Class Codes: G06T 15/04 Texture mapping

Abstract

A graphic processing system to compute a texture coordinate. An embodiment of the graphic processing system includes a memory device, a texture coordinate generator, and a display device. The memory device is configured to store a plurality of texture maps. The texture coordinate generator is coupled to the memory device. The texture coordinate generator is configured to compute a final texture coordinate using an arithmetic operation exclusive of a division operation. The display device is coupled to the texture coordinate generator. The display device is configured to display a representation of one of the plurality of texture maps according to the final texture coordinate. Embodiments of the graphic processing system facilitate a simple hardware implementation using operations other than division.

Claims

1. A graphic processing system to compute a texture coordinate, the graphic processing system comprising:

a memory device to store a plurality of texture maps;

a texture coordinate generator coupled to the memory device, the texture coordinate generator to compute a final texture coordinate using an arithmetic operation exclusive of a division operation; and

a display device coupled to the texture coordinate generator, the display device to display a representation of one of the plurality of texture maps according to the final texture coordinate.

2. The graphic processing system of claim 1, the texture coordinate generator comprising identification logic, the identification logic to identify an integer component and a fraction component of a floating point texture coordinate.

3. The graphic processing system of claim 2, the identification logic further configured to identify the integer component of the floating point texture coordinate, according to the following:

S_i=float(int(S)),

where S designates the floating point texture coordinate, and S_i designates the integer component of the floating point texture coordinate.

4. The graphic processing system of claim 3, the texture coordinate generator further comprising intermediate computation logic, the intermediate computation logic to convert the integer component of the floating point texture coordinate to a fixed point integer component, according to the following:

I=int(S_i),

where I designates the fixed point integer component.

5. The graphic processing system of claim 2, the identification logic further configured to identify the fraction component of the floating point texture coordinate, according to the following:

S_f=float(frac(S)),

where S designates the floating point texture coordinate, and S_f designates the fraction component of the floating point texture coordinate.

6. The graphic processing system of claim 5, the texture coordinate generator further comprising intermediate computation logic, the intermediate computation logic to compute an intermediate texture coordinate, according to the following:

U′

=int(S_f×

width),

where U′

designates the intermediate texture coordinate, and width designates a dimension of one of the plurality of texture maps.

7. The graphic processing system of claim 6, the texture coordinate generator further comprising final computation logic, the final computation logic to compute the final texture coordinate equal to the intermediate texture coordinate for a texture wrapping operation, according to the following:

U=U′

,

where U designates the final texture coordinate.

8. The graphic processing system of claim 6, the texture coordinate generator further comprising LSB logic, the LSB logic to identify a value of a least significant bit of a fixed point integer component, the fixed point integer component based on the integer component of the floating point texture coordinate.

9. The graphic processing system of claim 8, the texture coordinate generator further comprising final computation logic, the final computation logic to compute the final texture coordinate for a texture mirroring operation, according to the following:

U=U′

for LSB=0, and

U=width−

U′

for LSB=1,

where U designates the final texture coordinate, and LSB designates the value of the least significant bit of the fixed point integer component.

10. The graphic processing system of claim 1, the division operation comprising a quotient division operation or a remainder division operation.

11. A method for computing a texture coordinate, the method comprising:

identifying an integer component of a floating point texture coordinate;

identifying a fraction component of the floating point texture coordinate; and

computing an intermediate texture coordinate based on the fraction component of the floating point texture coordinate using an arithmetic operation other than a division operation.

12. The method of claim 11, further comprising identifying the integer component and the fraction component of the floating point texture coordinate, according to the following:

S_i=float(int(S)), and

S_f=float(frac(S)),

where S designates the floating point texture coordinate, S_i designates the integer component of the floating point texture coordinate, and S_f designates the fraction component of the floating point texture coordinate.

13. The method of claim 11, the division operation comprising a quotient division operation.

14. The method of claim 11, the division operation comprising a remainder division operation.

15. The method of claim 11, further comprising computing the intermediate texture coordinate, according to the following:

U′

=int(S_f×

width),

where U′

designates the intermediate texture coordinate, S_f designates the fraction component of the floating point texture coordinate, and width designates a dimension of a texture map.

16. The method of claim 11, further comprising computing a final texture coordinate based on the intermediate texture coordinate.

17. The method of claim 16, further comprising computing the final texture coordinate equal to the intermediate texture coordinate for a texture wrapping operation, according to the following:

U=U′

,

where U designates the final texture coordinate, and U′

designates the intermediate texture coordinate.

18. The method of claim 16, further comprising converting the integer component of the floating point texture coordinate to the fixed point integer component, according to the following:

I=int(S_i),

where I designates the fixed point integer component, and S_i designates the integer component of the floating point texture coordinate.

19. The method of claim 18, computing the final texture coordinate in response to a determination of a value of a least significant bit of the fixed point integer component for a texture mirroring operation, according to the following:

U=U′

for LSB=0, and

U=width−

U′

for LSB=1,

where U designates the final texture coordinate, U′

designates the intermediate texture coordinate, width designates a dimension of a texture map, and LSB designates the value of the least significant bit of the fixed point integer component.

20. A computer readable storage medium embodying a program of machine-readable instructions, executable by a digital processor, to perform operations to facilitate computation of a texture coordinate, the operations comprising:

identify an integer component of a floating point texture coordinate;

identify a fraction component of the floating point texture coordinate; and

compute an intermediate texture coordinate based on the fixed point fraction of the floating point texture coordinate component using an arithmetic operation other than a division operation.

21. The computer readable storage medium of claim 20, the operations further comprising an operation to compute a final texture coordinate equal to the intermediate texture coordinate for a texture wrapping operation, according to the following:

U=U′

,

where U designates the final texture coordinate, and U′

designates the intermediate texture coordinate.

22. The computer readable storage medium of claim 20, the operations further comprising:

convert the integer component of the floating point texture coordinate to a fixed point integer component, according to the following;

I=int(S_i),

where I designates the fixed point integer component, and S_i designates the integer component of the floating point texture coordinate;

identify a value of a least significant bit of the fixed point integer component; and

compute a final texture coordinate for a texture mirroring operation, according to the following;

U=U′

for LSB=0, and

U=width−

U′

for LSB=1,

where U designates the final texture coordinate, and LSB designates the value of the least significant bit of the fixed point integer component.

23. An apparatus for computing a texture coordinate, the apparatus comprising:

means for identifying an integer component of a floating point texture coordinate;

means for identifying a fraction component of the floating point texture coordinate; and

means for computing an intermediate texture coordinate based on the fraction component of the floating point texture coordinate using an arithmetic operation other than a division operation.

24. The apparatus of claim 23, further comprising means for computing a final texture coordinate equal to the intermediate texture coordinate for a texture wrapping operation, according to the following:

U=U′

,

where U designates the final texture coordinate, and U′

designates the intermediate texture coordinate.

25. The apparatus of claim 23, further comprising:

means for converting the integer component of the floating point texture coordinate to a fixed point integer component, according to the following;

I=int(S_i),

where I designates the fixed point integer component, and S_i designates the integer component of the floating point texture coordinate;

means for identifying a value of a least significant bit of the fixed point integer component; and

means for computing a final texture coordinate for a texture mirroring operation, according to the following;

U=U′

for LSB=0, and

U=width−

U′

for LSB=1,

where U designates the final texture coordinate, and LSB designates the value of the least significant bit of the fixed point integer component.