Depth buffer compression
Depth buffer compression
 CN 101,802,872 A
 Filed: 09/12/2008
 Published: 08/11/2010
 Est. Priority Date: 09/12/2007
 Status: Active Grant
First Claim
1. One kind be used for selecting will be in the method for the compact model of depth buffer (20) application, described method comprises:
 Acquisition is corresponding to the value of the sheet (30) of described depth buffer (20);
Select first compact model (34) from more than first compact model;
Compress the value of described (30), based on described first pattern (34) value of being compressed is write first, and use the main bit (34) of predetermined number store with described first pattern (34) and described first compression sizes in the relevant information of at least one item;
Select the set of less important compact model (32) from more than second compact model;
Determine to form a plurality of subsheet (28) of described (30), compress the value of described subsheet, described set based on less important compact model (32) writes second with the value of being compressed, and use the main bit (34) of identical predetermined number store with the described set of less important compact model (32) and described second compression sizes at least one relevant information;
The compression sizes of the compression sizes of more described (30) and described subsheet (28);
Choose the described set and the corresponding sheet size of described first pattern (34) or less important pattern (32) based on the result of described comparison;
AndThe sheet of selected compression or subsheet stored in the described depth buffer (20) and the information relevant with the set of selected first pattern (34) or less important compact model (32) or corresponding size write into the main bit (34) of described predetermined number.
Chinese PRB Reexamination
Abstract
Device, computer readable medium, and method for selecting compression modes to be applied in a depth buffer (20). The method includes obtaining values corresponding to a tile (30) of the depth buffer (20), selecting a first compression mode (34) from a first plurality of compression modes, compressing the values of the tile (30), writing the compressed values into a first block, based on the first mode (34), and using a preset number of primary bits (34) to store information related to at least one of the first mode (34) and a compression size of the first block, selecting a set of secondary compression modes (32) from a second plurality of compression modes, determining a number of subtiles (28), that make up the tile (30), compressing values of the subtiles, writing the compressed values into a second block, based on the set of secondary modes (32), and using the same preset number of primary bits (34) to store information related to at least one of the set of secondary modes (32) and a compression size of the second block, comparing the compression size of the tile (30) to the compression size of the subtiles (28), choosing the first mode (34) or the set of secondary modes (32) and a corresponding tile size based on a result of the comparing, and storing the selected compressed tile or subtiles into the depth buffer (20) and writing into the preset number of primary bits (34) information related to the chosen first mode (34) or set of secondary compression modes (32) or a corresponding size.

3 Citations
Depth buffer compression for stochastic motion blur rasterization  
Patent #
CN 103,518,227 A
Filed 04/26/2012

Current Assignee

Compressed depth cache  
Patent #
CN 103,546,158 A
Filed 04/08/2013

Current Assignee

Constant buffer size multisampled antialiasing depth compression  
Patent #
CN 106,575,440 A
Filed 08/20/2015

Current Assignee

No References
32 Claims

1. One kind be used for selecting will be in the method for the compact model of depth buffer (20) application, described method comprises:

Acquisition is corresponding to the value of the sheet (30) of described depth buffer (20); Select first compact model (34) from more than first compact model; Compress the value of described (30), based on described first pattern (34) value of being compressed is write first, and use the main bit (34) of predetermined number store with described first pattern (34) and described first compression sizes in the relevant information of at least one item; Select the set of less important compact model (32) from more than second compact model; Determine to form a plurality of subsheet (28) of described (30), compress the value of described subsheet, described set based on less important compact model (32) writes second with the value of being compressed, and use the main bit (34) of identical predetermined number store with the described set of less important compact model (32) and described second compression sizes at least one relevant information; The compression sizes of the compression sizes of more described (30) and described subsheet (28); Choose the described set and the corresponding sheet size of described first pattern (34) or less important pattern (32) based on the result of described comparison;
AndThe sheet of selected compression or subsheet stored in the described depth buffer (20) and the information relevant with the set of selected first pattern (34) or less important compact model (32) or corresponding size write into the main bit (34) of described predetermined number.


2. the method for claim 1, wherein each subsheet comprises 4 * 4 pixels, and described comprises 8 * 8 pixels.

3. the method for claim 1, wherein the described predetermined number of main bit is 4 or 8.

4. the method for claim 1, that wherein said more than second compact model comprises is unpressed, remove, biplane and depth shift pattern.

5. the method for claim 1, the main bit of wherein said predetermined number are enough to only expression corresponding to some combination of the compact model of described subsheet.

6. the method for claim 1, wherein said more than first compact model comprises monoplane and biplane mode.

7. the method for claim 1 also comprises:
For k set of the main bit assigned value of described predetermined number being used to indicating described compact model, and be described predetermined number main bit assigned value 2 ^{K1}2 ^{K2}2 ^{K3}... 2 ^{Kn}k set is to be used to indicate the compact model of described subsheet, and wherein k, k1, k2, k3, kn and n are integers.

8. the method for claim 1 also comprises:

Prevent combination corresponding to the compact model of the subsheet of combination with minimum appearance;
AndAssign the bit combination of these combinations of expression to transfer to represent whole compact model.


9. method as claimed in claim 8, the combination that is prevented of the compact model of wherein said subsheet changes to another pattern by a compact model in the combination that each is prevented and revises when taking place.

10. the method for claim 1, the result of wherein said comparison is associated with the largest percentage of the compression of described and subsheet.

11. the method for claim 1, wherein each step is carried out by processor.

12. method as claimed in claim 11, the main bit storage of wherein said predetermined number is in described processor.

13. the method for claim 1 also comprises:
The size of the piece of storage described first or second compression in the main bit of described predetermined number.

14. the method for claim 1 also comprises:
In the less important bit of the predetermined number of the main bit that is different from described predetermined number, storage and the first selected pattern or the relevant information of set of less important compact model.

15. method as claimed in claim 14 is at least one in the piece of described first and second compressions of the less important bit storage of wherein said predetermined number.

16. method as claimed in claim 14, the main bit storage of wherein said predetermined number are in processor, and the less important bit storage of described predetermined number is in the storer of described processor outside.

17. one kind be used for selecting will be at the device (10) of the compact model of depth buffer (20) application, described device comprises:

Input end (22) is configured to the value of acquisition corresponding to the sheet (30) of described depth buffer (20); Processor (12) is configured to Select first compact model (34) from more than first compact model, Compress the value of described (30), based on described first pattern (34) value of being compressed is write first, and the main bit (34) of predetermined number is used for storing at least one relevant information with described first pattern and described first compression sizes Select the set of less important compact model (32) from more than second compact model, Determine to form a plurality of subsheet (28) of described (30), based on the described set of less important compact model (32) value of the being compressed compression of described subsheet is write second, and the main bit (34) of identical predetermined number is used for the described set of storage and less important compact model (32) and the relevant information of at least one item of described second compression sizes The compression sizes of the compression sizes of more described (30) and described subsheet (28), and Choose the described set and the corresponding sheet size of described first pattern (34) or less important pattern (32) based on the result of described comparison;
AndDescribed depth buffer (20) is configured to store the main bit (34) of the described predetermined number relevant with the described set of selected first pattern (34) or less important compact model (32) or described corresponding sheet size.


18. device as claimed in claim 17, wherein each subsheet comprises 4 * 4 pixels, and described comprises 8 * 8 pixels.

19. device as claimed in claim 17, wherein the described predetermined number of main bit is 4 or 8.

20.
That 20. device as claimed in claim 17, wherein said more than second compact model comprise is unpressed, remove, biplane and depth shift pattern.

21. device as claimed in claim 17, the main bit of wherein said predetermined number are enough to only expression corresponding to the combination of the compact model of described subsheet.

22. device as claimed in claim 17, wherein said more than first compact model comprises monoplane and biplane mode.

23. device as claimed in claim 17, wherein said processor also is configured to:
For k set of the main bit assigned value of described predetermined number being used to indicating described compact model, and be described predetermined number main bit assigned value 2 ^{K1}2 ^{K2}2 ^{K3}... 2 ^{Kn}k set is to be used to indicate the compact model of described subsheet, and wherein k, k1, k2, k3, k4 and n are integers.

24. device as claimed in claim 17, wherein said processor also is configured to:

Prevent combination corresponding to the compact model of the subsheet of two combinations with minimum appearance;
AndDescribed two combinations are assigned to described compact model.


25. device as claimed in claim 24, wherein said processor also is configured to:
When taking place, change to the combination that is prevented that compact model is not revised the compact model of described subsheet by a compact model in the combination that each is prevented.

26. device as claimed in claim 24, the result of wherein said comparison is associated with described largest percentage with the compression of subsheet.

27. device as claimed in claim 17, the main bit storage of wherein said predetermined number is in described processor.

28. device as claimed in claim 17, the size of the piece of the main bit storage described first of wherein said predetermined number and/or second compression.

29. device as claimed in claim 17 wherein is different from the relevant information of set of less important bit storage and selected compact model of predetermined number of the main bit of current quantity.

30. device as claimed in claim 29 is at least one in the piece of described first and second compressions of the less important bit storage of wherein said predetermined number.

31. device as claimed in claim 29 also comprises:

The storer of described processor outside, The main bit storage of wherein said predetermined number is in described processor, and the less important bit storage of described predetermined number is in described storer.


32. the compact model that the computerreadable media of a storage computation machine executable instruction, wherein said instruction impel described processor selection to use in depth buffer (20) when being carried out by processor, described instruction comprises:

Acquisition is corresponding to the value of the sheet (30) of described depth buffer (20); Select first compact model (34) from more than first compact model; Compress the value of described (30), based on described first pattern (34) value of being compressed is write first, and the main bit (34) of predetermined number is used for the relevant information of at least one item of storage and described first pattern (34) and described first compression sizes; Select the set of less important compact model (32) from more than second compact model; Determine to form a plurality of subsheet (28) of described (30), compress the value of described subsheet, described set based on less important compact model (32) writes second with the value of being compressed, and the main bit (34) of identical predetermined number is used for the described set of storage and less important compact model (32) and the relevant information of at least one item of described second compression sizes; The compression sizes of the compression sizes of more described (30) and described subsheet (28); Choose the described set and the corresponding sheet size of described first pattern (34) or less important pattern (32) based on the result of described comparison;
AndThe sheet of selected compression or subsheet stored in the described depth buffer (20) and the information relevant with the described set of selected first pattern (34) or less important compact model (32) or corresponding size write into the main bit (34) of current quantity.

Specification(s)