Systems and methods for 3-D scene acceleration structure creation and updating
Systems and methods for producing an acceleration structure provide for subdividing a 3-D scene into a plurality of volumetric portions, which have different sizes, each being addressable using a multipart address indicating a location and a relative size of each volumetric portion. A stream of primitives is processed by characterizing each according to one or more criteria, selecting a relative size of volumetric portions for use in bounding the primitive, and finding a set of volumetric portions of that relative size which bound the primitive. A primitive ID is stored in each location of a cache associated with each volumetric portion of the set of volumetric portions. A cache location is selected for eviction, responsive to each cache eviction decision made during the processing. An element of an acceleration structure according to the contents of the evicted cache location is generated, responsive to the evicted cache location.
1. A method of building an acceleration structure for use in ray tracing, the method comprising:
mapping a part of a scene to a memory location in a memory, said mapping being performed using a hash function; and
evicting the memory location so as to form an element of the acceleration structure that aggregates one or more parts of the scene mapped to that memory location.
2. The method of claim 1 , further comprising performing ray tracing using the acceleration structure.
3. The method of claim 1 , wherein the memory is a cache, and the memory location is a cache location.
4. The method of claim 1 , wherein the part of the scene is a primitive, a part of a primitive or a coverage formed for an element of the acceleration structure at a more granular level of granularity relative to the level of granularity for which said mapping is being performed.
5. The method of claim 4 , wherein parts of the scene are distributed across a number of levels of granularity, each level of granularity dividing the scene into one or more voxels, wherein more granular levels of granularity represent the scene using a greater number of smaller voxels relative to less granular levels of granularity, and less granular levels of granularity represent the scene using a smaller number of larger voxels relative to more granular levels of granularity.
6. The method of claim 5 , wherein a coverage is a portion of a voxel that bounds all geometry at least partially contained in that voxel.
7. The method of claim 5 , wherein the scene is a three-dimensional scene, and a voxel is a volume containing a portion of the scene.
8. The method of claim 1 , wherein the hash function is configured to hash an address of the part of the scene to produce a hash value.
9. The method of claim 8 , wherein the hash value is used to identify one or more candidate memory locations.
10. The method of claim 8 , wherein the address comprises a spatial coordinate and data identifying a level of granularity for the part of the scene.
11. The method of claim 1 , further comprising forming a coverage for the element of the acceleration structure, that coverage bounding each of the one or more parts of the scene mapped to same memory location.
12. A system configured to build an acceleration structure for use in ray tracing, the system comprising:
a memory; and
processing logic configured to:
map a part of a scene to a memory location in the memory, said mapping being performed using a hash function; and
evict the memory location so as to form an element of the acceleration structure that aggregates one or more parts of the scene mapped to that memory location.
13. The system of claim 12 , wherein the processing logic is further configured to perform ray tracing using the acceleration structure.
14. The system of claim 12 , wherein the memory is a cache, and the memory location is a cache location.
15. The system of claim 12 , wherein the part of the scene is a primitive, a part of a primitive or a coverage formed for an element of the acceleration structure at a more granular level of granularity relative to the level of granularity for which said mapping is being performed.
16. The system of claim 12 , wherein the hash function is configured to hash an address of the part of the scene to produce a hash value.
17. The system of claim 16 , wherein the hash value is used to identify one or more candidate memory locations.
18. The system of claim 16 , wherein the address comprises a spatial coordinate and data identifying a level of granularity for the part of the scene.
19. The system of claim 12 , wherein the processing logic is further configured to form a coverage for the element of the acceleration structure, that coverage bounding each of the one or more parts of the scene mapped to same memory location.