Method and apparatus for sampling pattern generation for a ray tracing architecture
An apparatus and method for sampling pattern generation in a ray tracing architecture. For example, one embodiment of a graphics processing apparatus comprises: a ray generation circuit to generate a ray stream from one or more image tiles; and a sample pattern generation circuit to generate samples for rays in the ray stream, the samples generated to exhibit at least some randomness across pixels of a given frame but be repeatable across multiple frames.
1. A graphics processing apparatus comprising:
a ray generation circuit to generate a ray stream from one or more image tiles, wherein samples are generated for rays in the ray stream, the samples generated to exhibit at least some randomness across pixels of a given frame but be repeatable across multiple frames, wherein the exhibited randomness of the samples is in accordance with a probability distribution function, wherein the sample generation uses a sample pattern index to identify a particular sample pattern stored in a memory, and wherein the particular sample pattern is used to generate a set of rays in the ray stream.
2. The graphics processing apparatus as in claim 1 , wherein identifying the particular sample pattern comprises retrieving an address of a sample table stored in the memory.
3. The graphics processing apparatus as in claim 2 , wherein a sample set index is used to identify a particular set of samples from the sample table.
4. The graphics processing apparatus as in claim 3 , wherein the sample table comprises a hierarchical arrangement comprising N strata with each strata including M clusters, each cluster comprising O samples.
5. The graphics processing apparatus as in claim 4 , wherein the particular sample pattern is formed by reading all samples from one particular cluster from each of the strata.
6. The graphics processing apparatus as in claim 5 , wherein the sample table comprises 8 strata, each with 8 clusters, and 4 samples per cluster.
7. The graphics processing apparatus as in claim 3 , wherein the particular sample pattern is to be selected based on one or more of: different shapes of each light source, different bidirectional reflectance distribution functions (BRDFs) for different materials, and cosine-weighted hemispherical sampling patterns for ambient occlusion.
8. The graphics processing apparatus as in claim 1 , further comprising:
an intersection circuit to perform intersection tests for each ray against one or more primitives to generate intersection results.
9. The graphics processing apparatus as in claim 8 , further comprising:
a plurality of shaders to be dispatched to perform shading operations on the intersection results.
10. A system comprising:
a memory to store data and program code;
a central processing unit (CPU) comprising an instruction cache for caching a portion of the program code and a data cache for caching a portion of the data, the CPU further comprising execution logic to execute at least some of the program code and responsively process at least some of the data, at least a portion of the program code comprising graphics commands; and
a graphics processing subsystem to process the graphics commands and responsively render a plurality of image frames, the graphics processing subsystem comprising:
a ray generation circuit to generate a ray stream from one or more image tiles, wherein samples are generated for rays in the ray stream, the samples generated to exhibit at least some randomness across pixels of a given frame but be repeatable across multiple frames, wherein the exhibited randomness of the samples is in accordance with a probability distribution function, wherein the sample generation uses a sample pattern index to identify a particular sample pattern stored in the memory, and wherein the particular sample pattern is used to generate a set of rays in the ray stream.
11. The system as in claim 10 wherein identifying the particular sample pattern comprises retrieving an address of a sample table stored in the memory.
12. The system as in claim 11 , wherein a sample set index is used to identify a particular set of samples from the sample table.
13. The system as in claim 12 , wherein the sample table comprises a hierarchical arrangement comprising N strata with each strata including M clusters, each cluster comprising O samples.
14. The system as in claim 13 , wherein the particular sample pattern is formed by reading all samples from one particular cluster from each of the strata.
15. The system as in claim 14 , wherein the sample table comprises 8 strata, each with 8 clusters, and 4 samples per cluster.
16. The system as in claim 12 , wherein the particular sample pattern is to be selected based on one or more of: different shapes of each light source, different bidirectional reflectance distribution functions (BRDFs) for different materials, and cosine-weighted hemispherical sampling patterns for ambient occlusion.
17. The system as in claim 10 , further comprising:
an intersection circuit to perform intersection tests for each ray against one or more primitives to generate intersection results.