Claims:1. An apparatus comprising:
a ray generator to generate a plurality of rays;
ray direction evaluation circuitry/logic to generate approximate ray direction data for each of the plurality of rays; and
ray sorting circuitry/logic to sort the rays into a plurality of ray queues based, at least in part, on the approximate ray direction data.

2. The apparatus of claim 1 wherein the approximate ray direction data comprises a quantized direction value associated with each ray of the plurality of rays.

3. The apparatus of claim 2 wherein the quantized direction value for each ray comprises first data indicating a side of a volume intersected by the ray and second data comprising quantized intersection coordinates of an intersection between the ray and the side of the volume.

4. The apparatus of claim 2 wherein the ray sorting circuitry/logic is to group one or more of the plurality of rays into the plurality of ray queues based on a combination of the quantized direction value and a shader record key associated with the ray.

5. The apparatus of claim 4 wherein the ray sorting circuitry/logic is to first attempt to match a ray with a ray queue using both the quantized ray direction value and the shader record key and to attempt to match the ray with a ray queue using only the shader record key only when a match is not found.

6. The apparatus of claim 5 wherein when a match is not found using the both the quantized ray direction value and the shader record key, the ray sorting circuitry/logic is to attempt to allocate a new ray queue containing the ray.

7. The apparatus of claim 6 wherein the sorting circuitry/logic is to attempt to match the ray with a ray queue using only the shader record key only after determining that the new ray queue cannot be allocated.

8. The apparatus of any of claims 1 to 7 further comprising:
a ray dispatcher to dispatch the plurality of rays in groups defined by the ray queues in which the rays are stored.

9. The apparatus of any of claims 1 to 8 further comprising:
ray traversal circuitry to traverse one or more of the plurality of rays through a bounding volume hierarchy; and
ray intersection circuitry to determine intersections between one or more of the plurality of rays and one or more objects in a scene.

10. A method comprising:
generating a plurality of rays;
determining approximate ray direction data for each of the plurality of rays; and
sorting the rays into a plurality of ray queues based, at least in part, on the approximate ray direction data.

11. The method of claim 10 wherein the approximate ray direction data comprises a quantized direction value associated with each ray of the plurality of rays.

12. The method of claim 11 wherein the quantized direction value for each ray comprises first data indicating a side of a volume intersected by the ray and second data comprising quantized intersection coordinates of an intersection between the ray and the side of the volume.

13. The method of claim 11 wherein sorting further comprises:
grouping the plurality of rays into the plurality of ray queues based on a combination of the quantized direction value and a shader record key associated with the ray.

14. The method of claim 13 further comprising:
initially attempting to match a ray with a ray queue using both the quantized ray direction value and the shader record key; and
attempting to match the ray with a ray queue using only the shader record key only when a match is not found.

15. The method of claim 14 further comprising:
attempting to allocate a new ray queue containing the ray when a match is not found using the both the quantized ray direction value and the shader record key.

16. The method of claim 15 wherein attempting to match the ray with a ray queue using only the shader record key is performed only after determining that the new ray queue cannot be allocated.

17. The method of any of claims 10 to 16 further comprising:
dispatching the plurality of rays in groups defined by the ray queues in which the rays are stored.

18. The method of any of claims 10 to 17 further comprising:
traversing one or more of the plurality of rays through a bounding volume hierarchy; and
determining intersections between one or more of the plurality of rays and one or more objects in a scene.

19. A machine-readable medium having program code stored thereon which, when executed by a machine, causes the machine to perform the operations of:
generating a plurality of rays;
determining approximate ray direction data for each of the plurality of rays; and
sorting the rays into a plurality of ray queues based, at least in part, on the approximate ray direction data.

20. The machine-readable medium of claim 19 wherein the approximate ray direction data comprises a quantized direction value associated with each ray of the plurality of rays.

21. The machine-readable medium of claim 20 wherein the quantized direction value for each ray comprises first data indicating a side of a volume intersected by the ray and second data comprising quantized intersection coordinates of an intersection between the ray and the side of the volume.

22. The machine-readable medium of claim 20 wherein sorting further comprises:
grouping the plurality of rays into the plurality of ray queues based on a combination of the quantized direction value and a shader record key associated with the ray.

23. The machine-readable medium of claim 22 further comprising program code to cause the machine to perform the operations of:
initially attempting to match a ray with a ray queue using both the quantized ray direction value and the shader record key; and
attempting to match the ray with a ray queue using only the shader record key only when a match is not found.

24. The machine-readable medium of claim 23 further comprising:
attempting to allocate a new ray queue containing the ray when a match is not found using the both the quantized ray direction value and the shader record key.

25. The machine-readable medium of claim 24 wherein attempting to match the ray with a ray queue using only the shader record key is performed only after determining that the new ray queue cannot be allocated.