IP Library Granted Patent US 9,858,704
Granted Patent B2
US 9,858,704 · App. 15/182,694 · Granted Jan 2, 2018

Reduced precision ray traversal with plane reuse

Inventors: Karthik Vaidyanathan (Berkeley, CA); Marco Salvi (Seattle, WA)
Assignee: Intel Corporation
G06T15/06
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,858,704
App. No.
15/182,694
Granted
Jan 2, 2018
Kind
B2
Abstract

A computation for a parent node may be reused in a child node in a reduced precision bounding volume hierarchy ray traversal for graphics processing.

Claims (34)

1. A method comprising:

reusing, in a child node, a computation for a parent node in a reduced precision bounding volume hierarchy ray traversal for graphics processing.

2. The method of claim 1 including for minimum node coordinates, identifying coordinates that are different from the minimum parent node coordinates.

3. The method of claim 2 including calculating differences between parent and child node coordinates.

4. The method of claim 3 including quantizing the differences.

5. The method of claim 2 including for maximum node coordinates, identifying coordinates that are different from the maximum parent node coordinates.

6. The method of claim 5 including calculating differences between parent and child node coordinates.

7. The method of claim 6 including quantizing the differences.

8. The method of claim 1 including determining where a ray enters a bounding box and using a first quantized coordinate r i to get a first parametric distance λ i .

9. The method of claim 8 including determining where the rays exists in the bounding box using a second quantized coordinate s i to get a second parametric distance μ i .

10. The method of claim 9 including comparing μ i and λ i to determine ray/bounding box intersection by multiplying the ray's slope and λ i and the ray's slope and μ i using reduced precision multiplication.

11. One or more non-transitory computer readable media storing instructions to perform a sequence comprising:

reusing, in a child node, a computation for a parent node in a reduced precision bounding volume hierarchy ray traversal for graphics processing.

12. The media of claim 11 including for minimum node coordinates, identifying coordinates that are different from the minimum parent node coordinates.

13. The media of claim 12 including calculating differences between parent and child node coordinates.

14. The media of claim 13 including quantizing the differences.

15. The media of claim 12 including for maximum node coordinates, identifying coordinates that are different from the maximum parent node coordinates.

16. The media of claim 15 including calculating differences between parent and child node coordinates.

17. The media of claim 16 including quantizing the differences.

18. The media of claim 11 including determining where a ray enters a bounding box and using a first quantized coordinate r i to get a first parametric distance λ i .

19. The media of claim 18 including determining where the rays exists in the bounding box using a second quantized coordinate s i to get a second parametric distance μ i .

20. The media of claim 19 including comparing μ i and λ i to determine ray/bounding box intersection by multiplying the ray's slope and λ i and the ray's slope and μ i using reduced precision multiplication.

21. An apparatus comprising:

a processor to reuse, in a child node, a computation for a parent node in a reduced precision bounding volume hierarchy ray traversal for graphics processing; and

a memory coupled to said processor.

22. The apparatus of claim 21 , said processor to include for minimum node coordinates, identify coordinates that are different from the minimum parent node coordinates.

23. The apparatus of claim 22 , said processor to calculate differences between parent and child node coordinates.

24. The apparatus of claim 23 , said processor to quantize the differences.

25. The apparatus of claim 22 , said processor to include for maximum node coordinates, identify coordinates that are different from the maximum parent node coordinates.

26. The apparatus of claim 25 , said processor to calculate differences between parent and child node coordinates.

27. The apparatus of claim 26 , said processor to quantize the differences.

28. The apparatus of claim 21 , said processor to determine where a ray enters a bounding box and using a first quantized coordinate r i to get a first parametric distance λ i .

29. The apparatus of claim 28 , said processor to determine where the rays exists in the bounding box using a second quantized coordinate s i to get a second parametric distance μ i .

30. The apparatus of claim 29 , said processor to compare μ i and λ i to determine ray/bounding box intersection by multiplying the ray's slope and λ i and the ray's slope and μ i using reduced precision multiplication.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 30, 2016
From: VAIDYANATHAN, KARTHIK; SALVI, MARCO
To: INTEL CORPORATION
Reel/Frame 039057/0969 →
Continuity (2)
Provisional Application 62317684 · Apr 4, 2016
Related Publication 20170287203A1 · Oct 5, 2017