Varying effective resolution by screen location in graphics processing by approximating projection of vertices onto curved viewport

1. A method for graphics processing with a graphics processing system having a graphics processing unit and memory, comprising:

receiving data representing one or more vertices for a scene in a virtual space;

performing a mathematical projection of the vertices onto a screen space before rendering the scene on a display device having a plurality of pixels, wherein the mathematical projection includes a piecewise linear projection approximates the projection of the vertices onto a curved viewport by mathematically projecting the vertices onto a flat screen space followed by transformation of subsections of the flat screen space representing approximately equal solid angle to corresponding subsections of approximately equal area, wherein the transformation of the subsections reduces an effective area of the screen space that is to be rendered;

performing primitive assembly on the vertices to generate a one or more primitives in screen space from the projection of the vertices onto the screen space;

performing scan conversion on the one or more primitives to determine which pixel or pixels of the plurality of pixels are part of corresponding primitives of the one or more primitives;

generating a finished frame by performing pixel processing to assign pixel values to the pixel or pixels that are part of the corresponding primitives, wherein performing the primitive assembly and the scan conversion after performing the projection of the vertices reduces a computational load associated with rendering; and

storing the finished frame in the memory or displaying the finished frame on the display device.

2. The method of claim 1 , performing the projection of the vertices includes:

computing initial projections of the vertices from virtual space onto a planar viewport of the scene, the planar viewport being subdivided into a plurality of subsections, wherein each subsection of the plurality corresponds to an approximately equal solid angle of the screen space;

performing coarse rasterization for each primitive converted to screen space coordinates to determine which subsection or subsections the primitive overlaps; and

for each primitive of the one or more primitives performing a screenspace correction for each subsection that the primitive overlaps to map the coordinates of each vertex of the primitive to a corresponding corrected vertex coordinates associated with the subsection, wherein the screen space correction is equivalent to mapping each subsection that the primitive overlaps to a corresponding planar corrected subsection of a plurality of planar corrected subsections of the curved viewport, wherein the corrected subsections of the curved viewport have equal area.

3. The method of claim 2 , wherein performing the screen space correction includes performing a computation involving applying a piecewise linear approximation of an arctangent formula to screen space coordinates of one or more vertices of the primitive to generate corresponding corrected vertices.

4. The method of claim 2 , wherein performing the screen space correction includes performing a computation involving applying a piecewise linear approximation of the image distortion produced by physical optics in a display device to screen space coordinates of one or more vertices of the primitive to generate corresponding corrected vertices.

5. The method of claim 2 , wherein the scan conversion is performed per corrected subsection, wherein for a given primitive of the one or more primitives in a given subsection of the plurality of subsections scan conversion is performed on that portion of the given primitive located in the given subsection using the corrected vertex coordinates for the given primitive that are associated with the given subsection.

6. The method of claim 2 , wherein the subsections of the plurality of subsections of the planar viewport are rectangular.

7. The method of claim 1 , wherein the projection of the vertices, primitive assembly, and scan conversion are implemented in software by a compute shader.

8. The method of claim 1 , wherein the projection of the vertices, primitive assembly, and scan conversion are implemented in hardware.

9. The method of claim 1 , further comprising determining a portion of a screen of the display device that a user is looking at and wherein the projection is configured to vary an effective pixel resolution such that the effective pixel resolution is highest for one or more subsections of the screen containing the portion the user is looking at.

10. The method of claim 1 , wherein the projection is static for given optics and field of view of the display device.

11. The method of claim 1 , wherein the projection is static for given optics and field of view of the display device and wherein the projection is configured such that, one or more central subsections of the screen have a full effective pixel resolution, and subsections further from the center have progressively lower effective pixel resolution.

12. The method of claim 1 , wherein the display device is characterized by a field of view of 90 degrees or more.

13. The method of claim 1 , wherein the display device is a head-mounted display device.

14. A system for graphics processing, comprising:

a graphics processing unit (GPU); and

a memory,

wherein the graphics processing unit is configured to receive data representing one or more vertices for a scene in a virtual space;

wherein the GPU is configured to perform a mathematical projection of the vertices onto a screen space before rendering the scene on a display device having a plurality of pixels, wherein the mathematical projection approximates a projection of the vertices onto a curved viewport by projecting the vertices onto a flat screen space followed by transformation of subsections of the flat screen space representing approximately equal solid angle to corresponding subsections of approximately equal area, wherein the transformation of the subsections reduces an effective area of the screen space that is to be rendered;

wherein the GPU is configured to perform primitive assembly on the vertices to generate a one or more primitives in screen space from the projection of the vertices onto the screen space;

wherein the GPU is configured to perform scan conversion on the one or more primitives to determine which pixel or pixels of the plurality of pixels are part of corresponding primitives of the one or more primitives;

wherein the GPU is configured to generate a finished frame by performing pixel processing to assign pixel values to the pixel or pixels that are part of the corresponding primitives, wherein performing the primitive assembly and the scan conversion after performing the projection of the vertices reduces a computational load associated with rendering; and

wherein the GPU is configured to store the finished frame in the memory or displaying the finished frame on the display device.

15. The system of claim 14 , wherein the GPU is configured to perform the projection of the vertices by:

computing initial projections of the vertices from virtual space onto a planar viewport of the scene, the planar viewport being subdivided into a plurality of subsections, wherein each subsection of the plurality corresponds to an approximately equal solid angle of the screen space;

performing coarse rasterization for each primitive converted to screen space coordinates to determine which subsection or subsections the primitive overlaps; and

for each primitive of the one or more primitives performing a screenspace correction for each subsection that the primitive overlaps to map the coordinates of each vertex of the primitive to a corresponding corrected vertex coordinates associated with the subsection, wherein the screen space correction is equivalent to mapping each subsection that the primitive overlaps to a corresponding planar corrected subsection of a plurality of planar corrected subsections of the curved viewport.

16. The system of claim 15 , wherein the GPU is configured to perform the screen space correction by performing a computation involving applying a piecewise linear approximation of an arctangent formula to screen space coordinates of one or more vertices of the primitive to generate corresponding corrected vertices.

17. The system of claim 15 , wherein the GPU is configured to perform the screen space correction by performing a computation involving applying a piecewise linear approximation of image distortion produced by physical optics in a display device to screen space coordinates of one or more vertices of the primitive to generate corresponding corrected vertices.

18. The system of claim 15 , wherein the GPU is configured to perform the scan conversion per corrected subsection, wherein for a given primitive of the one or more primitives in a given subsection of the plurality of subsections the GPU performs the scan conversion on that portion of the given primitive located in the given subsection using the corrected vertex coordinates for the given primitive that are associated with the given subsection.

19. The system of claim 15 , wherein the subsections of the plurality of subsections of the planar viewport are rectangular.

20. The system of claim 15 , wherein the corrected subsections of the curved viewport have equal area.

21. The system of claim 20 , wherein the subsections of the plurality of subsections of the planar viewport are rectangular.

22. The system of claim 14 , wherein the GPU is configured to implement projection of the vertices, primitive assembly, and scan conversion in software with a compute shader.

23. The system of claim 14 , wherein the GPU is configured to implement the projection of the vertices, primitive assembly, and scan conversion in hardware.

24. The system of claim 14 , wherein the GPU is configured to implement coarse division of primitives between subsections in software, and the projection of the vertices, primitive assembly, and scan conversion in hardware.

25. The system of claim 24 , wherein the GPU is configured to associate subsection indices to primitive vertices in software, with each subsection index selecting a screen space projection and viewport from a palette implemented in hardware.

26. The system of claim 24 , wherein the GPU is configured to associate subsection indices to primitive vertex indices in software, with each subsection index selecting a screen space projection and viewport from a palette implemented in hardware.

27. The system of claim 14 , further comprising the display device.

28. The system of claim 14 , wherein the display device is characterized by a field of view of 90 degrees or more.

29. The system of claim 14 , wherein the display device is a head-mounted display device.

30. The system of claim 14 , wherein the system is configured to determine a portion of a screen of the display device that a user is looking at and wherein the projection is configured to vary an effective pixel resolution such that the effective pixel resolution is highest for one or more subsections of the screen containing the portion the user is looking at.

31. The system of claim 14 , wherein the projection is static for given optics and field of view of the display device.

32. The system of claim 14 , wherein the projection is static for given optics and field of view of the display device and wherein the projection is configured such that, one or more central subsections of the screen have a full effective pixel resolution, and subsections further from the center have progressively lower effective pixel resolution.

33. A non-transitory computer-readable medium having computer executable instructions embodied therein that, when executed, implement the method of claim 1 .