Abstract: A hierarchical acceleration structure may be built for graphics processing using a 32 bit format. In one embodiment, the acceleration structure may be a k-d tree, but other acceleration structures may be used as well. 64 bit offsets are only used when 64 bit offsets are needed.

Abstract: A hierarchical acceleration structure may be built for graphics processing using a 32 bit format. In one embodiment, the acceleration structure may be a k-d tree, but other acceleration structures may be used as well. 64 bit offsets are only used when 64 bit offsets are needed.

Abstract: A technique to promote load balancing in parallel graphics rendering. In one embodiment, one or more threads are associated with one or more rendering tiles and scheduled in a balanced manner on one or more graphics processing resources.

Abstract: A technique to improve ray tracing performance. In one embodiment, polygons not intersecting a specially created frustum are excluded from further ray-triangle tests, thereby enabling more efficient traversal of an acceleration structure corresponding to the given scene.

Abstract: A technique to promote load balancing in parallel graphics rendering. In one embodiment, one or more threads are associated with one or more rendering tiles and scheduled in a balanced manner on one or more graphics processing resources.

Abstract: A technique to improve ray tracing performance. In one embodiment, polygons not intersecting a specially created frustum are excluded from further ray-triangle tests, thereby enabling more efficient traversal of an acceleration structure corresponding to the given scene.

Abstract: A block decomposition based method is provided for compressing textures based on sharing color information between texture blocks. The texture map is divided into a plurality of texture blocks containing texels. During compression a first color value is stored for a first block and a second color value is stored for a second block. A palette is created for each texture block using the color values stored for several neighboring texture blocks. Each texel in each of the texture blocks is indexed to one of the colors in the palette. Accordingly the method compresses the information in the texture map by storing for each texture block bits representing a color value and bits to index each of the texels to a color in the palette, which allows sharing of color information between blocks. During decompression the palettes are regenerated and colors from the palette are assigned to the texels according to the stored index values.

Abstract: A method and apparatus for binary classification includes using signs of float values to detect different subgroups, detecting whether all entries in the group belong to the same subgroup, splitting original subgroup into uniform subgroups and classifying subgroups using array of float values. Coherency in groups of rays is detected by generating a group of rays, determining an originating point and a direction for each ray in the group, determining coherency of the group of rays and determining a group of rays as coherent as one in which all rays determined to travel in the same direction for each coordinate x, y, and z and determining a group of rays as incoherent otherwise and traversing the group of incoherent rays differently from the coherent group of rays.

Abstract: A method and apparatus for a frustum culling algorithm suitable for hardware implementation. In one embodiment, the method includes the separation of coordinates of a normal vector of each frustum plane of a frustum view into positive normal coordinates and negative normal coordinates. In one embodiment, the separation of the coordinates of each normal vector of the frustum planes enables implicit selection of the coordinates of a negative vertex (N-vertex) of an axis-aligned bounded box (AABB). Once implicitly selected, it is determined whether the N-vertex of the AABB is outside at least one frustum plane. In one embodiment, a determination that the N-vertex of the AABB is outside at least one of the frustum planes provides a trivial reject of objects enclosed by the AABB that are therefore is excluded from the rendering process. Other embodiments are described and claimed.

Abstract: Embodiments provide for ray tracing traversal that relies on selected geometrical properties of the application to reduce the number of operations required during each traversal step. The traversal algorithm does not depend on the number of rays in the group. As a result, multi-level traversal schemes may be implemented, starting with a large number of rays in a group and then reducing it as needed to maintain group coherency. Multi-level traversal schemes may be created by splitting large groups of rays while traversing acceleration structures.

Abstract: Embodiments of the invention provide for accelerated polygon intersection testing of rays against a set of polygons. The amount of computation required in the rendering process is reduced by preprocessing the scene into a data structure that can be more efficiently traversed. During the preprocessing stage, triangles such as triangle may be converted into vertex and edge representation.

Abstract: A given computer graphics scene may be rendered as a set of triangles. A set of photons may be distributed over the scene, and a number of steps may be performed for each triangle. For each triangle, a list of photons may be constructed. For that triangle, a set of control points may be identified for purposes of determining global illumination. For each control point, a specific illumination estimate may be computed. A kd-tree of the control points of the triangle may be built. An illumination estimate may then be drive for the triangle.

Abstract: A block decomposition based method is provided for compressing textures based on sharing color information between texture blocks. The texture map is divided into a plurality of texture blocks containing texels. During compression a first color value is stored for a first block and a second color value is stored for a second block. A palette is created for each texture block using the color values stored for several neighboring texture blocks. Each texel in each of the texture blocks is indexed to one of the colors in the palette. Accordingly the method compresses the information in the texture map by storing for each texture block bits representing a color value and bits to index each of the texels to a color in the palette, which allows sharing of color information between blocks. During decompression the palettes are regenerated and colors from the palette are assigned to the texels according to the stored index values.

Abstract: A block decomposition based method is provided for compressing textures based on sharing color information between texture blocks. The texture map is divided into a plurality of texture blocks containing texels. During compression a first color value is stored for a first block and a second color value is stored for a second block. A palette is created for each texture block using the color values stored for several neighboring texture blocks. Each texel in each of the texture blocks is indexed to one of the colors in the palette. Accordingly the method compresses the information in the texture map by storing for each texture block bits representing a color value and bits to index each of the texels to a color in the palette, which allows sharing of color information between blocks. During decompression the palettes are regenerated and colors from the palette are assigned to the texels according to the stored index values.