Walter Donovan - Saratoga CA, US Anders Kugler - Sunnyvale CA, US Christopher Donham - San Mateo CA, US
Assignee:
NVIDIA Corporation - Santa Clara CA
International Classification:
G09G 5/00
US Classification:
345586000
Abstract:
Floating-point texture filtering units leverage existing fixed-point filter circuits. Groups of floating-point texture values are converted to products of a fixed-point mantissa and a scaling factor that is the same for each texture value in the group. The fixed-point mantissas are filtered using a fixed-point filter circuit, and the filtered mantissa is combined with the scaling factor to determine a floating-point filtered value. Multiple floating-point filter results may be combined in a floating-point accumulator circuit. The same fixed-point filter circuit may also be used to filter fixed-point texture data by providing fixed-point input path that bypasses the format conversion and a fixed-point accumulator.
Walter E. Donovan - Saratoga CA, US Anders M. Kugler - Sunnyvale CA, US Christopher D.S. Donham - San Mateo CA, US
Assignee:
NVIDIA Corporation - Santa Clara CA
International Classification:
G09G 5/00
US Classification:
345582
Abstract:
Floating-point texture filtering units leverage existing fixed-point filter circuits. Groups of floating-point texture values are converted to products of a fixed-point mantissa and a scaling factor that is the same for each texture value in the group. The fixed-point mantissas are filtered using a fixed-point filter circuit, and the filtered mantissa is combined with the scaling factor to determine a floating-point filtered value. Multiple floating-point filter results may be combined in a floating-point accumulator circuit. The same fixed-point filter circuit may also be used to filter fixed-point texture data by providing fixed-point input path that bypasses the format conversion and a fixed-point accumulator.
Brian Cabral - San Jose CA, US Edward A. Hutchins - Mountain View CA, US Christopher Donham - San Mateo CA, US
International Classification:
G06T 15/40 G06T 1/20 G09G 5/36
US Classification:
345422, 345506
Abstract:
Early Z scoreboard tracking systems and methods in accordance with the present invention are described. Multiple pixels are received and a pixel depth raster operation is performed on the pixels. The pixel depth raster operation comprises discarding a pixel that is occluded. In one exemplary implementation, the depth raster operation is done at a faster rate than a color raster operation. Pixels that pass the depth raster operation are checked for screen coincidence. Pixels with screen coincidence are stalled and pixels without screen coincidence are forwarded to lower stages of the pipeline. The lower stages of the pipeline are programmable and pixel flight time can vary (e.g., can include multiple passes through the lower stages). Execution through the lower stages is directed by a program sequencer which also directs notification to the pixel flight tracking when a pixel is done processing.
Parallelogram Unified Primitive Description For Rasterization
Edward A. Hutchins - Mountain View CA, US Christopher D.S. Donham - San Mateo CA, US
International Classification:
G06T 1/00
US Classification:
345501, 345420
Abstract:
In a graphics pipeline of a graphics processor, a method for a unified primitive description for rasterization. The method includes receiving a group of primitives from a graphics application, wherein the group includes different types of primitives and the types of primitives include line primitives, point primitives and triangle primitives. For each of the types of primitives, the method includes generating a corresponding parallelogram, wherein the parallelogram has four sides disposed along an x-axis and a y-axis, and computing an inside y-axis mid point and an outside y-axis mid point based on the four sides. The parallelogram is controlled to represent to each of the primitive types respectively by adjusting a location of the inside y-axis mid point or the outside y-axis mid point.
Method And System For Evaluating Derivatives In Screen Space Using Perspective Corrected Barycentric Coordinates
Method and system for evaluating derivatives in screen space using perspective corrected barycentric coordinates. A preferred embodiment provides a method for computing the LOD at individual pixels directly without relying on approximations. In this embodiment, screen coordinates of a pixel and derivatives of the texture coordinates at the vertices of the triangle enclosing the pixel are determined. Derivatives of texture coordinates at the pixel with respect to screen space are evaluated by interpolation of the derivatives of the texture coordinates at the vertices using barycentric coordinates of the pixel. Then, using the derivatives of the texture coordinates and without relying on neighboring pixels texture coordinates, the LOD at the pixel is computed, such that the LOD as computed is unbiased in any particular direction. Significantly, such direct computation allows exact LOD values to be easily computed and eliminates the artifacts that are inherent in prior art implementations. This embodiment also enables a higher level of parallel processing by eliminating the prerequisite of computing texture coordinates before computing LOD, thereby enhancing system performance.