Nvidia Corporation

Brian K Angell

Edward Anthony Hutchins

Paul Kim

A rasterizer stage configured to implement multiple interpolators for graphics pipeline. The rasterizer stage includes a plurality of simultaneously operable low precision interpolators for computing a first set of pixel parameters for pixels of a geometric primitive and a plurality of simultaneously operable high precision interpolators for computing a second set of pixel parameters for pixels of the geometric primitive. The rasterizer stage also includes an output mechanism coupled to the interpolators for routing computed pixel parameters into a memory array. Parameters may be programmably assigned to the interpolators and the results thereof may be programmably assigned to portions of a pixel packet.

Method and system for implementing multiple high precision and low precision interpolators for a graphics pipeline

A present invention pixel processing system and method permit complicated three dimensional images to be rendered with shallow graphics pipelines including reduced gate counts and facilitates power conservation by utilizing a single unified data fetch stage (e.g., unified data fetch module) that retrieves a variety of different pixel surface attribute values (e.g., depth, color, and/or texture values) in a single stage. Different types of pixel surface attribute data (e.g., depth, color, texture) associated with multiple graphics processing functions (e.g., color blending, texture mapping, etc.) are retrieved in the single unified data fetch graphics pipeline stage. The pixel surface attribute values may be placed in corresponding variable fields of a pixel packet row. The pixel packet rows including the pixel surface attribute values are forwarded to downstream graphics pipeline stages (e.g., an arithmetic logic pipestage).

Unified data fetch graphics processing system and method

A present invention pixel processing system and method permit complicated three dimensional images to be rendered with shallow graphics pipelines including reduced gate counts and also facilitates power conservation. Pixel packet information includes pixel surface attribute values are retrieved in a single unified data fetch stage. At a data fetch pipestage a determination may be made if the pixel packet information contributes to an image display presentation (e.g., a depth comparison of Z values is performed determine if the pixel is occluded). A pixel packet status indicator (e.g., a kill bit) is set in the sideband portion of a pixel packet and the pixel packet is forwarded for processing in accordance with the pixel packet status indicator. The status indicator is a kill bit is set to prevent logic components from clocking information for a payload portion of the pixel packet if the status indicator indicates the pixel packet payload does not contribute to the image display presentation while continuing to clock pixel packet sideband information.

Kill bit graphics processing system and method

A present invention pixel processing system and method permit complicated three dimensional images to be rendered with shallow graphics pipelines including reduced gate counts and facilitates power conservation by utilizing a single unified data fetch stage (e.g., unified data fetch module) that retrieves a variety of different pixel surface attribute values for different attribute types (e.g., depth, color, and/or texture values) in a single stage. Different types of pixel surface attribute data (e.g., depth, color, texture) associated with multiple graphics processing functions (e.g., color blending, texture mapping, etc.) are retrieved in the single unified data fetch graphics pipeline stage. The pixel packet rows including the pixel surface attribute values are forwarded to other graphics pipeline stages for single thread processing (e.g. to a universal arithmetic logic unit capable of performing multiple graphics functions on the pixel surface attribute values).

Single thread graphics processing system and method

A pixel processing system and method which permits rendering of complicated three dimensional images using a shallow graphics pipeline including reduced gate counts and low power operation. Pixel packet information includes pixel surface attribute values retrieved in a single unified data fetch stage. A determination is made if the pixel packet information contributes to an image display presentation (e.g., a depth comparison of Z values may be performed). The pixel packet information processing is handled in accordance with results of the determining. The pixel surface attribute values and pixel packet information are removed from further processing if the pixel surface attribute values are occluded. In one exemplary implementation, the pixel packet includes a plurality of rows and the handling is coordinated for the plurality of rows. Any of a number of downstream pipestages may remove the occluded pixel information, and in response thereto, may notify a gatekeeper pipestage of the slack increase so that more pixels can be allowed into the pipeline.

Early kill removal graphics processing system and method

A graphics processor is disclosed having a programmable Arithmetic Logic Unit (ALU) stage for processing pixel packets. Scalar arithmetic operations are performed in the ALUs to implement a graphics function.

Method for operating low power programmable processor

A system and method for a data write unit in a 3-D graphics pipeline including generic cache memories. Specifically, in one embodiment a data write unit includes a first memory, a plurality of cache memories and a data write circuit. The first memory receives a pixel packet associated with a pixel. The pixel packet includes data related to surface characteristics of the pixel. The plurality of cache memories is coupled to the first memory for storing pixel information associated with a plurality of surface characteristics of a plurality of pixels. Each of the plurality of cache memories is programmably associated with a designated surface characteristic. The data write circuit is coupled to the first a memory and the plurality of cache memories. The data write circuit is operable under program control to obtain designated portions of the pixel packet for storage into the plurality of cache memories.

System and method for a universal data write unit in a 3-D graphics pipeline including generic cache memories

Embodiments of the present invention include an arithmetic logic unit for use in a graphics pipeline. The arithmetic logic unit comprising a plurality of scalar arithmetic logic subunits wherein each subunit performs a resultant arithmetic logic operation in the form of [a*b 'op' c*d] on a set of input operands a, b, c and d. The arithmetic logic unit also for produces a result based thereon wherein 'op' represents a programmable operation and wherein further the resultant arithmetic logic operation is software programmable to implement a plurality of different graphics functions.

Arithmetic logic unit and method for processing data in a graphics pipeline

A configurable graphics pipeline has more than one possible process flow of pixel packets through elements of the graphics pipeline. In one embodiment, a data packet triggers an element of the graphics pipeline to discover an identifier.

Auto software configurable register address space for low power programmable processor

A graphics processor has elements of a graphics pipeline coupled by distributors. The distributors permit the process flow of pixel packets through the pipeline to be reconfigured in response to a command from a host.

Reconfigurable pipeline for low power programmable processor

A graphics processor includes a graphics pipeline having a set of tap points. A configurable test point selector monitors a selected subset of tap points and counts statistics for at least one condition associated with each tap point of the subset of tap points.

Statistics instrumentation for low power programmable processor

A graphics processor includes an arithmetic logic unit (ALU) stage for processing pixel packets. Pixels are assigned as either even pixels or odd pixels. The pixel packets of odd and even pixels are interleaved to account for ALU latency.

Interleaving of pixels for low power programmable processor

A graphics processor includes programmable arithmetic logic units (ALUs) for performing scalar arithmetic operations on pixel packets. For a selected scalar arithmetic operation, operands in pixel packets may be formatted in a S1.8 format to improve dynamic range. For at least one other scalar arithmetic operation, the pixel packets may be formatted in a different data format.

Growing community of inventors

San Jose, CA, United States of America

Brian K Angell