Media coprocessor with graphics video and audio tasks partitioned by time division multiplexing

• US 6,275,239 B1
• Filed: 08/20/1998
• Issued: 08/14/2001
• Est. Priority Date: 08/20/1998
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Silicon Graphics Incorporated (Graphics Properties Holdings, Inc.)
• Inventors: Rao, Sudhaker, Ezer, Gulbin, Nicholson, Ronald, Van Hook, Timothy J.
• Primary Examiner(s): NGUYEN, PHU K
• Application Number:
  US09/137,017
• Time in Patent Office: 1,090 Days
• Field of Search: 345/473, 345/474, 345/418, 345/475, 345/118, 345/121
• US Class Current: 345/473
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining
• CPC Class Codes: H04N 19/42 characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation H04N19/635 takes precedence
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

A media coprocessor for performing 3-D graphics, video, and audio functions. The media coprocessor is comprised of a single IC semiconductor chip which is coupled with a host processor chip, one or more memory chips, and an I/O controller chip. The media coprocessor includes a digital bitstream processor, a digital signal processor, and a display processor. An update interval, synchronized to a video frame, is defined. This update interval is divided into a number of partitions. Audio data is processed during one of the partitions. Video data is processed during another partition. And 3-D graphics is processed in another partition. Thereby, the processing is performed in a sequential, time-division multiplex scheme whereby the single media coprocessor chip processes all three partitions in a single video frame.

Claims

1. A single semiconductor chip for performing audio, video, and 3-D graphics functions comprising:

a digital bitstream processor for decoding variable length code digital bitstream data;

a digital signal processor coupled to the digital bitstream processor which performs video and audio decompressing, filtering, geometry processing, and performing wavetable synthesis for audio;

a display processor coupled to the digital signal processor which supports both geometric 3-D graphics and image data types, wherein an update interval sychronized to a video frame is defined, the update interval is divided into a plurality of partitions such that audio data is processed during a first partition, video data is processed during a second partition, and graphics data is processed during a third partition, the processing being performed in a sequential, time-division multiplex scheme whereby the single semiconductor chip processes all three partitions in a single video frame;

a memory controller coupled to an external memory array, said memory array used to store image data and instruction data;

an audio/video input/output port coupled to the display processor.

2. The single semiconductor chip of claim 1, wherein the display processor renders textured, shaded, antialiased, Z-buffered, alpha blended polygons.

3. The single semiconductor chip of claim 2, wherein the display processor performs motion compensation and reconstruction for video.

4. The single semiconductor chip of claim 3, wherein the display processor scales, rotates, translates, filters, and composites primitives with per-pixel transparency.

5. The single semiconductor chip of claim 1, wherein the digital signal processor includes a floating point unit coupled to the interface which calculates 3-D graphics, video and audio operations.

6. The single semiconductor chip of claim 1 further comprising an internal memory coupled to the floating point unit for storing texture data and video data.

7. The single semiconductor chip of claim 1, wherein the digital signal processor includes a filter coupled to the internal memory for use in performing both mip-mapping and motion compensation.

8. A system for performing 3-D graphics, video, and audio functions, comprising:

a central processing unit for processing data;

a memory array coupled to the central processing unit for storing instructions data and image data pertaining to 3-D graphics and video;

a media coprocessor wherein an update interval sychronized to a video frame is defined, the update interval is divided into a plurality of partitions such that audio data is processed during a first partition, video data is processed during a second partition, and graphics data is processed during a third partition, the processing being performed in a sequential, time-division multiplex scheme whereby the media coprocessor processes all three partitions in a single video frame;

an input/output controller coupled to the media coprocessor for accepting commands and outputting audio, video, and 3-D graphics data.

9. The system of claim 8, wherein the media coprocessor includes:

a digital bitstream processor for decoding variable length code digital bitstream data;

a digital signal processor coupled to the digital bitstream processor which performs video and audio decompressing, filtering, and geometry processing;

a display processor coupled to the digital signal processor which supports both geometric 3-D graphics and image data types;

a memory controller coupled to an external memory array, said memory array used to store image data and instruction data;

an audio/video input/output port coupled to the display processor.

10. The single semiconductor chip of claim 9, wherein the display processor renders textured, shaded, antialiased, Z-buffered, alpha blended polygons;

performs motion compensation and reconstruction for video;

scales, rotates, translates, filters, and composites primitives with per-pixel transparency;

operates in a full color image type for all 3-D graphics rendering as well as video and display.

11. The single semiconductor chip of claim 9, wherein the digital signal processor includes a filter coupled to the internal memory for use in performing both mip-mapping and motion compensation.

12. A method for performing audio, video, and 3-D graphics functions comprising the steps of:

defining an update time;

partitioning the update time into a plurality of partitions;

processing audio data in a first partition;

processing video data in a second partition;

processing 3-D graphics data in a third partition, wherein an update interval sychronized to a video frame is defined, the update interval is divided into a plurality of partitions such that audio data is processed during a first partition, video data is processed during a second partition, and graphics data is processed during a third partition, the processing being performed in a sequential, time-division multiplex scheme wherein all three partitions in a single video frame;

outputting an audio signal;

displaying the video and 3D graphics onto a display screen.

13. The method for performing audio, video, and 3-D graphics functions of claim 12, wherein the steps of the method are performed by a single semiconductor chip.

14. The method for performing audio, video, and 3-D graphics functions of claim 12, wherein the update time is a frame time.

15. The method for performing audio, video, and 3-D graphics functions of claim 14, wherein the first partition is approximately twenty-five percent of the frame time, the second partition is approximately fifty percent of the frame time, and the third partition is approximately twenty-five percent of the frame time.

16. The method for performing audio, video, and 3-D graphics functions of claim 12, wherein the processing steps are performed sequentially in series.

17. The method for performing audio, video, and 3-D graphics functions of claim 12 further comprising the step of storing a task list in memory and storing a 3-D graphics task, an audio task, and a video task in the memory.

18. The method for performing audio, video, and 3-D graphics functions of claim 17 further comprising the step of storing an application task and a resident dispatcher in an instruction memory and resident data in a data memory.

19. The method for performing audio, video, and 3-D graphics functions of claim 12 further comprising the steps of:

performing decoding, quantization, filtering, and resampling on the audio data in the first partition;

performing decoding, quantization, discrete cosine transformation, motion compensation, reconstruction, image scaling, filtering, and compositing on the video data in the second partition;

performing lighting, polygon setup, rasterization, texturing, shading, color combination, blending, and depth comparisons on the 3-D graphics data in the third partition.