Nvidia Corporation

Edward M Veeser

Brad W Simeral

David G Reed

Stefano A Pescador

Koen Bennebroek

Karthik Bhat

Krishnan Sreenivas

Sanford S Lum

Roman Surgutchik

Sreenivas Krishnan

Joshua Titus

Anand Srinivasan

James P Reilley

Display data used in display frame generation are compressed for efficient storage in a local memory within a graphics processing unit. The compression technique used is difference encoding and before performing difference encoding, display data in RGB format are converted into YCbCr format. Since the component values of adjacent pixels in YCbCr format typically vary less than the component values of the same adjacent pixels in RGB format, converting the display data to YCbCr format before performing difference encoding improves the compression efficiency.

Compression of display data stored locally on a GPU

One embodiment of the invention sets forth a technique for compressing and storing display data and optionally compressing and storing cursor data in a memory that is local to a graphics processing unit to reduce the power consumed by a mobile computing device when refreshing the screen. Compressing the display data and optionally the cursor data also reduces the relative cost of the invention by reducing the size of the local memory relative to the size that would be necessary if the display data were stored locally in uncompressed form. Thus, the invention may improve mobile computing device battery life, while keeping additional costs low.

Screen compression for mobile applications

Display data and video data are stored within a graphics processing unit to reduce power consumed by the computing device during video playback. Storing display data and video data within the GPU reduces power consumption, because bus transaction activity is reduced and the need to read data from a larger, common main memory is avoided.

Power savings in a computing device during video playback

A speculative transfer mechanism transfers a source synchronous read request from a first clock domain to a second clock domain. The address portion having address information is transferred to the second clock domain in response to detecting a source synchronous address strobe latching signal. A pointer is generated in response to detecting the address strobe latching signal and passed into the second clock domain. In one embodiment, a pointer is retimed to be stable for a timing window for which a crossover of the address portion may be performed in the second clock domain. Request logic in the second clock domain generates a read command based on the address portion and the pointer.

Apparatus, system, and method for fast read request transfer through clock domains

Growing community of inventors

Austin, TX, United States of America

Edward M Veeser