Adobe, Inc.

Xerox Corporation

Edward R Fiala

Daniel H Greene

Michael D Cooper

James G Sandman, Jr

Michael L Hodgson

Paul D Rovner

Noam Lorberbaum

Patrick Swee-Hock Ong

Loretta G Reid

Patrick S-h Ong

Raymond Wai-Keung Wong

A method and apparatus for allocating memory usage in a visual output device having a render engine, a display list memory for storing display list objects and a compressed band buffer memory for storing compressed bitmap data rendered by the render engine. The method includes storing display list objects in the display list memory and detecting that the display list memory is full. Upon detecting the display list memory is full and determining that the compressed band buffer memory is not full, display list data is transferred from display list memory to the compressed band buffer memory, thereby freeing space in the display list memory space for more display list objects.

Printer memory boost

A method of generating halftone output on an output device. The method includes retrieving at least one table which associates gray levels with halftone results for a region of an output frame including more than one pixel, determining a reference gray level for a tile of an output frame including more than one pixel, using the at least one table to determine a halftone result for the reference gray level, storing the halftone result for the reference gray level to the tile of the output frame, and delivering the output frame to the output device. The table may be generated from a set of thresholds, which may be a threshold array. A first table may associates gray levels with codes and a second table may associates codes with halftone results, and the method may include converting the reference gray level to a reference code with the first table and converting the reference code to the halftone result for the reference gray level with the second table. Converting the reference gray level to a code may occur during a front-end interpretation and converting the reference code to the halftone result may occur during a back-end rendering.

Method and apparatus for generating halftone output

A page printer uses alternative internal representations for the print data in each of multiple lateral strips that make up the page. Page description language input is converted to an intermediate display list format that is sorted into display commands associated with each strip. If during input the display list representation of the data fills the memory allotted to its storage, one or more of the strips is converted from display list format to raster format and compressed using a lossless compression algorithm. If storage of compressed strips subsequently exceeds available memory, raster data is further compressed using a lossy compression algorithm. Additional display list commands are rasterized by decompressing the compressed strips, processing the new commands and recompressing the data using lossless or lossy compression as required. If the display list does not overflow memory during input, the complete strips are rasterized and compressed one at a time in the order in which they will be printed. The rasterizing time is stored. Should memory overflow at any point in the process, then the rasterizing times for the remaining strips are determined without storing the raster output. A decision can then be made for each strip to rasterize it to the print engine on the fly, pre-compute and store it as lossless compressed raster data, or revert to a lossy-compressed pre-computed version.

Method and apparatus for saving printer memory

A page printer uses alternative internal representations for the print data in each of multiple lateral strips that make up the page. Page description language input is converted to an intermediate display list format that is sorted into display commands associated with each strip. If during input the display list representation of the data fills the memory allotted to its storage, one or more of the strips is converted from display list format to raster format and compressed using a lossless compression algorithm. If storage of compressed strips subsequently exceeds available memory, raster data is further compressed using a lossy compression algorithm. Additional display list commands are rasterized by decompressing the compressed strips, processing the new commands and recompressing the data using lossless or lossy compression as required. If the display list does not overflow memory during input, the complete strips are rasterized and compressed one at a time in the order in which they will be printed. The rasterizing time is stored. Should memory overflow at any point in the process, then the rasterizing times for the remaining strips are determined without storing the raster output. A decision can then be made for each strip to rasterize it to the print engine on the fly, precompute and store it as lossless compressed raster data, or revert to a lossy-compressed precomputed version.

Non-redundant encoding for textual substitution data compression systems is provided by encoding the structure of search tree constructed and maintained by the compressor for the compression of the source data, thereby enabling an encoder to reconstruct an identical search tree for expanding the data.

Search tree data structure encoding for textual substitution data

In accordance with the present invention source data is encoded by literal codewords of varying length value, with or without the encoding of copy codewords of varying length and displacement value. Copy codeword encoding is central to textual substitution-style data compression, but the encoding of variable length literals may be employed for other types of data compression.

Growing community of inventors

Sunnyvale, CA, United States of America

Edward R Fiala