Silicon Graphics, Incorporated

Microsoft Technology Licensing, LLC

Nvidia Corporation

Graphics Properties Holdings, Inc.

Graphic Properties Holdings, Inc.

Christopher J Migdal

Robert A Drebin

John Stephen Montrym

Mark S Peercy

David L Dignam

John Milligan Airey

Philippe Lacroute

Alexander Lev Minkin

Michael Timothy Jones

David S Wang

James L Foran

Danny Loh

Danny D Loh

Christopher Clark Tanner

Walter E Donovan

William P Newhall, Jr

Yury Y Uralsky

Angus Dorbie

Anders M Kugler

David L Morgan, Iii

Mehmet Cem Cebenoyan

Pemith Randima Fernando

David (Chi-Shung) Wang

Amy J Migdal

Lup-Yen Peter Young

A floating point rasterization and frame buffer in a computer system graphics program. The rasterization, fog, lighting, texturing, blending, and antialiasing processes operate on floating point values. In one embodiment, a 16-bit floating point format consisting of one sign bit, ten mantissa bits, and five exponent bits (s10e5), is used to optimize the range and precision afforded by the 16 available bits of information. In other embodiments, the floating point format can be defined in the manner preferred in order to achieve a desired range and precision of the data stored in the frame buffer. The final floating point values corresponding to pixel attributes are stored in a frame buffer and eventually read and drawn for display. The graphics program can operate directly on the data in the frame buffer without losing any of the desired range and precision of the data.

Floating point computer system with fog

One embodiment of the present invention sets forth a technique for transitioning from bilinear sampling to filter-4 sampling, while avoiding the visual artifacts along the boundary between the two different types of filters. The technique may be implemented using a linear transition function or an arbitrary transition function stored in a lookup table. The transition to filter-4 sampling occurs when the view of a textured object includes both minified and magnified levels of texture detail. Using this technique, high image quality is maintained for texture mapped images that include both highly minified pixels as well as highly magnified pixels, without suffering the performance penalty associated with using a filtering operation such as filter-4 sampling across all pixels.

Systems and methods for smooth transitions to bi-cubic magnification

Display system having floating point rasterization and floating point framebuffering

A method and system for performing a texture operation with user-specified offset positions are disclosed. Specifically, one embodiment of the present invention sets forth a method, which includes the steps of deriving a first destined texel position based on an original sample position associated with a pixel projected in a texture map and a first offset position specified by a user and fetching texel attributes at the first destined texel position for the texture operation.

Method and system for processing texture samples with programmable offset positions

The present invention provides for a method of and apparatus for compressing and uncompressing image data. According to one embodiment of the present invention, the method of compressing a color cell comprises the steps of: defining at least four luminance levels of the color cell; generating a bitmask for the color cell, the bitmask having a plurality of entries each corresponding to a respective one of the pixels, each of the entries for storing data identifying one of the luminance levels associated with a corresponding one of the pixels; calculating a first average color of pixels associated with a first one of the luminance levels; calculating a second average color of pixels associated with a second one of the luminance levels; and storing the bitmask in association with the first average color and the second average color. In one embodiment, the color cell includes a matrix of 4&#xd7;4 pixels, the bitmask includes 32-bits and each of the color values includes 16-bits such that a compression rate of 4-bits per pixel is achieved. The present invention is particularly applicable to compress texture data such that the texture data can be more efficiently cached and moved during texture mapping. In that embodiment, the present invention can also support the compression of luminance, intensity and alpha textures.

Method of and apparatus for compressing and uncompressing image data

The present invention provides for a method of and apparatus for compressing and uncompressing image data. According to one embodiment of the present invention, the method of compressing a color cell comprises the steps of: defining at least four luminance levels of the color cell; generating a bitmask for the color cell, the bitmask having a plurality of entries each corresponding to a respective one of the pixels, each of the entries for storing data identifying one of the luminance levels associated with a corresponding one of the pixels; calculating a first average color of pixels associated with a first one of the luminance levels; calculating a second average color of pixels associated with a second one of the luminance levels; and storing the bitmask in association with the first average color and the second average color. In one embodiment, the color cell includes a matrix of 4x4 pixels, the bitmask includes 32-bits and each of the color values includes 16-bits such that a compression rate of 4-bits per pixel is achieved. The present invention is particularly applicable to compress texture data such that the texture data can be more efficiently cached and moved during texture mapping. In that embodiment, the present invention can also support the compression of luminance, intensity and alpha textures.

A scaleable network based computer system having a distributed texture memory architecture. A network residing within the computer system is used to transmit packets between a host processor and a number of subsystems. Three basic types of subsystems are coupled to the network: a geometry subsystem is used to process primitives; a rasterization subsystem is used to render pixels; and a display subsystem is used to drive a computer monitor. Any number and combination of these three types of subsystems can be coupled to one or more network chips to implement a wide variety of configurations. One or more memory chips can be coupled to any of the rasterization subsystems. These memory chips are used to store texture data. A rasterization subsystem can access texture data from its associated memory chips or can request texture data residing within any of the other memory chips. The request is sent over the internal network; the requested texture data is packetized and sent over the internal network to the requesting rasterization subsystem. Because texture data is distributed across the internal network and is accessible to any and all rasterization subsystems, there is no need to store duplicate copies. Furthermore, the computer system is readily scaleable simply by adding the appropriate geometry, rasterization, or display subsystem. Likewise, texture memory is readily scalable simply by adding more memory chips.

Computer system having a distributed texture memory architecture

A method and apparatus for efficiently managing texture memory in computer graphics systems is provided. Texture images are stored in discrete memory-aligned tiles to avoid fragmentation in the texture memory. Larger texture images are divided up into smaller tiles so that they will fit in any available tile region in texture memory. Small texture images usually fit into a single tile and therefore do not usually have to be divided up. Texture images that are larger than a tile region are split up into tile-sized images that are stored individually in any available tile region of texture memory. By dividing up the larger texture images this way, the texture memory is used more efficiently because any gaps that appear in the texture memory due to fragmentation may be filled by the tile-sized images.

Method and apparatus for texture memory management

A cache memory for high latency and out-of-order return of texture data. The present invention includes a texture cache memory that prefetches texture data before it is needed. Further, the texture cache memory counts the number of times a cache line is requested and a number of times the cache line is read, and determines whether the cache line is free by keeping track of the different between the two numbers. The texture cache memory of the present invention is capable of working efficiently in computer systems where there is a long latency from the time the texture data is requested and the time the texture data is available for use. In addition, the present invention is capable of handling texture data which enters into the texture cache memory in a different order from which it was requested. The present invention significantly improves performance of the texture data retrieval subsystem within network based or memory hierarchy based computer systems.

Cache memory for high latency and out-of-order return of texture data

An apparatus and method for quickly and efficiently providing texel data relevant for displaying a textured-image. A large amount of texture source data, such as photographic terrain texture, is stored as a two-dimensional or three-dimensional texture MIP-map on one or more mass storage devices. Only a relatively small clip-map representing selected portions of the complete texture MIP-map is loaded into faster, more expensive memory. These selected texture MIP-map portions forming the clip-map consist of tiles which contain those texel values at each respective level of detail that are most likely to be mapped to pixels being rendered for display based upon the viewer's eyepoint and field of view. To efficiently update the clip-map in real-time, texel data is loaded and discarded from the edges of tiles. Attempts to access a texel lying outside of a particular clip-map tile are accommodated by utilizing a substitute texel value obtained from the next coarser resolution clip-map tile which encompasses the sought texel.

Method and system for providing texture using a selected portion of a texture map

A system and method for multiple texture rendering on a primitive using a fine grain multi-pass at a pixel level. The present invention has hardware capable of processing one texture and either hardware, software, and/or firmware capability to hold state information associated with several textures. The hardware rapidly switches between processing different textures and allows a very fine grain multi-pass implementation of multiple texture rendering. In one embodiment, a computer graphics raster subsystem renders multiple textures on a primitive. A fine grain scan converter takes in a primitive description with multiple sets of texture coordinates defined for each vertex. Each set of texture coordinates defines an independent texture. Each texture is associated with a texture number. The fine grain scan converter produces a set of texture coordinates at a given pixel for each of the multiple textures before moving on to the next pixel. A texture address generator and texture filter use the texture numbers and the texture coordinates associated with the texture numbers to produce a filtered texture value for each of the multiple textures. A recirculating texture blender combines color and these filtered texture values to produce a final textured color value. The fine-grain multi-pass technique for rendering multiple textures on a primitive is faster than a coarse grain multi-pass technique because the geometry does not need to be sent multiple times from a geometry engine to a raster subsystem. In addition, the fine-grain multi-pass technique is cheaper and more efficient than a parallel approach because it does not require permanent multiple texture units in a hardware connected together by texture blending units.

Fine grain multi-pass for multiple texture rendering

An apparatus and method for quickly and efficiently providing texel data relevant for displaying a textured image. A large amount of texture source data, such as photographic terrain texture, is stored as a two-dimensional or three-dimensional texture MIP-map on one or more mass storage devices. Only a relatively small clip-map representing selected portions of the complete texture MIP-map is loaded into faster, more expensive memory. These selected texture MIP-map portions forming the clip-map consist of tiles which contain those texel values at each respective level of detail that are most likely to be mapped to pixels being rendered for display based upon the viewer's eyepoint and field of view. To efficiently update the clip-map in real-time, texel data is loaded and discarded from the edges of tiles. Attempts to access a texel lying outside of a particular clip-map tile are accommodated by utilizing a substitute texel value obtained from the next coarser resolution clip-map tile which encompasses the sought texel.

Growing community of inventors

Cupertino, CA, United States of America

Christopher J Migdal