Computer graphics processing and selective visual display system – Computer graphics display memory system – Frame buffer
Reexamination Certificate
1999-12-22
2003-12-23
Bella, Matthew C. (Department: 2676)
Computer graphics processing and selective visual display system
Computer graphics display memory system
Frame buffer
C345S546000, C345S568000
Reexamination Certificate
active
06667745
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of computer memory address mapping. More particularly, the present invention relates to a device for more efficiently converting a linear configuration virtual memory address to a physical memory address via an XY coordinate configuration system.
BACKGROUND OF THE INVENTION
Electronic systems and circuits have made a significant contribution towards the advancement of modem society and are utilized in a number of applications to achieve advantageous results. Numerous electronic technologies such as digital computers, calculators, audio devices; video equipment, and telephone systems include computer graphics systems that have facilitated increased productivity and reduced costs in analyzing and communicating data, ideas and trends in most areas of business, science, education and entertainment. Usually, applications designed to provide these results utilize information stored in a memory of a computer system. To be effective and operate properly, most graphics applications require the information to be communicated efficiently to and from the memory via a computer graphics system.
Computer graphics'systems typically provide a method for interfacing between a computer and a user. Often, this interfacing involves the graphical representation of images (graphics) on a display screen, other visualization device or a hard copy printout. Graphics are generated by computer graphics systems that simulate and display images of real or abstract objects. Graphics enable a user to visualize and comprehend the configuration of a single object or the interaction and relationships between a group of objects. The images usually comprise pictures in which the objects remain still, or video displays in which the objects move. Most modern computer graphics systems are interactive, permitting a user to input changes to a display or modify the images on the fly.
In a complex or three dimensional (3D) computer generated graphical image, objects are typically described by graphics data models. These coverage masks typically define the shape of the object, the object's attributes, and where the object is positioned. The shape of the object is normally described in terms of “primitives”, which usually comprise mathematically described circular disks, vectors, polygons or polyhedra. The graphics information is input into memory and a central processing unit (CPU) interprets instructions and image data in order to perform the appropriate processing. Some computer graphics systems may include special-purpose processors, each custom tailored to specific graphics functions. The main graphical processing function of the CPU (or special-purpose processors) is to simplify very complex models by taking the specifications of graphical primitives specified by application programs and to assign pixels parameter values that best represent characteristics of an image.
In most computer graphic systems an image is represented as a raster (an array) of logical picture elements (pixels). A pixel corresponds to a small area of the image, usually a rectangle (but it can be other shapes). The computer graphics system assigns parameter values to each pixel as part of a rasterization process. Rasterization can proceed on a pixel basis or primitive basis. These parameter values are digital values corresponding to certain attributes of the image (e.g. color, depth, etc.) measured over a small area of the image represented by the pixel. Typically each graphical image is represented by thousands of combined pixels. The pixel parameter values associated with an image are usually stored in a portion of a memory referred to as a frame buffer. The resolution and detail of the image are largely determined by the number of pixels in the frame buffer. The number of bits that are used for each pixel defines the depth of the frame buffer and determines properties such as how many colors can be represented on a given system. For example, a 1-bit-deep frame buffer allows only two colors. Frame buffers play an important role in rasterization.
A frame buffer is a portion of a memory that stores pixel information associated with an image. There are usually a number of frame buffers in a computer graphics system. The frame buffers may be scattered throughout a memory and are not necessarily contiguous. Frame buffers often comprise information associated with a particular image configuration system because some data configuration or reference systems are easier and more efficient for a computer graphics system to manipulate. For example, frame buffers are often configured and allocated memory space in a manner that accommodates a two dimensional “XY” tile coordinate system.
Numerous computer graphics systems prefer to operate in a two dimensional “XY” tile coordinate system.
FIG. 1
is a conceptual example of an XY coordinate image configuration system
100
. Image configuration system
100
includes 2 by 2 pixel regions, such as pixel region
191
, set in 16 by 16 region tiles
170
through
185
arranged on an “x axis”
110
, “y axis”
120
. Image configuration system
100
includes “z axis”
130
. The 2 by 2 pixel regions can be arranged in any place within the texture images and can slide around, sometimes it may fall in a single tile, sometimes it may fall within two tiles and other times it may fall within four tiles. “Slices” of a 3D image are defined by the “z axis”, for example pixel region
191
is in slice
133
and pixel region
192
is in slice
134
. Slice
134
is actually another xy plane at a different location on the Z axis and pixel region actually lies in tiles (not shown) behind tiles
275
,
276
,
279
and
280
. While a pixel region may move around the coordinate system, the boundaries of the tiles do not change and the tiles are defined by the particular XY coordinates. There are numerous tile boundary configurations, for example a 64 kilo-byte (KB) tile comprising 128 by 128 pixels and each pixel being define by a 32 bit parameter value.
Operating in a tile, XY coordinate system provides certain advantages for most graphics systems and a tile configuration system is particularly beneficial towards the end of a typical graphics pipeline. For example, scan conversion operations can be performed on a number of tiles in parallel permitting much faster scan conversions. Thus, a number of graphics buffers (e.g., image buffer, frame buffer, Z buffer, texture map buffer, etc.) are organized to reference stored data by two-dimensional tiled configuration addresses. While utilizing reference addresses that are configured in accordance with a tile configuration system is efficient for most computer graphics systems, it is not necessarily optimal in all situations.
Memory hardware is usually limited to specific configurations in which the physical location of storage spaces are arranged in a consecutive fashion and are addressed accordingly. Since memory hardware is typically set to a predefined physical location, address references in applications or systems that utilize different configuration schemes are virtual addresses (referencing a “virtual memory”) that are translated into addresses that refer to an actual physical storage location. Efficient computer systems usually handle the translations between virtual and physical reference addresses and make the distinction between a physical address that identifies a particular location in memory and a virtual address that identifies or refers to a piece of information transparent to a user.
Differing applications and systems within a computer system often prefer to communicate with memories utilizing reference addresses arranged in particular configurations. Tiled memory addressing is usually not conducive to user processes or applications that formulate memory read and write requests in a linear address format and expect responses in a linear configuration. For example, user graphics processes or applications are typically designed to refer to a contiguous linear virtual address “space” in which each s
Bella Matthew C.
Microsoft Corporation
Monestime Mackly
Woodcock & Washburn LLP
LandOfFree
System and method for linearly mapping a tiled image buffer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for linearly mapping a tiled image buffer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for linearly mapping a tiled image buffer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3137696