Hewlett-packard Company

Hewlett-packard Development Company, L.p.

Don W Dyer

Samuel C Sands

John Michael Brown

Gautam Mehrotra

Kevin T Lefebvre

Teresa M Morrison

Alan D Ward

Carol Lang

Rex A Barzee

Paul E Martz

James G Dugger

Noel D Scott

Kenneth William Shrum

Carol L Lang

A system for identification and assessment of performance optimizations implemented in the graphics environment, the identification and assessment of the performance optimizations based upon an optimized graphics call sequence generated by an application of one or more optimizations applied to a captured graphics call sequence occurring between said first and second drivers. The resulting optimized graphics call sequence causes the same graphics rendering to occur when provided to the graphics system as the original graphics call sequence. As such, the graphics application and associated graphics interface driver may then be analyzed by the application developer to identify specific modifications which, when implemented, would generate such an optimized graphics call sequence. This may include implementing specific modifications to the graphics application as well as implementing portions or all of the present invention into the driver for real-time execution.

System and method for assessing performance optimizations in a graphics system

A system for providing a graphics tool access to a computer graphics system to evaluate and control a graphics application executing on the computer graphics system. The system includes application program interface (API) event generators for performing predetermined operations relating to a graphics library function call and for generating a hook event containing results of the predetermined operations; dispatch table manger for selecting an active dispatch table from a normal operations dispatch table having function pointer to the graphics library functions and a hooks dispatch table having pointer to the API event generators; and hook event manager for enabling and configuring selected ones of the API event generators in response to a graphics tool event request. The system further includes internal event generators, integrated along various locations of a graphics pipeline managed by the graphics library, for performing predetermined diagnostic operations in the graphics system. In this embodiment the hook event manager also enables and configures selected ones of the internal event generators.

Accessing a graphics system for graphics application evaluation and control

A high precision, memory efficient method for the compression of surface normals and the inverse method for the decompression of those compressed surface normals back into surface normals. The normals are first scaled to unit length in Cartesian coordinates. Then, each of the smallest two vector components of the unit length normal is stored along with an indicator of which of the three vector components is not stored plus the algebraic sign of that vector component. Decompression of the surface normal requires first converting the two stored vector components into floating-point values and then using the equation 1&equals;x &plus;y &plus;z in order to obtain the non-stored vector component of the unit length normal.

Surface normal compression/decompression storing two vector components

A high precision, memory efficient method for the compression of surface normals into quantized normals and the inverse method for the expansion of those quantized surface normals back into surface normals. The surface of a three dimensional figure is conceptually divided into small areas, and the effective surface normal for each of these areas is related to the surface normal of a unit sphere tessellated into a similar number of small areas or tiles. A quantized normal is defined to be the tile number on the surface of the unit sphere. For a particular three dimensional figure, instead of storing surface unit normal values of &lcub;X,Y,Z&rcub; for each coordinate, the quantized surface normal value (i.e., the tile number) is stored. Thus, for a surface normal expressed in Cartesian coordinates, a compression ratio of 6:1 is possible depending upon the memory required to store real and integer values and the desired accuracy. Efficient computational methods for compressing the surface normal by obtaining the appropriate tile number and the inverse expansion of the quantized surface normal into surface normals are described. Regardless of the initial length of the surface normal, the resultant of the combined compression-expansion process is a surface normal of unit length which is the format expected by most standard graphics libraries.

Methods for high precision, memory efficient surface normal compression and expansion

A system for providing a graphics tool access to a computer graphics system to evaluate and control a graphics application executing on the computer graphics system. The system includes application program interface (API) event generators for performing predetermined operations relating to a graphics library function call and for generating a hook event containing results of the predetermined operations; dispatch table manger for selecting an active dispatch table from a normal operations dispatch table having function pointer to the graphics library finctions and a hooks dispatch table having pointer to the API event generators; and hook event manager for enabling and configuring selected ones of the API event generators in response to a graphics tool event request. The system further includes internal event generators, integrated along various locations of a graphics pipeline managed by the graphics library, for performing predetermined diagnostic operations in the graphics system. In this embodiment the hook event manager also enables and configures selected ones of the internal event generators.

System and method for asynchronously accessing a graphics system for graphics application evaluation and control

A high performance method for the compression of floating point format surface normals and the inverse method for the decompression of those compressed surface normals. Each of the three vector components of the surface normal is compressed by subtracting a constant from the floating point format value, then extracting a predefined field, and finally storing the extracted field. Decompression of the compressed surface normal requires first converting the three stored vector components into floating-point format and then adding a predefined constant to each. The compressed and decompressed surface normals may be of unit length or not as specified by the user.

Memory efficient surface normal decompression

A high-speed method for the compression and decompression of floating point numbers. The floating point numbers are biased using a predefined value and then stored in compressed format occupying less memory than that of the non-compressed floating point numbers. Decompression of the compressed floating point number follows a reverse process. These techniques are useful with applications in which the numbers to be compressed fall within a given range.

Compression of limited range floating point numbers

Memory efficient surface normal compression

Decompression of limited range floating point numbers

A high performance method for the compression of floating point format surface normals and the inverse method for the decompression of those compressed surface normals. Each of the three vector components of the surface normal is compressed by subtracting a constant from the floating point format value, then extracting a predefined field, and finally storing the extracted field. Decompression of the compressed surface normal requires first converting the three stored vector components into floating-point format and then adding a predefined constant to each. Typically the surface normals are of unit length.

High performance surface normal decompression

High performance surface normal compression

The invention provides improved methods for analyzing library calls made by a program and, particularly, improved methods of call tracing. Such methods include executing the program, recording source code instructions corresponding to at least selected calls issued during such execution, and recording at least selected data referred to in those calls. The data is recorded separately from the source code, e.g., in a separate file.

Growing community of inventors

Fort Collins, CO, United States of America

Don W Dyer