Microsoft Technology Licensing, LLC

David Floyd Aronson

Anuj Bharat Gosalia

Craig C Peeper

Loren McQuade

Charles Neil Boyd

Paul Lawrence Bleisch

Daniel Kurt Baker

David John Martin

Parham Mohadjer

Matthew Russell Kimball

Jonathan Steed

Michael David Anderson

Daniel Paul Horowitz

Michael Burrows

Daniel Reed Baker

J Andrew Goossen

Amar Dinesh Patel

Juan Carlos Arevalo Baeza

Jason Matthew Gould

Matthew William Lee

John H Palevich

Iouri Tarassov

Karen Elaine Stevens

Anantha R Kancheria

Bruce Michael Dawson

Randal N Linden

Andrew Goossen

Systems and methods for using memory usage to pinpoint sub-optimal code for gaming systems are provided herein. Memory usage characteristics, such as latency, cache misses, load-hit-store, memory address misuse, and wasted cache bandwidth are presented, preferably in a graphical format, to provide the developer with information for optimizing source code. A trace analysis is performed on source code undergoing optimization. Relevant data is extracted from the trace analysis, sorted as necessary, and presented to the user. The user may be presented with multiple results sorting mechanisms as well as ways to change the presentation of the results.

Using memory usage to pinpoint sub-optimal code for gaming systems

A system and methods for providing shading effects include a shading network possessing the ability to create, maintain, and incrementally evolve shading techniques and effects. The shader network provides the ability to encapsulate CPU and GPU code into a self-contained node that can be transported between renderers. As a result, authors can develop reusable shading techniques that can express the wide range of effects available via utilization of CPU and GPU code. The shader network also provides the capability to compose multiple nodes into a reusable 'macro' node that affords authors the ability to build derivative shading effects without the need to modify an application's renderer or constituent nodes.

Shader network

A high level shader language compiler incorporates transforms to optimize shader code for graphics processing hardware. An instruction reordering transform determines instruction encapsulations of dependent instructions that reduce concurrent register usage by the shader. A phase pulling transform re-organizes the shader's instructions into phases that reduce a measure of depth of texture loads. A register assigning transform assigns registers to lower register usage by the shader.

Optimizing compiler transforms for a high level shader language

The performance of a video game is analyzed using non-intrusive capture and storage of game data. A non-linear capture format is used for capturing run-time game data. The run-time game data includes run-time parameters associated with execution of an application code as well as run-time parameters associated with hardware of a game platform upon which the application code is being executed. The captured data is stored in a storage medium using a non-contiguous storage format.

Analyzing video game performance using non-intrusive capture and storage of run-time game data

An application renders a visual simulation to a display as a series of visual frames according to a plurality of calls. Each frame contains objects. A particular frame n of the simulation is captured for re-rendering. An identification of the frame n to be captured is received, a run file for the identified frame n is created, and for the frame n each call issued in connection with such frame n by the application is captured. For each issued call in order, all establishing calls necessary to establish state for the issued call are firstly written to the run file, and the issued call is then secondly written to the run file. Thus, the run file as produced represents an ordered list of calls that may be executed by an executable on any of a plurality of computing devices to re-render the frame independent from the application.

Encapsulating frame of graphics data for replay

Various technologies for visualizing a multi-dimensional view object in a two-dimensional format are described. In accordance with one described embodiment, an interface for visualizing a multi-dimensional view object in two dimensions includes a surface selection tree panel, a view object information panel, and a visualization panel. The surface selection tree panel displays a surface selection tree associated with the multi-dimensional view object. A number of other surface selection trees associated with the multi-dimensional view object also exist and can be displayed if selected. The surface selection trees include one or more surfaces associated with the multi-dimensional view object. The view object information panel complements the surface selection tree panel by displaying information associated with the multi-dimensional view object. Furthermore, the visualization panel displays an image or a spreadsheet-like numeric representation of the surface area that corresponds to a currently selected surface.

Viewing multi-dimensional data in two dimensions

A method for analyzing the performance of a video game uses a diagnostic tool that is associated with application code of the video game. The diagnostic tool is activated when the video game is in operation, and real-time performance data is captured and displayed. A warning is generated when a performance metric violates a pre-set condition. The warning may be displayed on a display screen that is used to provide information for rectifying the violation.

Real-time analysis of performance data of a video game

A shader program capable of execution on a GPU is analyzed for constant expressions. These constant expressions are replaced with references to registers or memory addresses on the GPU. A preshader is created that comprises two executable files. The first executable file contains the shader program with the each constant expression removed and replaced with a unique reference accessible by the GPU. The first file is executable at the GPU. A second file contains the removed constant expressions along with instructions to place the values generated at the associated reference. The second executable file is executable at a CPU. When the preshader is executed, an instance of the first file is executed at the GPU for each vertex or pixel that is displayed. One instance of the second file is executed at the CPU. As the preshader is executed, the constant expressions in the second file are evaluated and the resulting intermediate values are passed to each instance of the first file on the GPU.

Preshaders: optimization of GPU pro

A standard graphics specification for use by both developers of graphics files as well as of applications developed to execute in a runtime environment is disclosed. The graphics files are developed to conform to the graphics specification and therefore will be executable by applications in any runtime environment that likewise conform to the graphics specification. The specification includes program syntax standards and standards for metadata in the form of semantics and annotations that further describe the code. The specification additionally includes standards to which applications may conform to ensure that the applications will be capable of executing any graphics files that conform to the graphics specification.

Standard graphics specification and data binding

Preshaders: optimization of GPU programs

Complex computer graphics forms and motions can be constructed either by hand or with motion or geometry capture technologies, once they are created, they are difficult to modify, particularly at runtime. Interpolation provides a way to leverage artist-generated source material. Methodologies for efficient runtime interpolation between multiple forms or multiple motion segments enables computers to perform more realistic animation in real-time. Shape interpolation is applied to predefined figures to create smoothly skinned figures that deform in natural ways. Predefined figures are selected using a search technique that reduces the amount of interpolation required to produce real-time animation.

Methods and system for general skinning via hardware accelerators

Usage semantics allow for shaders to be authored independently of the actual vertex data and accordingly enables their reuse. Usage semantics define a feature that binds data between distinct components to allow them to work together. In various embodiments, the components include high level language variables that are bound by an application or by vertex data streams, high level language fragments to enable several fragments to be developed separately and compiled at a later time together to form a single shader, assembly language variables that get bound to vertex data streams, and parameters between vertex and pixel shaders. This allows developers to be able to program the shaders in the assembly and high level language with variables that refer to names rather than registers. By allowing this decoupling of registers from the language, developers can work on the language separately from the vertex data and modify and enhance high level language shaders without having to manually manipulate the registers. This also allows the same shaders to work on different sets of mesh data, allowing the shaders to be reused. Generally, semantics can be used as a data binding protocol between distinct areas of the programmable pipeline to allow for a more flexible workflow.

Growing community of inventors

Woodinville, WA, United States of America

David Floyd Aronson