Abstract: A mechanism is described for dynamically optimizing color buffer clear performance in graphics processing units. A method of embodiments, as described herein, includes allocating and initializing a first set of control bits associated with a framebuffer in a graphics processing unit (GPU), and rendering a first frame, wherein the first set of control bits are associated with the first frame. The method may further include allocating a second set of control bits associated with a second frame, and rendering the second frame. The method may further include facilitating an expedited resolve operation of the second frame based on a frame-to-frame coherence associated with the first and second frames.

Abstract: In position-only shading, two geometry pipes exist, a trimmed down version called the Cull Pipe and a full version called the Replay Pipe. Thus, the Cull Pipe executes the position shaders in parallel with the main application, but typically generates the critical results much faster as it fetches and shades only the position attribute of the vertices and avoids the rasterization as well as the rendering of pixels for the frame buffer. Furthermore, the Cull Pipe uses these critical results to compute visibility information for all the triangles whether they are culled or not. On the other hand, the Replay Pipe consumes the visibility information to skip the culled triangles and shades only the visible triangles that are finally passed to the rasterization phase. Together the two pipes can hide the long cull runs of discarded triangles and can complete the work faster in some embodiments.

Abstract: Various embodiments are generally directed to techniques for reducing storage access bandwidth requirements in retrieving a texture image from a storage for applying textures to rendered objects by rendering the texture image itself into the storage to reduce the storage space in which the texture image is stored and to arrange portions of the texture image to be retrieved with fewer accesses. A device to render images includes a processor component; a color analyzer to determine a clear color of a texture image stored as source texture data; and a rendering routine to render the texture image into a storage as reduced texture data, the rendering routine to selectively store in the reduced texture data pixel color values retrieved from the source texture data that are associated with pixels of the texture image not colored with the clear color. Other embodiments are described and claimed.

Abstract: Various embodiments are generally directed to an apparatus, method and other techniques to determine color information for multiple graphical layers of a graphical display at a location of a pixel, and to determine a pixel color information for the pixel at the location based on the color information for each of the multiple graphical layers.

Abstract: Apparatus, systems and methods for handling portrait mode oriented display surfaces without requiring expensive hardware in the display sub-system are disclosed. For example, an apparatus is disclosed such that the rendering of graphics data to the portrait mode display surfaces is redirected at rendering time such that there is no need for adding complicated hardware in the display part of the graphics adapter in order to handle conventional displays—all of which have no circuitry to deal with data natively stored in a portrait mode surface. Additionally, an apparatus to handle direct surface access of a surface through a surface lock which has already been rotated is already described. This can either be done by copying of surface data or by an optimized proposed apparatus which eliminates this copy. Other implementations are also disclosed.

Abstract: According to some embodiments of the present invention, pixel throughput may be improved by performing depth tests and recording the results on the granularity of an input geometry object. An input geometry object is any object within the depiction represented by a primitive, such as a triangle within an input triangle list or a patch within an input patch list.

Abstract: Apparatus, systems and methods for handling portrait mode oriented display surfaces without requiring expensive hardware in the display sub-system are disclosed. For example, an apparatus is disclosed such that the rendering of graphics data to the portrait mode display surfaces is redirected at rendering time such that there is no need for adding complicated hardware in the display part of the graphics adapter in order to handle conventional displays—all of which have no circuitry to deal with data natively stored in a portrait mode surface. Additionally, an apparatus to handle direct surface access of a surface through a surface lock which has already been rotated is already described. This can either be done by copying of surface data or by an optimized proposed apparatus which eliminates this copy. Other implementations are also disclosed.

Abstract: Apparatus, systems and methods for handling portrait mode oriented display surfaces without requiring expensive hardware in the display sub-system are disclosed. For example, an apparatus is disclosed such that the rendering of graphics data to the portrait mode display surfaces is redirected at rendering time such that there is no need for adding complicated hardware in the display part of the graphics adapter in order to handle conventional displays—all of which have no circuitry to deal with data natively stored in a portrait mode surface. Additionally, an apparatus to handle direct surface access of a surface through a surface lock which has already been rotated is already described. This can either be done by copying of surface data or by an optimized proposed apparatus which eliminates this copy. Other implementations are also disclosed.

Abstract: A method is described that involves redefining a primitive surface to create a redefined primitive surface. The redefining involves clipping a first edge of the primitive surface to a second edge of a displayable screen area. The first edge of the primitive surface extends outside a guard band region prior to the clipping. The guard band region surrounds the displayable screen area. The clipping causes the redefined primitive surface to have an edge running within the guard band that the primitive surface does not have.

Abstract: An apparatus and method to render concave polygons. In some embodiments, the method includes setting an alpha buffer of a hardware rasterizer to zero, generating an alpha mask by rendering the concave polygon as a plurality of convex polygons of connected vertices, wherein an alpha value for each of the plurality of the plurality of polygons is toggled between a zero value and a second non-zero value on successive renderings of each portion of the plurality of the convex polygons. The alpha values may be stored in the alpha buffer, and the plurality of convex polygons may be re-rendered based on the generated alpha mask, where the stored alpha value is the second non-zero value.

Abstract: An apparatus may include two or more frame buffers, a control module, a management module, and a display engine. The two or more frame buffers may each store frame data arranged in a plurality of lines. The control module may designate one of the frame buffers for output. This designation may change for each frame output to a display device. The management module identifies the lines associated with the designated frame buffer as either valid or invalid. More particularly, the management module identifies a line as invalid when the line has changed in at least one of the two or more buffers since the designated buffer's previous designation for output. The display engine fetches, from the designated buffer, any lines identified as invalid. These fetched lines may be sent to the display device for output. Additionally, the fetched lines may be compressed and stored by the display engine.

Abstract: Apparatus, systems and methods for handling portrait mode oriented display surfaces without requiring expensive hardware in the display sub-system are disclosed. For example, an apparatus is disclosed such that the rendering of graphics data to the portrait mode display surfaces is redirected at rendering time such that there is no need for adding complicated hardware in the display part of the graphics adapter in order to handle conventional displays—all of which have no circuitry to deal with data natively stored in a portrait mode surface. Additionally, an apparatus to handle direct surface access of a surface through a surface lock which has already been rotated is already described. This can either be done by copying of surface data or by an optimized proposed apparatus which eliminates this copy. Other implementations are also disclosed.

Abstract: A method is described to render a line stipple pattern in an ad hoc piecewise fashion by creating a texture map having information that represents the line stipple pattern. The method also includes, for each of a plurality of line primitives that each represent a different section of the line stipple pattern, mapping a section of the texture map's information to its corresponding line primitive.

Abstract: A method and apparatus for rasterizing a line, including determining a start point and initial line step count of the line, determining a line stipple pattern position and repeat count at each point of the line based on the start point of the line and initial line step count and rasterizing the line in an order independent manner. In some embodiments, the line may be rasterized backwards. An approximate reciprocal value for the repeat count may be used.

Abstract: An apparatus and method to render concave polygons. In some embodiments, the method includes setting an alpha buffer of a hardware rasterizer to zero, generating an alpha mask by rendering the concave polygon as a plurality of convex polygons of connected vertices, wherein an alpha value for each of the plurality of the plurality of polygons is toggled between a zero value and a second non-zero value on successive renderings of each portion of the plurality of the convex polygons. The alpha values may be stored in the alpha buffer, and the plurality of convex polygons may be re-rendered based on the generated alpha mask, where the stored alpha value is the second non-zero value.

Abstract: A method is described that involves storing 1D texture map information within a 2D texture map. The 1D texture map has a size that exceeds a maximum allowable 1D texture map size.

Abstract: A technique for detecting texture sharing between multiple contexts having unique ID's includes obtaining a texture usage mask of a subject texture and a context ID of a subject context. A first logic operation with the texture usage mask of the subject texture and the context ID of the subject context is performed to produce a resultant value. The subject texture is determined not to be shared by another context with the subject context upon the resultant value being equal to a first predetermined value. The subject texture is determined to be shared by another context with the subject context upon the resultant value being equal to a second predetermined value which is different from the first predetermined value. The texture usage mask of a subject texture may be revised prior to the subject texture being used by another context by performing a second logic operation with the texture usage mask and a context ID of another context to produce a resultant new texture usage mask for the subject texture.

Abstract: A technique for detecting texture sharing between multiple contexts having unique ID's includes obtaining a texture usage mask of a subject texture and a context ID of a subject context. A first logic operation with the texture usage mask of the subject texture and the context ID of the subject context is performed to produce a resultant value. The subject texture is determined not to be shared by another context with the subject context upon the resultant value being equal to a first predetermined value. The subject texture is determined to be shared by another context with the subject context upon the resultant value being equal to a second predetermined value which is different from the first predetermined value. The texture usage mask of a subject texture may be revised prior to the subject texture being used by another context by performing a second logic operation with the texture usage mask and a context ID of another context to produce a resultant new texture usage mask for the subject texture.

Abstract: A method is described that involves storing 1D texture map information within a 2D texture map. The 1D texture map has a size that exceeds a maximum allowable 1D texture map size.

Abstract: A method is described to render a line stipple pattern in an ad hoc piecewise fashion by creating a texture map having information that represents the line stipple pattern. The method also includes, for each of a plurality of line primitives that each represent a different section of the line stipple pattern, mapping a section of the texture map's information to its corresponding line primitive.