公开(公告)号：US08432410B1

公开(公告)日：2013-04-30

申请号：US11559545

申请日：2006-11-14

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G09G5/00 ,  G06T15/00 ,  G06F13/14

CPC分类号： G06T15/04 ,  G06T2200/28

摘要： A three dimensional (3D) graphics application programming interface (API) extension provides support for specifying images in a shared exponent format. The shared exponent format is used to represent high dynamic range textures in a compact encoding to reduce the memory footprint needed to store the image data compared with other high dynamic range formats. Image data is encoded to and decoded from the shared exponent format using a pixel processing pipeline. Image data encoded into the shared exponent format can be decoded and used as texture data during rendering.

摘要翻译： 三维（3D）图形应用编程接口（API）扩展提供了以共享指数格式指定图像的支持。 共享指数格式用于以紧凑编码表示高动态范围纹理，以减少与其他高动态范围格式相比存储图像数据所需的存储器占用空间。 图像数据使用像素处理流水线从共享指数格式编码和解码。 编码为共享指数格式的图像数据可以在渲染期间被解码并用作纹理数据。

公开(公告)号：US20110285743A1

公开(公告)日：2011-11-24

申请号：US13107698

申请日：2011-05-13

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G09G5/02

CPC分类号： G06T15/005 ,  G06T11/203

摘要： One embodiment of the present invention sets forth a technique for rendering paths by first generating a stencil buffer indicating pixels of the path that should be covered and then covering the path. The paths may be filled or stroked without tessellating the paths. Path rendering may be accelerated when a graphics processing unit or other processor that is configured to perform operations to generate the stencil buffer and cover the path to fill or stroke the path. When the paths are rendered in a top-to-bottom (front-to-back) order, an opacity stencil may be generated and used to avoid determining path coverage and shading for pixels that are opaque.

摘要翻译： 本发明的一个实施例提出了一种用于渲染路径的技术，首先产生一个模板缓冲器，该模板缓冲器指示应覆盖的路径的像素，然后覆盖该路径。 路径可以填充或抚摸，而无需细分路径。 当图形处理单元或其他处理器被配置为执行操作以生成模板缓冲器并覆盖路径以填充或冲程时，可以加速路径渲染。 当路径以从上到下（从前到后）的顺序呈现时，可以生成不透明度模板并用于避免确定不透明像素的路径覆盖和阴影。

公开(公告)号：US08044956B1

公开(公告)日：2011-10-25

申请号：US11833937

申请日：2007-08-03

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G06T15/50 ,  G09G5/00

CPC分类号： G06T15/503 ,  G06T11/40 ,  G06T2200/12

摘要： One embodiment of the present invention sets forth a technique for improving antialiasing quality, while minimizing performance degradation, by adaptively selecting between multisampling and supersampling on a per pixel basis. The resulting performance may be generally comparable to multisampling. At the same time, however, the resulting quality may be generally comparable to supersampling. The antialiasing technique disclosed herein determines whether to use multisampling or supersampling on a particular pixel being rendered, based on the specific coverage of the associated geometry primitive. Because many pixel centers are covered by a geometry primitive, a statistical performance advantage is gained when pixels in a rendered image can be generating using multisampling rather than supersampling. The cases where pixel centers are not covered tend to be less frequent, but are very significant to image quality. High image quality is maintained by rendering these cases using supersampling.

摘要翻译： 本发明的一个实施例提出了一种通过在每个像素的基础上自适应地选择多采样和超级采样之间来提高抗锯齿质量同时最小化性能降级的技术。 所得到的性能通常可以与多采样相当。 然而，同时，所得到的质量通常可以与超级取样相当。 本文公开的抗锯齿技术基于相关联的几何图形的特定覆盖来确定是否对正被渲染的特定像素使用多采样或超级采样。 因为许多像素中心被几何图元覆盖，所以当渲染图像中的像素可以使用多采样而不是超级采样生成时，可获得统计性能优势。 不包含像素中心的情况往往不那么频繁，但对图像质量非常重要。 通过使用超级采样来渲染这些情况来保持高图像质量。

公开(公告)号：US07009615B1

公开(公告)日：2006-03-07

申请号：US10006477

申请日：2001-11-30

申请人： Mark J. Kilgard ,  Patrick R. Brown

发明人： Mark J. Kilgard ,  Patrick R. Brown

IPC分类号： G06T1/20

CPC分类号： G06T1/60

摘要： A system, method and computer program product are provided for buffering data in a computer graphics pipeline. Initially, graphics floating point data is read from a buffer in a graphics pipeline. Next, the graphics floating point data is operated upon in the graphics pipeline. Further, the graphics floating point data is stored to the buffer in the graphics pipeline.

摘要翻译： 提供了一种用于在计算机图形管线中缓冲数据的系统，方法和计算机程序产品。 最初，图形流水线中的缓冲区读取图形浮点数据。 接下来，在图形管线中操作图形浮点数据。 此外，图形浮点数据被存储在图形流水线中的缓冲器中。

公开(公告)号：US09916674B2

公开(公告)日：2018-03-13

申请号：US13111148

申请日：2011-05-19

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G06T15/30 ,  G06T15/00 ,  G06T11/20

CPC分类号： G06T15/005 ,  G06T11/203

摘要： One embodiment of the present invention sets forth a technique for improving path rendering on computer systems with an available graphics processing unit. The technique involves reducing complex path objects to simpler geometric objects suitable for rendering on a graphics processing unit. The process involves a central processing unit “baking” a set of complex path rendering objects to generate a set of simpler graphics objects. A graphics processing unit then renders the simpler graphics objects. This division of processing load can advantageously yield higher overall rendering performance.

公开(公告)号：US09613451B2

公开(公告)日：2017-04-04

申请号：US13109763

申请日：2011-05-17

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G09G5/00 ,  G06T15/00 ,  G06T11/20

CPC分类号： G06T15/005 ,  G06T11/203

摘要： One embodiment of the present invention sets forth a technique for rendering anti-aliased paths by first generating an alpha buffer representing coverage data. To generate the alpha buffer, jittered versions of the rendered path are rendered and corresponding stencil buffers indicating sub-pixel samples of the path that should be covered are generated. After each stencil buffer is generated, the jittered path is rasterized to convert the sub-pixel coverage into coverage weights that are stored in the alpha component of a frame buffer. As each jittered path is rasterized, the coverage weights are accumulated. Finally, geometry representing the union of the jittered versions of the path is rendered to shade pixels based on the accumulated coverage weights. The anti-aliased rendered paths may be filled or stroked without tessellating the paths.

公开(公告)号：US09317960B2

公开(公告)日：2016-04-19

申请号：US13107698

申请日：2011-05-13

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G09G5/00 ,  G06T15/00 ,  G06T11/20

CPC分类号： G06T15/005 ,  G06T11/203

摘要： One embodiment of the present invention sets forth a technique for rendering paths by first generating a stencil buffer indicating pixels of the path that should be covered and then covering the path. The paths may be filled or stroked without tessellating the paths. Path rendering may be accelerated when a graphics processing unit or other processor that is configured to perform operations to generate the stencil buffer and cover the path to fill or stroke the path. When the paths are rendered in a top-to-bottom (front-to-back) order, an opacity stencil may be generated and used to avoid determining path coverage and shading for pixels that are opaque.

摘要翻译： 本发明的一个实施例提出了一种用于渲染路径的技术，首先产生一个模板缓冲器，该模板缓冲器指示应覆盖的路径的像素，然后覆盖该路径。 路径可以填充或抚摸，而无需细分路径。 当图形处理单元或其他处理器被配置为执行操作以生成模板缓冲器并覆盖路径以填充或冲程时，可以加速路径渲染。 当路径以从上到下（从前到后）的顺序呈现时，可以生成不透明度模板并用于避免确定不透明像素的路径覆盖和阴影。

公开(公告)号：US09024946B2

公开(公告)日：2015-05-05

申请号：US12879976

申请日：2010-09-10

申请人： Patrick R. Brown ,  Christopher T. Dodd ,  Mark J. Kilgard

发明人： Patrick R. Brown ,  Christopher T. Dodd ,  Mark J. Kilgard

IPC分类号： G06T15/30 ,  G06T15/00

CPC分类号： G06T15/005 ,  G06F13/1605 ,  G06T1/20 ,  G06T15/80 ,  G09G5/363

摘要： One embodiment of the present invention sets forth a technique for performing a computer-implemented method for tessellating patches. An input block is received that defines a plurality of input patch attributes for each patch as well as instructions for processing each input patch. A plurality of threads is launched to execute the instructions to generate each vertex of a corresponding output patch based on the input patch. Reads of values written during instruction execution are synchronized so threads can read and further process the values of other threads. An output patch is then assembled from the outputs of each of the threads; and emitting the output patch for further processing.

摘要翻译： 本发明的一个实施例提出了一种用于执行用于镶嵌补丁的计算机实现的方法的技术。 接收一个输入块，其定义每个补丁的多个输入补丁属性以及用于处理每个输入补丁的指令。 启动多个线程以执行指令以基于输入补丁生成相应输出补丁的每个顶点。 在指令执行期间写入的值的读取是同步的，因此线程可以读取并进一步处理其他线程的值。 然后从每个线程的输出组装输出补丁; 并发射输出补丁进行进一步处理。

公开(公告)号：US08698837B2

公开(公告)日：2014-04-15

申请号：US13110777

申请日：2011-05-18

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G09G5/02

CPC分类号： G06T15/005 ,  G06T11/203

摘要： One embodiment of the present invention sets forth a technique for rendering clipped paths by first generating clip stencil buffer state indicating pixels that are inside of the clip path. The clip stencil buffer state may also store an opacity value for each covered pixel to generate a mask that modulates the opacity of a draw path that is clipped. Clipped draw stencil buffer state is then generated indicating pixels of the draw path that should be covered based on the clip stencil buffer state and coverage of the draw path. The clipped draw path is then filled or stroked to produce the clipped draw path. The clip and draw paths may be filled or stroked without tessellating the paths. Path rendering may be accelerated when a GPU or other processor that is configured to perform operations to generate the clip stencil buffer state and the clipped draw stencil buffer state, and to fill or stroke the clipped draw path.

摘要翻译： 本发明的一个实施例提出了一种用于通过首先产生指示在剪辑路径内的像素的剪辑模板缓冲器状态来渲染剪切路径的技术。 夹子模板缓冲器状态还可以存储每个被覆盖的像素的不透明度值，以生成调制被剪切的绘图路径的不透明度的掩码。 然后生成剪切绘制模板缓冲区状态，指示基于剪辑模板缓冲区状态和绘制路径的覆盖范围应覆盖的绘制路径的像素。 然后，剪切的绘制路径被填充或抚摸以产生剪切的绘制路径。 剪辑和绘制路径可以填充或抚摸，而无需细分路径。 当被配置为执行操作以生成剪辑模板缓冲器状态的GPU或其他处理器并且剪切绘制模板缓冲器状态并且填充或描画剪切绘制路径时，可以加速路径渲染。

公开(公告)号：US08149243B1

公开(公告)日：2012-04-03

申请号：US11565993

申请日：2006-12-01

申请人： Mark J. Kilgard

发明人： Mark J. Kilgard

IPC分类号： G06T1/60

CPC分类号： G06T15/04 ,  G06T2200/28

摘要： A three dimensional (3D) graphics applications programming interface (API) extension provides support for specifying images in a packed float format. In the packed float format, floating point values of three color components are represented as N bits, where N is less than the total number of bits required for a standard half-precision or full precision format. For example, the blue, green, and red components may each be encoded to have a 5-bit exponent and a 5- or 6-bit mantissa with an implied leading 1. The packed float format is used to represent high dynamic range textures in a compact encoding to reduce the memory footprint needed to store the image data compared with other high dynamic range formats.

摘要翻译： 三维（3D）图形应用程序编程接口（API）扩展提供了以打包浮动格式指定图像的支持。 在打包浮动格式中，三个颜色分量的浮点值表示为N位，其中N小于标准半精度或全精度格式所需的总位数。 例如，蓝色，绿色和红色分量可以被编码为具有5位指数和5或6位尾数，具有隐含的前导1.打包浮动格式用于表示高动态范围纹理 与其他高动态范围格式相比，缩小编码以减少存储图像数据所需的存储器占用空间。