Samsung Electronics Co., Ltd.

Abhinav Golas

John Wakefield Brothers

Karthik Ramani

Santosh Abraham

Nicholas Sohre

Joohoon Lee

Christopher Thomas Cheng

Ki Fung Chow

Liangjun Zhang

Sangheon Lee

David Conrad Tannenbaum

Michael K Fertig

Vikash Kumar

Seonggun Kim

Manshila Adlakha

Pavel Krajcevski

Vandit Mehra

Santosh George Abraham

Christopher Cheng

Woody Chow

Inventive aspects include a binning unit for coarse depth culling during binning of pixel geometries. The binning unit includes a rasterizer to receive primitives, and generates pixel coverage information and depth information. The binning unit includes one or more local culling stages to perform local culling within a window of primitives. The local culling unit outputs a set of surviving coverage and surviving depth information. The binning unit includes one or more global culling stages to use the set of the surviving coverage and the surviving depth information to further cull based on an entirety of previously received coverage information and depth information.

Coarse depth culling during binning

A method of performing adaptive shading of image frames by a graphics processing unit (GPU) includes determining, by the GPU, a first shading rate based on determining that a change in a plurality of underlying assets between a first image frame and a second image frame is above a first threshold; determining, by the GPU, a second shading rate based on determining that one or more viewports in the second image frame is similar to one or more viewports in the first image frame; determining, by the GPU, a third shading rate based on determining that a quality reduction filter is used; and selecting, by the GPU, a shading rate from among the first shading rate, the second shading rate, and the third shading rate for the first image frame.

Motion based adaptive rendering

A method of variable rate compression including: partitioning the image data into a plurality of blocks, each including channel-blocks; and for one of the one or more non-flat channel-blocks: generating a residual block including residual values corresponding to values of the one of the one or more non-flat blocks; in response to determining that a particular residual value of the residual values is equal to a corresponding previous residual value: incrementing a run counter; and in response to determining that the run counter is greater than a threshold, generating a run-mode flag and writing the run-mode flag after the particular residual value in a compressed output stream; and in response to determining that the particular residual value is not equal to a corresponding previous value: in response to determining that the run counter is greater than the threshold, generating a run-length after the run-mode flag in the compressed output stream.

Online image compression in hardware

An apparatus, system and method is provided to determine a motion of pixels in local regions of a scene, classify the motion into a speed category, and make decisions on how to render blocks of pixels. In one implementation the motion in a tile is classified into at least three different speed regimes. If the pixels in a tile are in a quasi-static speed regime, a determination is made whether or not to reuse a fraction of pixels from the previous frame. If the pixels are determined to be in a high speed regime, a decision is made whether or not a sampling rate may be reduced.

A system includes a plurality of hardware resources, one or more processors configured to execute one or more programs to perform one or more operations; and a resource allocator configured to implement resource allocation of the plurality of hardware resources to a set of hardware threads. The resource allocation of the plurality of hardware resources is performed by: implementing a linked list comprising a plurality of nodes, wherein each of the plurality of nodes includes respective information regarding at least one of a next node or a previous node; allocating in a first order one or more ranges of free resources of the plurality of hardware resources to one or more nodes of the plurality of nodes; and releasing allocated nodes out-of-order in a second order distinct from the first order.

Flexible in-order and out-of-order resource allocation

There is provided a method of a graphics processing system, the method including receiving dependency information for a set of interdependent images indicating a dependency across one or more compute shader and graphics workloads, and interleaving processing of the compute shader and graphics workloads for the set of interdependent images in accordance with the dependency information without recompiling a compute shader generating the one or more compute shader workloads.

Automated graphics and compute tile interleave

One or more embodiments of the present disclosure provide an apparatus used in source data compression, comprising a memory and a at least one processor. The memory is configured to store vertex attribute data and a set of instructions. The processor is coupled to the memory. The processor is configured to receive a source data stream that includes one or more values corresponding to the vertex attribute data. The processor is also configured to provide a dictionary for the one or more values in the source data stream, wherein the dictionary includes a plurality of index values corresponding to the one or more values in the source data stream. The processor is also configured to lace at least some of the one or more values in the source data stream with corresponding index values of the plurality of index values.

Vertex attribute compression and decompression in hardware

A fixed rate compressor is used to perform variable rate texture compression. A texture image is accessed. A block size used to compress the image is automatically varied over the image to achieve variable rate texture compression. The block size may be selected to reduce the compressed texture image size and adapted in local regions of the texture image based on quality considerations, such as a quality condition that an error for each block be less that a threshold error. The restriction on block size and block types may be selected to perform decompression with hardware conventionally used to perform decompression of fixed-rate blocks. The quality condition may be user-selectable by a user input to provide additional control over the tradeoffs between quality and compression.

Variable-rate texture compression using fixed-rate codes

A method of variable rate compression of image data in an image pipeline of a graphics processing system, the method includes identifying, by a processor of the graphics processing system, a set of cTiles associated with the image data, each cTile including a plurality of pixels, for each cTile of the set of cTiles identifying, by the processor, a pivot pixel from among the plurality of pixels, identifying, by the processor, a compression type of the cTile by comparing, bit-by-bit , pixels within the cTile with the pivot pixel, and compressing, by the processor, the cTile based on the identified compression type, and generating, by the processor, a metadata entry associated with the set of cTiles, the metadata entry indicating the compression type of each one of the set of cTiles and defining a mapping between an uncompressed address space of the set of cTiles and a compressed address space.

Online texture compression and decompression in hardware

A graphics system interleaves a combination of graphics renderer operations and compute shader operations. A set of API calls is analyzed to determine dependencies and identify candidates for interleaving. A compute shader is adapted to have a tiled access pattern. The interleaving is scheduled to reduce a requirement to access an external memory to perform reads and writes of intermediate data.

A graphics system supports variable rate compression and decompression of texture data and color data. An individual block of data is analyzed to determine a compression data type from a plurality of different compression data types having different compression lengths. The compression data types may include a compression data type for a block having a constant (flat) pixel value over n&#xd7;n pixels, compression data type in which a subset of 3 or 4 values represents a plane or gradient, and wavelet or other compression type to represent higher frequency content. Additionally, metadata indexing provides information to map between an uncompressed address to a compressed address. To reduce the storage requirement, the metadata indexing permits two or more duplicate data blocks to reference the same piece of compressed data.

Texture pipeline with online variable rate dictionary compression

In a graphics processing system pixel data and vertex coordinate information from a previous frame is buffered and provided to the current frame. A decision is made in the current frame whether pixel data from the previous frame may be reused. In one implementation if the speed of pixels in a tile is less than a quasi-static speed threshold a decision is made whether or not to reuse a fraction of pixels from the previous frame.

Graphics processing with advection to reconstruct missing sample data points

In a pipelined application having different stages of processing, such as a graphics application or an image processing application, there may be a dependence of one compute kernel upon another. Data associated with individual kernels needs to be written and read. A technique to minimize a need to read and write kernel data to external memory utilize at least one of fusing kernels, resizing workgroups, and performing interleaving of kernels.

Automated compute kernel fusion, resizing, and interleave

Graphics processing is performed in which a decision is made in individual tiles whether or not to sample at a reduced sampling rate. A sampling pattern is selected from a set of sampling patterns having the same reduced sampling rate. The sampling pattern is dithered over a set of frames to reduce the visual appearance of visual artifacts via temporal color averaging.

Growing community of inventors

Mountain View, CA, United States of America

Abhinav Golas