Via Alliance Semiconductor Co., Ltd.

Glenfly Technology Co., Ltd.

Via Technologies, Inc.

Shanghai Jiaotong University

Shanghai Zhaoxin Semiconductor Co., Ltd.

Glenfly Tech Co., Ltd.

Shanghai Normal University

Shanghai Xiyuan Biotechnology Co., Ltd.

Yuanfeng Wang

Fengxia Wu

Zhou Hong

Heng Que

Deming Gu

Wei Zhang

Liang Li

John Wakefield Brothers

Boris Prokopenko

Wen-Chung Chen

Tian Shen

Jiangming Xu

Wei Wang

Xin Ye

Ruiyang Chen

Guixiang He

Yongtao Zhao

Wenlin Hao

Kang Wei

Kuan Li

Xinlin Wei

Jinmin Zhang

Lizeng Cheng

Quanquan Xu

A neural network computing device and a cache management method thereof are provided. The neural network computing device includes a computing circuit, a cache circuit and a main memory. The computing circuit performs a neural network calculation including a first layer calculation and a second layer calculation. After the computing circuit completes the first layer calculation and generates a first calculation result required for the second layer calculation, the cache circuit retains the first calculation result in the cache circuit until the second layer calculation is completed. After the second layer calculation is completed, the cache circuit invalidates the first calculation result retained in the cache circuit to prevent the first calculation result from being written into the main memory.

Neural network computing device and cache management method thereof

The present disclosure provides a functional beverage of dark tea with zero energy for reducing weight and lipid and a preparation method thereof, and relates to the technical field of processing health-promoting beverage. The functional beverage includes 1.5-3.5 parts of dark tea, 0.3-0.7 parts of mulberry leaf, 0.1-0.5 parts of lotus leaf, 0.1-0.3 parts of stevia leaf, and 95.4-97.4 parts of purified water. The preparation process thereof includes treatment of raw materials, circulatory microwave extraction, static settlement, membrane ultrafiltration, and ultra-high temperature instantaneous sterilization (UHT). Compared with the prior art, the functional beverage uses purely natural herb plants without any additives. The obtained product has zero sugar while has the effects of quenching thirst and reducing weight and lipid. Key technologies such as integrated formula, circulatory microwave extraction, and membrane ultrafiltration enable the best flavor and efficacy of the beverage.

Preparation method for functional beverage of dark tea with zero energy for reducing weight and lipid

A graphics processing unit includes a pixel shader, an output merger, a cache, and a memory. The pixel shader is configured to output a pixel data. The output merger is coupled to the pixel shader and configured to receive the pixel data. The output merger outputs the pixel data and a sample mask corresponding to the pixel data. The cache is coupled to the output merger and configured to receive the pixel data and the sample mask. The cache generates a sample data according to the pixel data and the sample mask. The memory is coupled to the cache. The cache writes the sample data into the memory. A data size of the sample data is a multiple of a data size of the pixel data. An operation method thereof is also provided.

Graphics processing unit having pixel shader, output merger, cache, memory and operation method thereof

A data transferring apparatus and a method for transferring data with overlap are provided. The data transferring apparatus includes a command splitter circuit and a plurality of tile processing circuits. The command splitter circuit splits a block level transfer command into a plurality of tile transfer tasks. The command splitter circuit may issue the tile transfer tasks to the tile processing circuits in a plurality of batches. The tile processing circuits may execute the tile transfer tasks in a current batch, so as to read data of a plurality of corresponding tiles among a plurality of source tiles of a source block to the tile processing circuits. After all the tile transfer tasks in the current batch have been executed by the tile processing circuits, the command splitter circuit issues the tile transfer tasks in a next batch of the batches to the tile processing circuits.

Data transferring apparatus and method for transferring data with overlap

The disclosure provides an image processing method and an image processing device. The method includes: obtaining a specific block from a specific block group of a specific image, wherein the specific block group includes blocks; compressing the specific block as a specific stream based on a predetermined compression ratio; when a length of the specific stream is not shorter than a predetermined block length, dividing the specific stream into a first sub-stream and a second sub-stream; storing the first sub-stream into a specific private block of a plurality of private blocks; finding an idle chunk out of shared chunks shared by the blocks; if the idle chunk is enough for storing the second sub-stream, storing the second sub-stream into the idle chunk and using a chunk usage information to record a number of the idle chunk.

Image processing method and image processing device

An image processing method and an image processing device are provided. The method includes: retrieving a first and second blocks and compressing the first and second blocks into a first and second streams; in response to determining performing a dynamic memory allocation mechanism to the first and second blocks, respectively allocating a first and second private storage spaces to the first and second blocks, and allocating a shared storage space shared by the first and second blocks and divided into multiple segments; when the first private storage space is insufficient to store the first stream, using the first private storage space and a first part of the segments to cooperatively store the first stream; when the second private storage space is insufficient to store the second stream, using the second private storage space and a second part of the segments to cooperatively store the second stream.

An apparatus for enqueuing kernels on a device-side is introduced to incorporate with at least a MXU (Memory Access Unit) and a CSP (Command Stream Processor): The CSP, after receiving a first command from the MXU, executes commands of a ring buffer, thereby enabling an EU (Execution Unit) to direct the MXU to allocate space of the ring buffer for a first hardware thread and subsequently write second commands of the first hardware thread into the allocated space of the ring buffer according to an instruction of a kernel.

Apparatuses for enqueuing kernels on a device-side

An image compression method based on JPEG-LS is presented. In the method, the M&#xd7;N pixels in the source image are divided into k groups. M, N, and k are all integers larger than one. Each group corresponds to a plurality of pixels among the M&#xd7;N pixels. The decorrelation procedure and the context modeling procedure are performed for each of the plurality of pixels in the igroup of the k groups. The compensation look-up table is not refreshed until all pixels in the igroup are performed with the decorrelation procedure and the context modeling procedure.

Image compressing method based on JPEG-LS

An electronic device, for integration with functional circuit modules, includes gates, monitor module, signal control module and record module. The functional modules are operated on clock signal for generating request instruction and response signal. The gate is coupled to the functional modules for transmitting request instruction and response signal to functional module on enable signals. The monitor module is coupled to the functional modules and the gates for generating hold signal. The monitor module generates enable signals on finish signal. The clock signal control module coupled to the functional modules and the monitor module for outputs main clock signal to generate clock signals. The clock signal control module generates record instruction and stop clock signals, and the clock signal control module re-outputs clock signals on finish signal. The record module coupled to the functional modules and the clock signal control module begins to record request instruction and response signal when receiving record instruction.

Electronic device with recording functionality and method for recording thereof

A graphics processing unit and associated graphics processing method are provided. The graphics processing unit includes: an execution unit, for performing shader execution and texture loading; a fixed-function unit, for executing a graphics rendering pipeline; a memory-access unit; a texture unit, for reading texture data from a memory via the memory-access unit according to the data requirement of the execution unit or the fixed-function unit; and a command stream parser, for receiving a draw command from a display driver, and transmitting the draw command to the execution unit or the fixed-function unit to perform graphics processing according to the type of draw command. When the command stream parser determines that the draw command is a specific draw command, the command stream parser transmits the draw command only to the fixed-function unit to perform graphics processing, and turns off power to the execution unit.

Low-power graphics processing using fixed-function unit in graphics processing unit

A device for image processing includes a first queue, a second queue, a cache, and a processor. The first queue is capable of receiving a first image tile. The processor is electrically connected to the first queue, the second queue, and the cache, respectively. The processor is capable of obtaining the first image tile from the first queue and obtaining mask information of the background mask corresponding to the first tile from the cache. The processor determines the relationship between the first image tile and the background mask based on the first image tile and the mask information so as to selectively transfer the first image tile to the second queue.

Method and device for image processing

The computer graphics system is configured to improve the performance of a stencil shadow volume method for rendering shadows. The apparatus and methods utilize a combination of compressed and uncompressed stencil buffers in coordination with compressed and uncompressed depth data buffers. An uncompressed stencil buffer is capable of storing stencil shadow volume data for each pixel and a compressed stencil buffer is capable of storing stencil shadow volume data for a group of pixels. The compressed stencil buffer is utilized with a compressed stencil buffer cache to perform a stencil shadow volume operation more efficiently than present methods.

Apparatus and method of an improved stencil shadow volume operation

The computer graphics system is configured to generate a shadow effect with a stencil shadow volume method using a combination of compressed and uncompressed stencil buffers in coordination with compressed and uncompressed depth data buffers. An uncompressed stencil buffer is capable of storing stencil shadow volume data for each pixel and a compressed stencil buffer is capable of storing stencil shadow volume data for a group of pixels.

Growing community of inventors

Shanghai, China

Yuanfeng Wang