Abstract: Disclosed herein are systems and methods for methods of developing a database of controllable objects in an environment. For example, a method includes a mobile device having a camera to capture images of objects in an environment. For each object, the method includes, in response to receiving a user selection of the object, training a machine-learning model to recognize the object. The method includes receiving a command associated with the object and receiving a plurality of images of the object and training the machine-learning model to recognize the object based on the plurality of images. The method further includes transmitting the trained model and the command to a wearable electronic device causing the wearable electronic device to save the trained machine-learning model to a data store and to associate the command with the trained machine-learning model.

Abstract: Methods and systems for verifying a hardware design for a component that implements a permutation respecting function. The methods include formally verifying that an instantiation of the hardware design produces the correct result to the function for a subset of the valid input vectors; and formally verifying that an instantiation of the hardware design is permutation respecting for each of one or more input vector permutations (i.e. that the instantiation of the hardware design produces the permutation related outputs for an input vector and the permutation of that input vector) over a set of input vectors. The subset and the input vector permutations are selected so that any valid input vector that is not in the subset can be generated from an input vector in the subset via a combination of the one or more input vector permutations.

Abstract: 3-D rendering systems include a rasterization section that can fetch untransformed geometry, transform geometry and cache data for transformed geometry in a memory. As an example, the rasterization section can transform the geometry into screen space. The geometry can include one or more of static geometry and dynamic geometry. The rasterization section can query the cache for presence of data pertaining to a specific element or elements of geometry, and use that data from the cache, if present, and otherwise perform the transformation again, for actions such as hidden surface removal. The rasterization section can receive, from a geometry processing section, tiled geometry lists and perform the hidden surface removal for pixels within respective tiles to which those lists pertain.

Abstract: Methods and compression units for compressing a block of image data, the block of image data comprising a plurality of image element values, the image element values being divisible into at least a first value and a second value such that the block of image data comprises a two-dimensional block of first values, the method comprising: compressing a first data set comprising all or a portion of the two-dimensional block of first values in accordance with a first fixed-length compression algorithm to generate a first compressed block by: identifying common base information for the first data set; and identifying a fixed-length parameter for each first value in the first data set, the fixed-length parameter being zero, one or more than one bits in length; and forming a compressed block for the block of image data based on the first compressed block.

Abstract: Methods and decompression units for decompressing data from a compressed block of image data, the compressed block of image data representing a block of image data comprising a plurality of image element values, the image element values being divisible into at least a first value and a second value such that the block of image data comprises a two-dimensional block of first values.

Abstract: The graphics processing unit described herein is configured to process graphics data using a rendering space which is sub-divided into a plurality of tiles. The graphics processing unit comprises a tiling unit and rendering logic. The tiling unit is arranged to generate a tile control list for each tile, the tile control list identifying each graphics data item present in the tile. The rendering logic is arranged to render the tiles using the tile control lists generated by the tiling unit. The tiling unit comprises per-tile hash generation logic arranged to generate, for each tile, a per-tile hash value based on a set of textures that will be accessed when processing the tile in the rendering logic, and the tiling unit is further arranged to store the per-tile hash value for a tile within the tile control list for the tile.

Abstract: A communications interface for interfacing between a host system and a state machine includes an event slot, the event slot comprising a plurality of registers including: a write register for writing by the host system, and a read register for reading by the host system, wherein the event slot is addressed from the host system by a single address location permitting the host system to write data to the write register and/or read data from the read register; and wherein the write register and the read register are individually addressable by the state machine.

Abstract: A method for a plurality of pipelines, each having a processing element having first and second inputs and first and second lines, wherein at least one of the pipelines includes first and second logic operable to select a respective line so that data is received at the first and second inputs respectively. A first mode is selected and for the at least one pipeline, the first and second lines of that pipeline are selected such that the processing element of that pipeline receives data via the first and second lines of that pipeline, the first line being capable of supplying data that is different to the second line. A second mode is selected and for the at least one pipeline a line of another pipeline is selected, the second line of the at least one pipeline is selected and the same data at the second line is supplied as the first line.

Abstract: Methods and arbiters for arbitrating between a plurality of ordered requestors and a shared resource based on priorities allocated to the requestors. The methods include determining whether there is at least one requestor that has requested access in the current cycle and has priority in the current cycle; in response to determining that there is at least one requestor that has requested access in the current cycle and has priority in the current cycle, selecting a lowest ordered requestor that has requested access in the current cycle and has priority in the current cycle; and in response to determining that there are no requestors that have requested access in the current cycle and have priority in the current cycle, selecting a highest ordered requestor that has requested access in the current cycle.

Abstract: Methods of memory allocation map registers referenced by different groups of instances of the same task to individual logical memories. Other example methods describe the mapping of registers referenced by a task to different banks within a single logical memory and in various examples this mapping may take into consideration which bank is likely to be the dominant bank for the particular task and the allocation for one or more other tasks.

Abstract: A function estimation hardware logic unit may be implemented as part of an execution pipeline in a processor. The function estimation hardware logic unit is arranged to calculate, in hardware logic, an improved estimate of a function of an input value, d, where the function is given by 1 ? / ? d i . The hardware logic comprises a plurality of multipliers and adders arranged to implement a mth-order polynomial with coefficients that are rational numbers, where m is not equal to two and in various examples m is not equal to a power of two. In various examples i=1, i=2 or i=3. In various examples m=3.

Abstract: A method and compression unit for compressing a block of image data to satisfy a target level of compression, wherein the block of image data comprises a plurality of image element values, each image element value comprising one or more data values relating to a respective channel. For each of the channels: (i) an origin value for the channel for the block is determined, (ii) difference values are determined representing differences between the data values and the determined origin value for the channel for the block, and (iii) a first number of bits for losslessly representing a maximum difference value of the difference values for the channel for the block is determined.

Abstract: Input data for a layer of a convolutional neural network (CNN) is provided by receiving input data values to be processed in a layer of the CNN. Addresses in banked memory of a buffer are determined in which the received data values are to be stored based upon format data indicating a format parameter of the input data in the layer and indicating a format parameter of a filter which is to be used to process the input data in the layer. The received input data values are stored at the determined addresses in the buffer for retrieval for processing in the layer.

Abstract: An integrated circuit includes a memory configured to store a plurality of functions; a mapping interface configured to perform a mapping from a received first signal to a first function of the plurality of functions; and a state machine configured to, in response to said mapping, execute the first function; wherein the integrated circuit is arranged to, in dependence on the execution of the first function at the state machine, modify said mapping between the first signal and the first function so as to re-map the first signal to a second function of the plurality of functions such that, on receiving a subsequent first signal, the state machine is configured to execute the second function.

Abstract: Apparatus includes hardware logic arranged to normalise an n-bit input number. The hardware logic comprises at least a first hardware logic stage, an intermediate hardware logic stage and a final hardware logic stage. Each stage comprises a left shifting logic element, the first and intermediate stages each also comprise a plurality of OR-reduction logic elements and the intermediate and final stages each also comprise one or more multiplexers. The OR-reduction logic elements operate on different subsets of bits from the number input to the particular stage. In the intermediate and final hardware logic stages, a first of the multiplexers selects an OR-reduction result received from a previous hardware logic stage and the left shifting logic element is arranged to perform left shifting on the updated binary number received from an immediately previous hardware logic stage dependent upon the selected OR-reduction result.

Abstract: A method of processing primitives within a tiling unit of a graphics processing system is described. The method comprises determining whether a primitive falls within a tile based on positions of samples within each pixel. If it is determined that the primitive does fall within a tile based on the positions of samples within pixels in a tile, an association between the tile and the primitive is stored to indicate that the primitive is present in the tile. For example, an identifier for the primitive may be added to a control stream for the tile to indicate that the primitive is present in the tile. Various different methods are described to make the determination and these may be used separately or in any combination.

Abstract: A graphics processing system for operation with a data store, comprising: one or more processing units for processing tasks; a check unit operable to form a signature which is characteristic of an output from processing a task on a processing unit; and a fault detection unit operable to compare signatures formed at the check unit; wherein the graphics processing system is operable to process each task first and second times at the one or more processing units so as to, respectively, generate first and second processed outputs, the graphics processing system being configured to: write out the first processed output to the data store; read back the first processed output from the data store and form at the check unit a first signature which is characteristic of the first processed output as read back from the data store; form at the check unit a second signature which is characteristic of the second processed output; compare the first and second signatures at the fault detection unit; and raise a fault signal i

Abstract: A binary logic circuit for determining y=x mod(2m?1), where x is an n-bit integer, y is an m-bit integer, and n>m, includes reduction logic configured to reduce x to a sum of a first m-bit integer ? and a second m-bit integer ?; and addition logic configured to calculate an addition output represented by the m least significant bits of the following sum right-shifted by m: a first binary value of length 2m, the m most significant bits and the m least significant bits each being the string of bit values represented by ?; a second binary value of length 2m, the m most significant bits and the m least significant bits each being the string of bit values represented by ?; and the binary value 1.

Abstract: A hierarchical acceleration structure is generated for intersection testing in a ray tracing system. Nodes of the hierarchical acceleration structure are determined, wherein each of the nodes represents a region in a scene, and wherein the nodes are linked to form the hierarchical acceleration structure. Data is stored representing the hierarchical acceleration structure. The stored data defines the regions represented by a plurality of the nodes of the hierarchical acceleration structure. At least one node is an implicitly represented node, wherein data defining a region represented by an implicitly represented node is not explicitly included as part of the stored data but can be inferred from the stored data.

Abstract: Methods for determining a fixed point format for one or more layers of a DNN based on the portion of the output error of the DNN attributed to the fixed point formats of the different layers. Specifically, in the methods described herein the output error of a DNN attributable to the quantisation of the weights or input data values of each layer is determined using a Taylor approximation and the fixed point number format of one or more layers is adjusted based on the attribution. For example, where the fixed point number formats used by a DNN comprises an exponent and a mantissa bit length, the mantissa bit length of the layer allocated the lowest portion of the output error may be reduced, or the mantissa bit length of the layer allocated the highest portion of the output error may be increased. Such a method may be iteratively repeated to determine an optimum set of fixed point number formats for the layers of a DNN.