Abstract:
A circuit and method are disclosed for relatively rapidly causing the current flowing through a write head to transition between steady states without generating an appreciable amount of capacitively-coupled noise. Embodiments of the present invention generally provide drive voltage signals to the write head that have no common mode voltage levels during transitions between steady state current levels in the write head. In other words, the drive voltage signals applied to the write head are substantially entirely differential during write head current transitions. In an exemplary embodiment of the present invention, a driver circuit (100) includes switching circuitry (111,112,113,114,116,117,118,119) connected between the terminals (42A,42B) of the write head (42) and reference voltage supplies, such as positive and negative voltage supplies. The driver circuit further includes timing circuitry (125) that generates control signals for controlling the switching circuitry.
Abstract:
A random access memory array includes random access memory elements arranged in a rows and columns. The elements of each row have a word line and a write digit line and the elements of each column have a bit line and a write bit line. A first selection circuit/transistor for each row has a first source-drain path coupled in the write digit line and a gate terminal coupled to the word line. A second selection circuit/transistor for each column has a second source-drain path coupling in the write bit line and a gate terminal coupled to the bit line. A first write signal is applied to one word line to actuate the first selection circuit/transistor for the row corresponding to that one word line and cause a write current to flow through the first source-drain path of the actuated first selection circuit/transistor and the corresponding write digit line to write data into certain memory elements in that row. A second write signal is applied to one bit line to actuate the second selection circuit/transistor for the column corresponding to that one bit line and cause a write current to flow through the second source-drain path of the actuated second selection circuit/transistor and the corresponding write bit line to write data into at least one memory element in that column.
Abstract:
A network-on-chip interconnects an array of integrated circuit resources. The network-on-chip includes at least one vertical communications ring per column of the array and at least one horizontal communications ring per row of the array. A network interface is associated:with each resource of the array and operates to interface the communications rings with each other and the resource with the communications rings. A ring hop is provided at each network interface and for each communications ring thereat. Each ring hop functions as an add/drop multiplexer with respect to inserting packets onto the associated communications ring and extracting packets from the associated communications ring. Packets are communicated over the vertical/horizontal rings using a logical transport channel that flows in a cyclic manner through the communications ring without interruption. A back pressure mechanism allows the ring hops on a given communications ring to address overflow conditions on that ring, and a request mechanism allows the ring hops on a given communications ring to request use of that ring to carry a packet communication.
Abstract:
A new technique for Hard Disk Drive (HDD) servo-burst demodulation is provided. A 4-samples per dibit Discrete Fourier Transform (DFT) amplitude estimation is used to calculate the read-head servo-position error signal. Comparatively, the conventional method of burst demodulation ― called burst integration ― typically uses more than 8 samples/dibit. Consequently, the new 4-samples/dibit DFT burst-demodulation scheme requires fewer samples per dibit than does burst integration, thus reducing the disk space occupied by the burst data while increasing the performance as compared to burst integration. Furthermore, the DFT scheme does not require the samples to be synchronized to any particular points of the servo burst, and can include an averaging algorithm that further improves performance for a given Signal to Noise Ratio (SNR). Moreover, the same sample-clocking circuit that detects the Gray Code servo information can also implement the DFT burst-demodulation scheme to demodulate the servo burst.
Abstract:
A priority encoder can be used for a Content-Addressable Memory (CAM) device that typically has an array of CAM cells arranged in columns and rows with each row having a match signal indicative that compare data has matched data within the respective row. A priority encoder is operatively connected to the array of CAM cells and determines a highest priority matching address for data within the array of CAM cells. The priority encoder includes match lines associated with respective rows and precharged bus lines connected into respective match lines that are discharged whenever there is a match signal such that the highest precharged bus line discharged results in an encoded address.
Abstract:
A method and device are disclosed for detecting successful transfers between a Universal Serial Bus (USB) port and a USB smart card and generating a signal that provides an indication of the USB transaction activity. This USB transaction activity signal is modulated according to the USB transaction activity and drives a Light Emitting Diode (LED) in a preferred embodiment of the invention. A counter internal to the USB smart card scales the transaction activity signal such that it is perceptible to the user. Because the current through the LED depends upon the USB transaction activity, the brightness of the LED varies according to the USB transaction activity. The LED may be driven from a current mirror sink or source, or a current switch sink or source.
Abstract:
A digital video decoder comprising a bit unpacker that utilizes at least an encoded video data stream to produce a prediction data stream and a coding data stream; a motion compensation instruction encoder communicably coupled to the bit unpacker, the motion compensation instruction encoder utilizing at least the prediction data stream to selectively produce a set of motion compensation instructions; and a block decoder communicably coupled to the bit unpacker, the block decoder utilizing at least the coding data stream to produce a set of error terms.
Abstract:
A processor executes one or more prefetch threads and one or more main computing threads. Each prefetch thread executes instructions ahead of a main computing thread to retrieve data for the main computing thread, such as data that the main computing thread may use in the immediate future. Data is retrieved for the prefetch thread and stored in a memory, such as data fetched from an external memory and stored in a buffer. A prefetch controller determines whether the memory is full. If the memory is full, a cache controller stalls at least one prefetch thread. The stall may continue until at least some of the data is.transferred from the memory to a cache for use by at least one main computing thread. The stalled prefetch thread or threads are then reactivated.