摘要:
Disclosed herein is a computer implemented method for performing multiply-add operations of binary numbers P, Q, R, S, B in an arithmetic unit of a processor, the operation calculating a result as an accumulated sum, which equals to B+n×P×Q+m×R×S, where n and m are natural numbers. Further disclosed herein is an arithmetic unit configured to implement multiply-add operations of binary numbers P, Q, R, S, B comprising at least a first binary arithmetic unit for calculating an aligned high part result and a second binary arithmetic unit for calculating an aligned low part result of the multiply-add operations.
摘要:
A system is provided for providing a product and a location from a code selected from the group consisting of a product code index, a location code index, a location code value, and a product code value. The system comprises: a processor, an input coupled to the processor and configured to receive the code, a first converter operable to convert a product code index to a product code value; a second converter operable to convert a product code value to a location code value; a third converter operable to convert a location code index to a location code value; a fourth converter operable to convert a location code value to a product code value; whereby, a product and a location are provided.
摘要:
Circuitry accepting a first input value and a second input value, and outputting (a) a first sum involving the first input value and the second input value, and (b) a second sum involving the first input value and the second input value, includes a first adder circuit, a second adder circuit, a compressor circuit and a preprocessing stage. The first input value and the second input value are input to the first adder circuit to provide the first sum. The first input value and the second input value are input to the preprocessing stage to provide inputs to the compressor circuit, which provides first and second compressed output signals which in turn are input to the second adder circuit to provide the second sum. The preprocessing stage may include circuitry to programmably zero the first input value, so that the first sum is programmably settable to the second input value.
摘要:
Electronic logic gates that operate using N logic state levels, where N is greater than 2, and methods of operating such gates. The electronic logic gates operate according to truth tables. At least two input signals each having a logic state that can range over more than two logic states are provided to the logic gates. The logic gates each provide an output signal that can have one of N logic states. Examples of gates described include NAND/NAND gates having two inputs A and B and NAND/NAND gates having three inputs A, B, and C, where A, B and C can take any of four logic states. Systems using such gates are described, and their operation illustrated. Optical logic gates that operate using N logic state levels are also described.
摘要:
Hardware and processes are provided for efficient interpretation of multi-value signals. The multi-value signals have a first voltage range with is used to indicate multiple numerical or logical values, and a second voltage range that is used to provide control functions. In one example, the multi-value circuitry is arranged as a set of rows and columns, which may be cascaded together. The control function can be implemented to cause portions of rows, columns, or cascaded connections to be powered off, thereby saving power and enabling more efficient operation.
摘要:
Hardware and processes are provided for efficient interpretation of multi-value signals. The multi-value signals have a first voltage range with is used to indicate multiple numerical or logical values, and a second voltage range that is used to provide control functions. In one example, the multi-value circuitry is arranged as a set of rows and columns, which may be cascaded together. The control function can be implemented to cause portions of rows, columns, or cascaded connections to be powered off, thereby saving power and enabling more efficient operation.
摘要:
N-valued re-circulating latches using n-valued reversible inverters with n>3 are disclosed. Latches using n-valued self-reversing inverters are provided; latches using n-valued universal inverters are provided; and latches using inverters which are not self-reversing or universal are also provided. A latch may use two individually controlled gates. It may also use one individually controlled gate. N-valued latches are provided wherein a state is represented by a signal being an independent instance of a physical phenomenon. A latch not using absence-of-signal as a state is also provided.
摘要:
A methodology for describing an input-output behavior of a multi-level logic gate to process simultaneously a multiplicity of independent Boolean logic functions, with each Boolean function processing signals carried on an individual, separate channel. An embodiment may simultaneously process the same data with the same function or with different functions, multiple data with the same function, or multiple data with different functions. In addition, multi-level logic signals (having more than two levels) may be processed, so that a higher communication bandwidth may be obtained without necessarily increasing the number of traces (wires). Other embodiments are described and claimed.
摘要:
Apparatus and method for performing multi-value arithmetic operations are disclosed. Multi-value signals can be added, subtracted and multiplied using a first truth table to generate a residue and a second truth table to generate a carry. Additionally, method and apparatus to efficiently perform the function a0b1+a1b0 on multi-value signals are disclosed. Also an efficient method of processing large binary signals is disclosed.
摘要:
Apparatus and method for performing multi-value arithmetic operations are disclosed. Multi-value signals can be added, subtracted and multiplied using a first truth table to generate a residue and a second truth table to generate a carry. Additionally, method and apparatus to efficiently perform the function a0b1+a1b0 on multi-value signals are disclosed. Also an efficient method of processing large binary signals is disclosed.