Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and system for processing floating-point numbers

A technique for floating-point numbers and numbers, applied in the field of processing floating-point numbers

Pending Publication Date: 2022-02-22
IMAGINATION TECH LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Therefore, there are shortcomings in existing methods and architectures for dealing with floating point numbers

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and system for processing floating-point numbers
  • Method and system for processing floating-point numbers
  • Method and system for processing floating-point numbers

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0065] The following description is given by way of example to enable any person skilled in the art to make and use the invention. The invention is not limited to the embodiments described herein, and various modifications to the disclosed embodiments will be apparent to those skilled in the art.

[0066] Embodiments are now described by way of example only.

[0067] When improving computational performance dealing with floating point numbers, such as performing addition or subtraction, the accuracy of the generated output and computational speed are the two main criteria considered.

[0068] Some arrangements of floating point adders that provide faster calculations include arrangements for performing parallel summations. These can also be used to reduce network latency. For example, a specific implementation of a network of floating-point adders for performing parallel sums and designed to reduce latency is shown in figure 2 shown. figure 2 The tree adder 200 in includ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a method and a system for processing floating-point numbers. A method and system for processing a set of 'k' floating-point numbers to perform addition and / or subtraction is disclosed. Each floating-point number includes a mantissa (mi) and an index (ei). The method includes receiving the set of "k" floating-point numbers in a first format, each floating-point number in the first format including a mantissa (mi) having a bit length of "b" bits. The method further comprises generating a set of k'numbers (y) based on the mantissa of the k 'floating-point number, the numbers having a bit length of n'bits obtained by adding additional most significant bits and additional least significant bits to the bit length'b' of the mantissa (mi). The method comprises identifying a maximum exponent (emax) of the exponents ei, aligning the magnitude bits of the number (y) based on the maximum exponent (emax), and simultaneously processing the set of'k 'numbers.

Description

technical field [0001] This application deals with handling floating point numbers. Background technique [0002] Most computing systems use a number format, usually in binary or base 2, for various calculations. These number formats include fixed-point or floating-point formats. Fixed-point number formats can provide additional precision, but are only used to represent a limited range of values. Therefore, most modern computing systems use a floating-point number format that provides a trade-off between range and precision. [0003] A floating point number includes a mantissa (m) with a bit length of 'b' bits, an exponent (e) with a bit length of 'a' bits, and an optional sign bit (s) to represent a binary number. In some widely used formats, the exponent is biased (ie shifted) by a value (c) to represent numbers less than 1 and used to encode outliers at its endpoints. For non-extreme values ​​e, the floating-point number x is said to be normalized, and the number x is...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F7/50G06F7/556
CPCG06F7/50G06F7/556G06F7/485G06F7/509G06F5/012G06F7/49936G06F7/49915
Inventor T·费列雷
Owner IMAGINATION TECH LTD