IEEE 754 standard

How would you use the IEEE 754 standard to represent a floating point number if you were given the bit pattern representation of it?
pcombAsked:
Who is Participating?
 
Joe Winograd, EE Fellow 2017, MVE 2016, MVE 2015Connect With a Mentor DeveloperCommented:
The IEEE 754 representation of a number contains three components:

(1) sign bit: if the sign bit is 0, the number is positive; if the sign bit is 1, the number is negative.

(2) significand: a number greater than or equal to 1 and less than 2, often referred to as "1 plus a fraction" (an older term for this component is mantissa)

(3) exponent: a power of 2.

Here's an excellent description of IEEE 754 format, with examples of converting to and from it:
http://www.oxfordmathcenter.com/drupal7/node/43

If you want to learn more about it, take a look at David Goldberg's comprehensive article, What Every Computer Scientist Should Know about Floating-Point Arithmetic.

Also, there was an interesting thread here at EE a few years ago that you may find helpful:
http://www.experts-exchange.com/Programming/Algorithms/Q_25116964.html

Regards, Joe
0
 
Dave BaldwinFixer of ProblemsCommented:
The IEEE 754 standard defines the bit patterns to use for floating point numbers along with methods and other details.  It is used in math coprocessors and in mathematical routines in almost all programming languages.  So I'm not sure what your question is about.  What 'bit pattern representation' are you talking about?
0
All Courses

From novice to tech pro — start learning today.