A field-programmable gate array (FPGA) is an integrated circuit designed to be configured by a customer or a designer after manufacturing

Welcome! Bienvenidos! английский Willkommen! Benvenuti! الإنجليزية Welkom! ਅੰਗਰੇਜ਼ੀ Boas-Vindas! Hoşgeldiniz! Wellkom! 英语
- Join
- Professional
  - Engineering
  - FPGA
  - Resume
- Contact
- Sports
  - Triathlon
  - Training Log
- Gallery
- Calendar
- Türkçe
- İletişim
- Misc.
- Site Map

Fixed Point, Floating point, Integer Fixed Point : Sign.Integer .Fraction
Floating Point: Sign.Exponent.Integer.Mantissa(Fraction)

Normalized Floating Point: Sign.Exponent.Mantissa(Fraction)

Normalized Floating Point saves 1 bit as the integer part is not represented as integer part of the number is normalized to 0 by shifting and implied to be 0.
Integer is a fixed point number where the number of fraction bits is 0.

IEEE754 Floating point standard

Single Precision: 1bit sign-8bit unsigned exponent - 23bit unsigned fraction, exponent bias:127 (=2^0)
Double Precision: 1bit sign-11bit exponent-52bit fraction, exponent bias:1023 (=2^0)

Fixed point binary to floating point Conversion:

Find the left most bit that has the value of '1'.
Mantissa is the digits on the right of the first 1 from the left.
Exp is the number of bits between the decimal point and the first '1' from the left.
If it is an integer (no fraction bits) then exp is the number of bits to the right of the first 1 from the left.Subtract this from the bias. the left most bit is not used but inferred.

Floating point binary to Fixed point Conversion:

The mantissa is shifted accordingly until the exponent becomes 0.

Example:

Fixed Point : 573.48
Floating Point: 573.48 10⁰

Normalized IEEE754 Single Precision Floating Point: 0.57348 10³ 0_10000010_mantissa(binary)

Normalized IEEE754 Double Precision Floating Point: 0.57348 10³ 010000000010_mantissa(binary)

Integer: 573 (rounded)