# Computing: Floating Point Binary Numbers

**URL:** <https://forum.kirupa.com/t/computing-floating-point-binary-numbers/262295>\
**Category:** random\
**Created:** [June 4, 2008, 6:59pm UTC](https://forum.kirupa.com/t/computing-floating-point-binary-numbers/262295 "2008-06-04T18:59:38Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ben\_is\_sparky](https://yyz1.discourse-cdn.com/flex011/user_avatar/forum.kirupa.com/ben_is_sparky/32/664_2.png) [@ben\_is\_sparky](https://forum.kirupa.com/u/ben_is_sparky)\
**Post date:** [June 4, 2008, 6:59pm UTC](https://forum.kirupa.com/t/computing-floating-point-binary-numbers/262295/1 "2008-06-04T18:59:38Z")

</div>

We’re trying to put -11/32 into a floating point binary number with 8 bits for the mantissa and exponent (each).

We get 10101000 11111111

mantissa: 1.0101  
exponent: -1 (dec)

but our textbook gets 10100000 11111111  
which we decode back to get -3/8

mantissa: 1.01  
exp: -1 (dec)

Can anyone confirm why these are different / who is right?  
(we are going mad)

Thanks!

Edit: I should say, we’re using two’s complement for both the mantissa and exponent parts.