# Timeline for Is floating point math broken?

### Current License: CC BY-SA 4.0

21 events
when toggle format what by license comment
Nov 4, 2019 at 13:05 history bounty ended
Apr 9, 2019 at 10:25 history edited
Provide representations in the form a * 2^b where 'a' is integer
May 23, 2017 at 11:55 history edited URL Rewriter Bot
replaced http://stackoverflow.com/ with https://stackoverflow.com/
Mar 31, 2017 at 8:30 history edited
Fixed dead link on Wikipedia, corrected horrific mistake (1 + 12 + 52 != 64), and used a non-mobile wiki URL
Jan 20, 2016 at 9:07 history edited
Explain the rounding method used instead of glossing over it by saying it's rounded away from zero (which in general is not quite true).
Jul 26, 2015 at 8:49 history edited
State how zeros are represented and explicitly state that they are not denormal (instead, they form a separate class along with NaN values, infinities, normals, and denormals)
S Jul 26, 2015 at 8:34 history edited
Make explicit how denormal numbers have an implied zero.
S Jul 26, 2015 at 8:34 history
Make explicit how denormal supports representing zero.
Jul 26, 2015 at 7:59 review
S Jul 26, 2015 at 8:34
Jul 26, 2015 at 5:31 review
Jul 26, 2015 at 6:50
Jul 25, 2015 at 21:33 review
Jul 25, 2015 at 23:59
Apr 17, 2015 at 14:04 history edited
Corrected comment on subnormal numbers. The largest subnormal number is not 2^-1023 (which has an offset exponent of 0 and mantissa 1000...000). Instead, it is 2^(-1023 + 1) - 2^(-1023 - 51), which has an offset exponent of 0 and mantissa 1111...111.
Apr 10, 2015 at 16:15 history edited
expanded on the denormal numbers footnote
Feb 25, 2015 at 20:08 history edited
Use a<sup>b</sup> instead of a^b when not in code markup
Feb 24, 2015 at 5:09 history edited
expanded more on introduction and conclusion, and changed some formatting
Feb 23, 2015 at 20:50 history edited