Similar Posts

Subscribe
Notify of
4 Answers
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Destranix
1 year ago

This is the format in which floating point numbers are usually stored.

This allows a large value range, but only with limited accuracy.

See also:

https://de.wikipedia.org/wiki/IEEE_754

To illustrate the distribution (this is logarithmic):

comment image

KarlRanseierIII
1 year ago
Reply to  Destranix

It is even more important that the accuracy is not homogeneously distributed over the value range.

Destranix
1 year ago

Jein. This is not important for understanding now, but it is already, yes.
Then you may want to explain what operations (addition/subtraction) tend to reduce the accuracy.

I’ll add a graphic.

KarlRanseierIII
1 year ago

Look at you:

https://docs.python.org/3/library/decimal.html

on.

There are good reasons why you don’t use IEEE flats for a calculator in case of doubt.