[Tutor] Associate decimal values with an integer range

[Steven D'Aprano]

On 30/04/13 08:00, Dave Angel wrote:

> One byte floating point?  Nonsense.  While it can be defined (see eryksun's response, which tracks the IEEE standard quite well), it wouldn't store enough precision to be used for anything but a toy.

Minifloats have their uses. Probably the best use is teaching.

http://en.wikipedia.org/wiki/Minifloat

Oh, by the way, my earlier reply missed the point, since it described a one-byte FIXED point number, not floating point. Sorry about that.



-- 
Steven