Re: [Flashforth-devel] Floating point math for FF 5.x - IEEE-754 format - Part II.

[om1zz <om...@vo...>]

Part II.
This is a short intro how the single precision (32bit) floating point numbers work in the lib:

We need to process some data coming from ADC, 9DOF, I2C sensors with our FF system, of course :)

To process integers you may use the s>f and d>f words for conversion to float.

For example, with "precison" set to 7 (6decimal places) for printing with fs. :

16bit signed integers:

100 s>f fs. 1.000000E2  ok<#,ram>
-100 s>f fs. -1.000000E2  ok<#,ram>

30001 s>f fs. 3.000100E4  ok<#,ram>
-30001 s>f fs. -3.000100E4  ok<#,ram>

32bit signed integers:

12000000. d>f fs. 1.200000E7  ok<#,ram>
-12000000. d>f fs. -1.200000E7  ok<#,ram>

12345678. d>f fs. 1.234568E7  ok<#,ram>

10000000. d>f 10000000. d>f f* fs. 1.000000E14  ok<#,ram>

-1000000. d>f 1. d>f f+ fs. -9.999990E5  ok<#,ram>


Happy floating!
I.