Floating Point output.

most languages limit the amount of precision when converting genuine IEEE floats to humanly readable decimal output.
DePbo.10.05.0.23.Installer.exe
the default precision for most C-family compilers is 6 decimal places.

This, is not sufficient for *some* values and the precision must be extended to greater detail to get an EXACT match when re-binarazing previously decoded output.

The dll needs to iteratively check that the resulting text "0.00123" will, in fact, produce the same 4 byte IEEE float when and if re-compiled.

This level of exactness is rarely necessary, yielding some rather ugly expressions like 0.00009888887 when 0.00001 would be more 'readable'.

However, to get exact binary re-compiled equivalents, and, as a confidence check, this extension to the standard compiler implementation has been made.

Some immediate side effects to this 'exactness' will be apparent.

Firstly, the process can be slow when large amounts of float data is present in the to-be-decoded binary. Tools using this revision will be slower than before. Not by much, but slower.


Secondly, comparison output of previously decoded config.bin's (eg) *might* no longer be the same. *SOME*, float values, for *SOME* configs will be longer than previously listed.


Third, warning messages are sent to stderr if the value is a nan or infinity

