This would be the possibility to bypass the utf-8 conversion using something like an option for /raw. This would allow to do some homemade codepage conversions(which can sometimes be useful).
Currently if utf-8 is enabled every single string that gets inputed gets converted to utf-8.
EDIT:
https://forums.mirc.com/ubbthreads.php?ub...true#Post144139 here is a reference to a similar post I made long ago.