Unless mIRC starts changing ASCII codes for specific characters, which would not be a good idea and I doubt will happen, the only thing mIRC could do would be to use a $chr for the underline control code. The problem with that is that mIRC would have to somehow support all old scripts that use the original $chr as well and that could prove to be a problem.
I think the best way is to try and find some method of removing the Japanese language code page from Vista. Perhaps some IT forum would have information on doing that. Vista's supposed to be secure, but there should be a way to get around it and remove that code page, even if it's changing the registry by hand to do so.
