Calibre is "not supposed" to do that.
I'm not the right person to respond to this but until someone who knows what they are doing does, here are some suggestions:
1. For epub, you would expect it to keep the underlying html and css information nearly intact.
Are there particular tags that disappeared?
(You can unzip the the epub and take a look at how it implemented the missing formatting, and then see what how calibre reflected that.)
2. Once you can find what didn't show up in the conversion, you can look to see whether the source complies with the HTML standard for that. Although the developers have made some effort to compensate for common errors, misformed HTML is the most likely reason for the conversion to come out wrong.
Good luck.
|