[Agama]

I have a 4MB plain text file which I have marked-down in preparation for calibre conversion to ePub. The markdown produces a 3 level TOC and doesn't require any structure detection, (so I set it to null in the calibre conversion options).

Conversion works perfectly and exploding the ePub shows a set of html files varying in size from 60K to 127K. As an experiment I switched on structure detection for H3 elements, (chapters), so that the output was now many, many more html files each of size 4K to 8K.

So my query is this: Is it better to have many small html files within the ePub or a fewer number of larger ones? Does it improve reader compatibility/presentation to break a book into separate html files per chapter or does it make no difference?

(Does calibre's conversion processing prefer one or the other?)

I'm only asking as I intend to post the book to the eBooks section of MR and want to make the book widely compatible. (I have had purchased books which caused my reader to hang).