PDFs aren't going to get better, and that's not a fault of ebook readers or Calibre.
PDF isn't so much a format as a wrapper for several different formats, intended to allow printouts that look like the original. It was created to get around Postscript driver problems & compatibility issues. It's a terrific archive format; it's a lousy use format unless it was designed for the screen it's going to be read on. (Doctorow's For The Win PDF, for 6" ebook readers.)
How well it can be converted depends on how it was made; Calibre can't do anything useful with PDFs that are just scans. And the ones made with layout programs like InDesign and XPress convert badly if a lot of tweaking went into the layout--if the header-with-shading involves a large-font word, and the same word again in a light color behind it, and the word *again* with a picture wrapped around the letters, what you see in the conversion is three copies of the word. Maybe all together; maybe mixed in with the other text, depending on what order it was done in.
Conversion is always going to preserve line breaks at the end of a column or page, because PDF doesn't have a way of recognizing "this sentence continues." Sometimes it puts header/footer bits at the end of every page; if it doesn't, that depends on the program the PDF was created/converted with. And so on.
Lots of serious ebook people scream about PDF not being an ebook format. I convert PDFs to Word, reformat them, and then throw them back into useful formats. But I'm a hardcore PDF geek.
Re: nice icon!
PDF isn't so much a format as a wrapper for several different formats, intended to allow printouts that look like the original. It was created to get around Postscript driver problems & compatibility issues. It's a terrific archive format; it's a lousy use format unless it was designed for the screen it's going to be read on. (Doctorow's For The Win PDF, for 6" ebook readers.)
How well it can be converted depends on how it was made; Calibre can't do anything useful with PDFs that are just scans. And the ones made with layout programs like InDesign and XPress convert badly if a lot of tweaking went into the layout--if the header-with-shading involves a large-font word, and the same word again in a light color behind it, and the word *again* with a picture wrapped around the letters, what you see in the conversion is three copies of the word. Maybe all together; maybe mixed in with the other text, depending on what order it was done in.
Conversion is always going to preserve line breaks at the end of a column or page, because PDF doesn't have a way of recognizing "this sentence continues." Sometimes it puts header/footer bits at the end of every page; if it doesn't, that depends on the program the PDF was created/converted with. And so on.
Lots of serious ebook people scream about PDF not being an ebook format. I convert PDFs to Word, reformat them, and then throw them back into useful formats. But I'm a hardcore PDF geek.