Format of book on Kindle is terrible

[ChipAHoy]

On Calibre, I have a book that is in MOBI, PDF and AZW3. I transferred the ebook to my Kindle Paperweight reader. The format is a mess, sometimes there is text just on the left side of the screen in a narrow column, there are page headings in the middle of text, the spacing is all over, etc.

I tried viewing both MOBI and AZW3, but it seems like there is no difference.

This is an ebook I did not obtain from Amazon. Do these kind of ebooks require different handling or conversion steps than a book obtained from Amazon?

Any help is appreciated, thank you.
****
Just tried PDF and this is the most readable version. Is converting to PDF the best option, then?

[JSWolf]

Where was this book obtained?

[dwig]

Originally Posted by ChipAHoy

On Calibre, I have a book that is in MOBI, PDF and AZW3. ... Do these kind of ebooks require different handling or conversion steps than a book obtained from Amazon?
...

Anytime you acquire a book in multiple formats and one of those is PDF you should expect problems. Such bundles often begin ebook life as PDFs, either scans of printed pages or PDF exports from the publisher's page layout files that were created for print use. More often than not, the other formats are created by conversion from the PDF with the usual horrid results. There is absolutely no easy fix in these cases.

[deback]

I can fix the formatting if you have a way to get the file to me.

[Tex2002ans]

Originally Posted by ChipAHoy

On Calibre, I have a book that is in MOBI, PDF and AZW3. I transferred the ebook to my Kindle Paperweight reader. The format is a mess, sometimes there is text just on the left side of the screen in a narrow column, there are page headings in the middle of text, the spacing is all over, etc.

As dwig has stated, the other formats could have been auto-generated.

For example, Archive.org offers their ebooks as the original PDF scans OR a multitude of alternate formats (EPUB, MOBI, TXT, [...]).

To get these alternate formats, usually the PDF is just fed through an automatic conversion. As you found out, these alternate formats are typically full of errors (typos, bad formatting, missing text, still have page numbers, broken paragraphs, etc. etc.).

Something like a basic Fiction book might turn out "okay", but the more complicated the book, the worse errors are going to be (footnotes messed up, tables broken, formulas missing, [...]).

Originally Posted by ChipAHoy

Just tried PDF and this is the most readable version. Is converting to PDF the best option, then?

PDF was built as a final OUTPUT format, but it is one of the absolute worst as an INPUT format... and it requires a lot of manual labor to clean and correct to create a great ebook.

If you want to read on your Kindle, your best bet would be to try to find an actual ebook version of the book: from the library, find on Project Gutenburg, purchase from B&N/Amazon, find on the MobileRead EPUB section, etc.

If you are stuck with the PDF, the best way is to read the PDF on a larger device (tablet, computer).

If you want to TRY to make the PDF a little more palatable for the Kindle, you may have some luck with willus's k2pdfopt program.

There is no easy/automatic way to convert from PDF to ebook with satisfactory results. If you still want to pursue manual cleanup, you need to get your hands dirty. There has been a lot of discussion about this over the years, here are two of the topics where PDF -> ebook was discussed:

https://www.mobileread.com/forums/sh...d.php?t=223817
https://www.mobileread.com/forums/sh....php?p=2878245

[BetterRed]

Originally Posted by Tex2002ans

[ . . . ]

PDF was built as a final OUTPUT format, but it is one of the absolute worst as an INPUT format... and it requires a lot of manual labor to clean and correct to create a great ebook.

[ . . . ]

See also ==>> Calibre User Manual FAQ - What are the best source formats to convert?

BR

[JSWolf]

I'm thinking the source is PDF and that maybe this book is not legit. I've read of these sorts of problems but the book in question did turn out to not be legit. So please wait until we find out what's going on before trying to help.

[Tex2002ans]

JSWolf, you really should stop this piracy inquisition in every single topic. It doesn't do any good.

There are plenty of legitimate sites where books/articles/journals are released as PDF versions.

Look at Archive.org/JSTOR, or all the complicated scientific papers on arXiv.org, or any of the places I do a lot of conversions for.

And the available EPUB/MOBI versions of those documents may just be auto-converted junk. I know a lot of the reason why I got into conversion initially was because the EPUBs were poorly converted and PDFs are/were the only "real" available format of a certain book.

[sjfan]

Originally Posted by JSWolf

I'm thinking the source is PDF and that maybe this book is not legit. I've read of these sorts of problems but the book in question did turn out to not be legit. So please wait until we find out what's going on before trying to help.

I've read about people stabbing other people, but I don't assume everyone who has a knife in their kitchen is a criminal.

As in all facets of life, we should assume innocence until there's some proof of guilt. It's both cynical and antithetical to basic tenets of due process to assume guilt, especially on the basis of something as threadbare as “it's a PDF conversion”. Legitimate PDFs are commonplace. And unless there's some evidence of malfeasance, it's none of our business where someone got their books from—it's not really even an appropriate question to ask, IMO (there may be legitimate privacy reasons they don't want to disclose that information, be they health reasons, political, social, or other).

To the OP:
PDF is a tough format to convert. The ideal solution—as others have noted—is finding a better source file to convert (epub, mobi, azw, even HTML is often okay).

(Side note: If it's a technical paper you may be able to get your hands on the LaTeX source if you contact the author, which is often more easily converted, be it directly or via HTML.)

But if you're stuck with the PDF as the only source, and k2pdfopt isn't working, you could try converting in Calibre. Playing around with the heuristics tab and/or the unwrapping factor can often produce dramatically better results than the default conversion.