An .azw3 is a Palm database like a .mobi, and often contains two books: an old MOBI at the front and a KF8 book behind it, with the header saying where the second starts. KF8 is real XHTML, so it converts properly.
Drop the book and it will either show you the text or tell you exactly why it cannot. DRM is the usual reason.
KF8 arrived with the Kindle Fire in 2011 and is what every Kindle book was until the format moved on again. If you have a choice between a .mobi and an .azw3 of the same book, this is the one to convert.
Two books in one file
For years Amazon shipped the old and the new book in one container so that a Kindle from 2009 and a Kindle Fire could both open the same download. The header of the first book carries a record number saying where the second begins, and a converter that ignores it reads the worse half without ever mentioning it.
- XHTML and CSSrather than 1990s HTML
- Headings, lists and tablesmarked as what they are
- Pictures by kindle:embedresolved to their records
- Same DRM questionand the same answer
The question that decides everything
Before anything else: does the book have DRM on it. If it was bought from the Kindle store, it almost certainly does, and that is the end of the conversation.
What does convert is everything else, and there is more of it than people expect: manuscripts, samples, manuals, documents sent to a Kindle by email, a great deal of independently published work, and every public domain book from Project Gutenberg and Standard Ebooks.
What comes through
The text, in full, in reading order.
Headings, lists and tables, which KF8 marks properly.
Bold, italic, underline and the rest of the inline styling.
Pictures, resolved from their kindle:embed addresses and placed.
The title and author from the file's own metadata.
The Kindle's typesetting, which is the device's work rather than the file's.
Reading position, notes and highlights, which live on the device.
Books with DRM, which cannot be read at all.
Files in the newer KFX format, which is a different container.
Fonts embedded in the book, which the PDF sets in its own.
Which is why an AZW3 converts better than a MOBI.
The KF8 book, not the old one in front of it.
Rather than a broken file and no explanation.
When to use Calibre instead
This converts one book, in a browser, with nothing to install and nothing leaving the machine. That is the right trade for a file or two.
A whole library
Calibre converts in bulk, keeps the metadata and the covers, and manages the collection afterwards.
A book this refuses
HUFF/CDIC compression and the newer KFX container are both outside what a browser tool reads sensibly.
You want EPUB rather than PDF
For reading on a device, EPUB is the better target, and Calibre does that conversion well.
You need the metadata edited
Series, tags, covers and the rest are a library manager's job.
Calibre is free, and it is the right answer often enough to be worth saying so on a page that competes with it.
Why this runs in your browser
Books are personal, and an e-book is often a manuscript, a review copy or something sent in confidence. Unpacking a Palm database and laying text on a page needs no server, so there is no reason for one.
The file is read in memory, the book laid out with real fonts, and the finished PDF checked by reopening it before it is handed over. Nothing is transmitted and nothing is stored.
The tool is right above
Drop an .azw3, look at how it read the book, and take the PDF.
Common questions
DRM, the two books in one file, and what the conversion keeps.
01My book will not open. Does it have DRM?
Almost certainly, if it came from the Kindle store. DRM encrypts the text against the account that bought it, the key is not in the file, and nothing here will remove it. The tool says so plainly as soon as you drop the file rather than producing a broken PDF.
02What is the difference between AZW3 and MOBI?
AZW3 carries KF8, which is real XHTML and CSS, so headings, lists, tables and styling are proper markup. MOBI is a cut-down HTML from the 1990s. An AZW3 therefore converts noticeably better, and if you have both versions of a book, use this one.
03What is in the file besides the book?
Often a second book. Amazon shipped dual files for years, with an old MOBI at the front for devices that could not read KF8 and the KF8 book behind it, and the header says where the second one starts. This reads the KF8 half, which is the better of the two.
04Are pictures included?
Yes. KF8 refers to them with a kindle:embed address in base32, which is resolved to the record holding the picture and placed in the page. Anything that cannot be read is left out and reported.
05What about .azw, without the 3?
An .azw is usually an old MOBI with a different extension and DRM on it. If it is DRM free it will be read like any other MOBI. The extension itself tells you very little.
06Will the PDF look like the book did?
It will carry the same text, headings, lists, tables and pictures on a page you choose the size of. It will not reproduce the Kindle's own typesetting, which was the device's work rather than the file's. Choose the text size and margins here to suit what you are going to do with it.
07Should I use Calibre instead?
For a library of books, yes: it converts in bulk, keeps metadata, and handles the compression this refuses. For one file and no software, this is faster and the book does not leave your machine.
08Is my book uploaded?
No. It is read and converted in your browser. Nothing is transmitted and nothing is stored.