Adventures in discovering the K4PC PID.

[skindle]

I am trying to decipher the topaz format based on the K4PC code, and it is slow going. K4PC makes huge use of the C++ stl and the template code is a real pain to reverse. Heavy use of maps and vectors, maps of vectors and vectors of vectors

[orwell2k]

I'm curious about why Amazon uses this apparently crappy Topaz format?

I don't have a Kindle, but Topaz ebooks seem to look much crappier on K4PC than the Mobi formatted books.

My limited browsing of the Kindle-related forums seems also to indicate that in general Kindle users also don't like the format - slow reader performance, poor display quality, etc.

And only a small % of Amazon books seem to be in Topaz compared to Kindle Mobi.

But then, I guess our clever software gurus would then have to find another challenge - although at this rate there won't be too many left, thank goodness!

[daffy4u]

Back in the day, some Topaz books had features that mobi books didn't. Embedded fonts and the ability to zoom some images.

Now Kindles can zoom images but we can only use the one font on the device (hacked or not).

[poohbear_nc]

Originally Posted by skindle

I am trying to decipher the topaz format based on the K4PC code, and it is slow going. K4PC makes huge use of the C++ stl and the template code is a real pain to reverse. Heavy use of maps and vectors, maps of vectors and vectors of vectors

If you ever manage to crack & unlock topaz you will get the Ebook version of the Nobel prize!
Ironically, such a coup would actually increase Amazon's sales of ebooks if owners of other devices were able to read topaz-only books.

[rcuadro]

Originally Posted by poohbear_nc

If you ever manage to crack & unlock topaz you will get the Ebook version of the Nobel prize!
Ironically, such a coup would actually increase Amazon's sales of ebooks if owners of other devices were able to read topaz-only books.

Absolutely...

[daffy4u]

Originally Posted by poohbear_nc

If you ever manage to crack & unlock topaz you will get the Ebook version of the Nobel prize!
Ironically, such a coup would actually increase Amazon's sales of ebooks if owners of other devices were able to read topaz-only books.

But, if they don't crack it, with all this attention, we could end up seeing more Topaz boosk.

[rcuadro]

Maybe... but they would be limited to the Kindle since I dont see other readers using it...

[poohbear_nc]

Originally Posted by daffy4u

But, if they don't crack it, with all this attention, we could end up seeing more Topaz boosk.

True - if Amazon really wanted to lock down titles.

But conversely - with all this attention - I think it will eventually be cracked if enough bright folks (now around the world with K2i, DXi, and K4PC) work on it.

[daffy4u]

Originally Posted by poohbear_nc

True - if Amazon really wanted to lock down titles.

The real horror would be that other publishers license Topaz from Amazon and we end up finding it everywhere.

[KevinH]

Hi,

You may want to check out the last couple of entries in the Dark Reverser's Blog. Some work has been done on understanding the contents of the Topaz files towards converting them to another format.

Hope someone picks it up and runs with it or something better.

http://darkreverser.wordpress.com/2008/02/13/new-blog/

KevinH

[phenomshel]

Originally Posted by daffy4u

The real horror would be that other publishers license Topaz from Amazon and we end up finding it everywhere.

Shhhhhhh, Daffy!!!!!!!!!!!!!

[daffy4u]

Originally Posted by phenomshel

Shhhhhhh, Daffy!!!!!!!!!!!!!

Sorry. But we all know it's not an original thought.

[phenomshel]

Originally Posted by daffy4u

Sorry. But we all know it's not an original thought.

Oh, I know, it's crossed my mind several hundred times. The bear of it is, I really don't think Amazon set out to make Topaz hard to convert, that just happened to be a very nice side benefit for them.

[labba]

Originally Posted by skindle

I am trying to decipher the topaz format based on the K4PC code, and it is slow going. K4PC makes huge use of the C++ stl and the template code is a real pain to reverse. Heavy use of maps and vectors, maps of vectors and vectors of vectors

yea i know what your saying ..
me and my friend wanted to work togather on this and split the work... but reversing as a team is the hardest thing i ever done .. it's looking like an impossible mission to reverse togather and the only way is by doing it alone..

i hate that!

i wanted to speed things up and all it was doing (the team work) it slow us down..

[KevinH]

FYI:

The topaz file format DRM has been broken using CMBTC's work (and CMBTC's program does work - he just put in some things to prevent the code from working without slight modifications - to prevent its easy use for theft).

By slightly modifying that code you can dump the "dict", "other", "glyphs", "img", "page", and "metadata" sections in unencrypted form *and* uncompressed form.

A new python program that converts the unencrypted data from the "other", "page" , "glyphs" sections ("img" sections are straightforward) using data from the "dict" section has been written.

Search for the Dark Reverser's New blog. Search for "cmbtc" in that log and start reading down from there. The code you want to play with after you have a fixed version of cmbtc working is called "decode_page.py".

The latest version works to decode both the page*.dat files and now the glyphs*.dat files.

More work is needed and help is wanted.

KevinH