I'm working on a (c#)-program to create books for the sony reader. However, my sony reader is still somewhere at a fedex-building waiting to be delivered, so I can't test how it looks like.

Can you guys help me out by loading this lrf in your reader and see if the quality is good enough to read?


http://www.mobileread.com/forums/attachment.php?attachmentid=3712&stc=1

How it was created:
I've used imagemagick to resize, filter and conver the images. I'm using the lanczos filter.
Then all the files are made into a pdf and finally into a lrf with the help of rasterfarian.

I've added some pretty tough samples for testing: Ah my goddess (known for the annoying small fonts), berserk (large fonts, but too many shades) and Gunnm( a lot of dithering and small fonts)

Hi! Looks like a great start. Not quite sized right. In small does not cover screen - blank on right hand side. In medium view then bottom of page cut off and no way to scroll down to it. Hope this helps and that you get your machine soon.

Hi! Looks like a great start. Not quite sized right. In small does not cover screen - blank on right hand side. In medium view then bottom of page cut off and no way to scroll down to it. Hope this helps and that you get your machine soon.


Thanks for the reply. When you say that it doesn't cover the screen, does it mean that you can see the full imagepage, but the screen is not fully used?

The screen is not fully used in any sizing and in medium and large view then not all of the imagepage is visible.

OK. My reader arrived today and just in time too because my litle tool is ready for my needs (still some bugs, and need to add more control, but it produces nice images and PDFs)

I've revisioned my tool a bit.

The way images are processed are different.

It won't convert the images anymore to gif.
Instead it will try to trim the image (remove borders) as good as it can (using imagemagick's trim function with fuzz for the techies here).
Colours will be changed to grayscale.
Then it will throw some anti-alias and filtering.
And finally resized to a max height of 754.

All the images will be made into a PDF. (one chapter per folder/zip)
Centered, without margins.
The chapters will be shown in the TOC

And that's where my tool end.

I'll use rasterfarian to convert the PDF to LRF. The newest version of rasterfarian uses netbpm, so that's more suitable for teh decoloring and edge enhancing.


And the end result are these files.
I like the TOC :)

(I'm now trying to test out the size limit. A file with 1600 images as pages is not loved by the reader....)


A few pictures I made of my new toy :)

http://img.photobucket.com/albums/v103/athlonkmf/IMG_1023.jpg
http://img.photobucket.com/albums/v103/athlonkmf/IMG_1021.jpg
http://img.photobucket.com/albums/v103/athlonkmf/IMG_1020.jpg

Looks great, I'll test it with some Naoki Urasawa mangas too as soon as you release this.

I just tested it and it was very fine.

One bug report though: the file "serial" number (I don't remember how it's really called) was the same in the 2 test files - I don't know how you implement that particular number though.

I also have 2 question:

Will your program work/compile under linux directly or will I have to go through wine or some such.
Do you intend to release it soon?:D


Thanks for the good work.

So, if I understand correctly, this tool will convert zip archives to PDF, and then use rasterfarian for the conversion to LRF? If so, this tool might be interesting to those of us using other devices that can read PDF, e.g. the iLiad.

you can use this thread http://www.mobileread.com/forums/showthread.php?t=11261 it's where I put the release.