k2pdfopt: optimizes PDFs for viewing on e-readers

[Gibook]

Hello everyone! Thanks for this wonderful software.
Any chance to configure properly ONYX BOOX Note2 10.3" E-reader?

[willus]

Originally Posted by tomas94

Hello everyone! I am in search of an E-Reader. The discovery of this wonderful app completely changed my perspective on these devices. Now, I am thinking of getting a 10th generation Amazon Kindle, does anyone know if the app works for that generation? It made me doubt the fact that the app lists "Kindle 1-5". Thank you very much!

From what I gather from the web, the standard Kindle all the way up through the 10th generation has not changed screen resolution, so choosing "Kindle 1-5" should work fine for a 10th generation Kindle as well. If you want to post some screen shots, I can confirm.

[willus]

Originally Posted by Gibook

Hello everyone! Thanks for this wonderful software.
Any chance to configure properly ONYX BOOX Note2 10.3" E-reader?

There is no built-in option directly for that device, but the web site lists the resolution as 1404 x 1872 and 227 dpi, so you would use these options to work with the ONYX BOOX Note2:

-w 1404 -h 1872 -odpi 227

...though if you post screen shots from your device displaying my test PDF file, I can determine how much of the screen is actually used for displaying the PDF.

[Gibook]

Thanks for your quick reply. In attachment the screenshots of your file test

[willus]

Originally Posted by Gibook

Thanks for your quick reply. In attachment the screenshots of your file test

I meant screenshots directly from the ONYX BOOX Note2--not from Adobe Acrobat. Most e-readers have the ability to capture their screen to a bitmap file with some kind of key press or button press.

[Gibook]

Sorry!

[willus]

The images are 1404 pixels wide and 1834 pixels high, so the right options would be:

-w 1404 -h 1834 -odpi 227

[Gibook]

[VladimirS]

@Mr.Willus,
creator of remarkable piece of software - K2pdfopt

First, thank you thousand times for K2pdfopt !

Mr. Willus, can you give us some basic tips how to create,
or maybe, better to say EDIT, any existing PDF file,
which is scanned PDF, with or without OCR layer,
in order to make it lighter for our e-ink devices
(lighter in terms of hardware requirements CPU, RAM etc),
which is easily recognised as surplus time delay when turning pages.

Your software is great tool which most of use use to
split scanned PDF files, which are originally scaned
as 2 pages on one page. Command line command is
(k2pdfopt -mode copy -grid 2x1x0 -w 0.5s -h 1s file.pdf)




Which properties of PDF file, like:

1) number of pages in PDF
(possible solution - reduce number of pages by splitting file)

2) page size in pixels
(possible solution - reduce resolution using -ds option)

3) resolution of scanned PDF (with or without OCR)
(possible solution - reduce resolution using -ds option)

4) PDF version (1.3, 1.4, 1.5, 1.6 etc)
(possible solution - convert PDF with some tool to lower version)

5) integrated elements inside pages (bookmarks, table of contents)
(possible solution - remove unnecessary TOC and bookmarks)



What are the steps (if they exist) we can take in order to
speed up file manupulation with our e-ink devices.
I think this question bothers many e-ink device owners.

Thank you in advance for your advices!

[willus]

VladimirS -- I have found that complex compressions such as JPEG 2000 (JP2/JPX) are usually the worst offenders for slower page turning times on readers. They make the PDF very small in size, but they take longer to decode. You can tell if your source PDF has these compressions by using this:

k2pdfopt -i source.pdf

In addition, complex layouts can also slow reading. That is why "-mode copy" works well for reducing the CPU load on the reader, because it converts each page into a bitmap using the standard "deflate" PDF compression which can quickly be decoded, with 4-bits-per-pixel gray level. In general, if the source PDF is slow to render and this is more important to you than the size of the document, converting to bitmaps with k2pdfopt should help, which is accomplished by being sure that native mode is turned off (-n-). The bits per pixel in the resulting bitmaps can be controlled with the -bpc option. The resolution of the bitmaps is most easily controlled with the -dr option. See my command-line usage page. I don't believe the version of the PDF inherently impacts the page turning time significantly, nor if the file has bookmarks.

[VladimirS]

Thank you very much, Mr.Willus, for all the help that you give to us!

[sunny_hill]

Hi,

Thanks for the software!

I've converted a PDF using the fitwidth mode and it reads great on an Oasis 2. The only trouble is that in order to read the PDF in landscape orientation, I have to set the orientation on the Oasis 2 to portrait which makes highlighting things difficult. Does anyone know if there is there a way to use k2pdfopt to produce a PDF that will display correctly in landscape orientation on the Oasis 2?

[willus]

You can see the dimensions used by the Oasis 2 by running this:

k2pdfopt -dev ?

Try just flipping the -w and -h values and turning off landscape:

k2pdfopt -mode fitwidth -ls- -dev ko2 -w 1583 -h 1200 ...

Or in the MS Windows GUI, select Oasis 2 and then flip the w and h values like shown in the screen shot.

[sunny_hill]

Originally Posted by willus

You can see the dimensions used by the Oasis 2 by running this:

k2pdfopt -dev ?

Try just flipping the -w and -h values and turning off landscape:

k2pdfopt -mode fitwidth -ls- -dev ko2 -w 1583 -h 1200 ...

Or in the MS Windows GUI, select Oasis 2 and then flip the w and h values like shown in the screen shot.

Thanks for the response. I tried what you mentioned and the document displays properly when in landscape orientation. However there is a border around each page. This border wasn't present in the previous PDF I generated using just the fitwidth mode. Do you by chance know if it's possible to remove this border?

[willus]

Do you mean a blank border? Or a black border? At this point it would help if you could post maybe a page or two of your source document so I could see what you are converting.