JE-Comics Converter

[Kosst Amojan]

Azayzel, maybe you can help me. I have a Reader and for some reason I cannot turn out readable comics. The files I use are standard comic sized scans at great resolutions but whenever I load them onto the reader after I use the converter, the text is way too small and I can only pick out a few words (my eyes are fine). Without the converter though I can't read anything.

Perhaps the text in manga is larger then in standard Marvel and DC comics, I don't read them so I don't know. If any of you have any suggestions, please help!

[Kirys]

I think that the problem is that anyway the resolution of the reader is too low for the size of the font of your comic.
is the comic readable (after the conversion) on the computer screen?
di you tried read it in landscape mode?

[Azayzel]

Originally Posted by Kosst Amojan

Azayzel, maybe you can help me. I have a Reader and for some reason I cannot turn out readable comics. The files I use are standard comic sized scans at great resolutions but whenever I load them onto the reader after I use the converter, the text is way too small and I can only pick out a few words (my eyes are fine). Without the converter though I can't read anything.

Perhaps the text in manga is larger then in standard Marvel and DC comics, I don't read them so I don't know. If any of you have any suggestions, please help!

I actually run the comics through JE-Comics Convertor and then through PDFRasterfarian. Before I used PDFR, I had the same problem as well; I could just barely make out some of the dialog without viewing it landscape mode (hate this, as it's strange to hold and I don't like to read a half page at a time). The JE-CC stitches all the images into one file, of the correct size for the reader, and PDFR applies a few filters to enhance the PDF output so that it is more easily viewable on the Reader; i.e., edge enhancement and noise removal. If you go to the Reader Developer's Corner, you can find that app and run it in conjunction with this one.

One issue you will find with US comics vs. manga is the physical size difference of the layouts. Here's a physical size comparison I've done myself, so you can see that they can fit more comic frames on a page in a US comic = more to fit on the eBook's page = smaller font; though the ratio length-to-width is almost the same -> .67 :: .65

Manga (TokyoPop) 5" x 7 3/8" or 126mm x 188mm
US Comic (Image) 6 5/8" x 10 3/16" or 169mm x 259mm

For demo purposes, I've attached a comic I converted with PDFRasterfarian before I found out about JE-Comics Convertor and a manga converted using both apps. If for some reason I need to remove either, I will happily do so, they are simply here for demostation. I'd u/l the PDF output of the manga, but it's 23MB rar'd. Secondly, I haven't read the manga I u/l, I just flipped through it to check quality; so if the story sucks, sorry. Finally, regarding reading manga so it makes sense... you start in the upper-right and work left, instead of upper-left and right like with US comics; sames goes for dialog (still top-to-bottom )

[Azayzel]

Originally Posted by Kirys

Actually it don't load PNG yet, just because I forgot to add the support :P (i focused only on the resolution issues), next version will I promise

Did someone tested the lossless format on the sony?

Kirys: Ok, I tested out the lossless vs original samples that you provided in the top post. The lossless is a bit clearer, but the font is still a little light in spots. Is there any type of edge or font enhancement you can apply to sharpen the fonts or bold them? They are still readable. Strangely enough, the differences I could notice in portrait are not as noticeable in landscape mode (would have thought they'd be more apparent due to the larger size.

Thanks for pointing out the rotation capability, I wasn't aware of it and this will save with some of those darn double-page images thrown in there every now and then (were horribly small in portrait).

Finally, before I let you guys have some rest for a while... I am not sure I follow what you are saying when you said that PNG format does not load yet. Do you mean that you haven't set it as a selectable option? If that's the case, it doesn't bother me in the least as I can select all images within the folder using the "All Files" option. I just converted a 199-page manga, where only the title page was jpg and the rest were png.

Have a good evening or day, whatever it happens to be!

[Kirys]

Originally Posted by Azayzel

Kirys: Ok, I tested out the lossless vs original samples that you provided in the top post. The lossless is a bit clearer, but the font is still a little light in spots. Is there any type of edge or font enhancement you can apply to sharpen the fonts or bold them?

I'm not an expert of image processing. I can try with some sharpening and contrast enhancement but is not easy to achieve the same result without knowing a good recipe

Originally Posted by Azayzel

They are still readable. Strangely enough, the differences I could notice in portrait are not as noticeable in landscape mode (would have thought they'd be more apparent due to the larger size.

Whe you go into landscape mode the reader stretch the image (that was resized to the effective resolution of the screen. And its interpolation is not as good as the one produced by our softwares

If you post a filtered version of the lossless pdf i can have an idea of the difference from my output (i'm not a win user so i cannot use PDFraster).
Bye

[narve]

Originally Posted by Kirys

I do plan to release the source code eventually. But actually I want to finish the base architecture of the software and its main features (i should also decide an os license compatible with the libs I'm using).

BTW you want to make an higher version of the output or you want to just rotate the image?
If you want to rotate the image select the checkbox under the R colum fo the images you want to rotate

Most of my pics are landscape, so I just want to get higher average screen usage. I guess rotating the picture would help -- I haven't tried it. However then I need a "rotate all" button, or: When multi-selecting rows, and then clicking on the "R"-column, the R-selection applies to all rows. Then you must override the rowSelected-event and stuff, it is un-necessary tricky but not too hard. I guess I can do it, even though I am a server-dude, not a swing-guru

[Azayzel]

Originally Posted by Kirys

If you post a filtered version of the lossless pdf i can have an idea of the difference from my output (i'm not a win user so i cannot use PDFraster).

Hmm, makes sense about the landscape stretching the image. Didn't think of that. Before I upload/post anything, I would like to clarify what you want to take a look at, the PDF of the comic I converted before running through the other program or the result in PDF or TIFF format? Also, which PDF, the manga or the comic? Thanks again for your time! These aren't really showstoppers for me, as the combined effort is more than adequate.

[Kosst Amojan]

Originally Posted by Azayzel

I actually run the comics through JE-Comics Convertor and then through PDFRasterfarian. Before I used PDFR, I had the same problem as well; I could just barely make out some of the dialog without viewing it landscape mode (hate this, as it's strange to hold and I don't like to read a half page at a time). The JE-CC stitches all the images into one file, of the correct size for the reader, and PDFR applies a few filters to enhance the PDF output so that it is more easily viewable on the Reader; i.e., edge enhancement and noise removal. If you go to the Reader Developer's Corner, you can find that app and run it in conjunction with this one.

Thanks man, I'll try that when I get home from work. I actually use both but didn't think to use them in conjunction.

[Kirys]

Originally Posted by Azayzel

Also, which PDF, the manga or the comic?

TestSonyLossLess.pdf that is available in the first post of this thread

[Kirys]

Originally Posted by narve

Most of my pics are landscape, so I just want to get higher average screen usage. I guess rotating the picture would help -- I haven't tried it. However then I need a "rotate all" button

The interface lack many things, it is because I've focused on the engine rather than the interface itself, I'll add a popupmenu to the list for this kind of things in next version

Originally Posted by narve

I am a server-dude, not a swing-guru

Yeah I usually do server side thing in java, that is the first real swing app I do, and also the first app that have to process images

[Kirys]

I've uploaded v0.3.0

[splat]

Originally Posted by Kirys

MAJOR: Now you can add also all the content of a directory.

Any possibility of in a future release getting it to do a deeper scan of a directory and include sub directories.

eg

I have directory structure manga\volume\chapter\pages.jpg where there could be more than one volume and chapter directory.

I'd like to be able to add the directory "volume" and have it scan the directories chapter1 - chapterN and add all the images in those directories into the PDF.

As always keep up the good work, this is a great application

[Kirys]

I a future there will be that feature, but now it would create a big problem to the way it does sort files, so I have to think the most flexible way to solve this problem.
Personally i do prefer having different chapters into different pdfs
First I want (and need) to focus on the output quality that still need work, but for that i do need someone with an iliad or a sony and a digital camera, to help me figure out how the changes reflect on the readers.

[splat]

Originally Posted by Kirys

Personally i do prefer having different chapters into different pdfs

I had been doing that, but it got a bit annoying creating 30-40 page pdfs, when a 200page one worked just as well.

EIther way, take your time on this feature, it's just as easy for me to move all the files into the one directory and then convet

[splat]

Only just getting around to using 3.0 (rev 1.19) fully and I am having a few issues.

1) converting PNG images does not work. I can add them to the image list fine
but when you click on generate it just sits at "completed 0%" and does
nothing

2) Adding a directory adds .DS_STORE files to the list for conversion.