Bug 323440 - Please do not [always] save as PNG by default
Summary: Please do not [always] save as PNG by default
Status: RESOLVED INTENTIONAL
Alias: None
Product: Skanlite
Classification: Applications
Component: general (show other bugs)
Version: 1.0
Platform: Debian testing All
: NOR wishlist
Target Milestone: ---
Assignee: Kåre Särs
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-08-13 01:57 UTC by Philippe Cloutier
Modified: 2013-08-13 12:52 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Philippe Cloutier 2013-08-13 01:57:43 UTC
I just scanned a simple monochrome text document on a letter-sized sheet of paper. Skanlite saved this to a 9 MB file by default. Disk space surely got cheap, but I was expecting more like a 100 kB file. OK, that's harder to achieve for a color 38 MP scan. By default, Skanlite saves scans as PNG-s (at least here). I tried saving as JPEG instead, and instead got a 1.6 MB file, which is at least several times better. Sure, JPEG is lossy but if I compare the 2 files, the JPEG actually looks better. They both have great quality of course, but if I focus on a certain letter and compare, JPEG is the smoothest. (Gwenview bug?)

This can be permanently worked around by selecting JPEG in the preferences dialog.

My gut feeling says lossless compression must not be very efficient for scanning. I suggest to make JPEG the default. If defaulting to PNG was intended, maybe the format selection could be improved? Could the content determine which of PNG or JPEG should be choosed? Like Skanlite compares the PNG size to the JPEG size and goes with JPEG unless PNG would be smaller for the content in question?
Comment 1 Kåre Särs 2013-08-13 06:21:50 UTC
Thanks for the report!

If you scanned a document you probably would like to have the document as a pdf with the image jpeg compressed internally and the text recognized and embedded in a pdf. 

My needs/itch has been more on the photo side and I have not added half-harted document scanning. It would take quite a lot of work to get on par with gscan2pdf.

For the moment I will mark this as a wontfix as I think the proper sollution to your problem is to add a separate/different spcialized interface for document scanning.

/Kåre
Comment 2 Philippe Cloutier 2013-08-13 12:52:54 UTC
I would be fine with a JPEG file.
Isn't using JPEG rather than PNG even more important for scanning photos than for scanning text?