Hi,

almost all BIMP-procedures alter the size of the files, usually compressing them- why?

Using v. 1.18, GIMP 2.8.

Thanks!

I've never used Bimp, but I suppose that if it does not have own quality settings,
it probably shares the same settings that you choose when saving (export to) normally with Gimp to some other than its native format.
The quality settings can be saved for each format.
You can test it by saving a file to .png for example, dragging the compress slider to zero and saving the export settings.
I always do this with Gimp installation first with jpg and png cause I'm used to use 95% for jpg quality and level 1 for png.

Assuming the point of using BIMP is that you are changing some property of the image, there will be cases where there will be better compression, more areas of the same colour etc.

However you need to compare like with like.

The BIMP defaults are different from the Gimp defaults, up to you to set them.

The main one for jpeg is the quality setting, it is lower in BIMP hence a smaller file size.



For png you might end up with a larger file. The compression setting is the same but extra information can be stored in the file, comments etc.

Thanks for the answers!

Rich2005:

But quite confusing: size of original jpg file: 7.457 KB.
With quality set to 85: 3.702 KB,
and with quality set to 100: 16.192 KB

How can this be explained?

It is because the jpeg quality versus file-size curve is not linear.
It is not a percentage.

The curve is like this.



100 quality still has a tiny bit of compression.
100 to 80 has a big saving in file size for not much loss in quality.
After that, small savings in file size can result in a big drop off in quality.
Lower than about 50 quality and jpeg artifacts become very noticeable.

Those BIMP jpeg settings you queried are very like the default Gimp 2.6 values. Changed a bit in Gimp 2.8

You should be exporting in Jpeg format only after all editing is finished.
Save your work in Gimp xcf format which keeps layers, paths, guides etc.

Thanks;)

So, in other words, diverse manipulation results in less information in the files > more compression, right?

But not necessarily loss of quality, right?

But how explain that to a client who wants his/her photos in highest resolution possible?

Do a search on jpeg compression and also jpeg artifacts

example this one: https://photo.stackexchange.com/questio ... about-them

Does your client insist on jpeg format? Is file size one of their criteria?

They probably will not be able to tell the difference between say 98 quality and 90 quality, apart from the file size.

If you are doing a lot of editing to a photograph, save it in xcf until complete. Then up to you (or client). Export alternatives tiff(lossless), png.

Back to the original subject, What operations are you doing in BIMP? Do you need a batch operation? There are other tools depending on what is needed.

Hi!

I use BIMP for batch operations only, mainly adding a watermark and/or applying automatic levels. Whatever the case, the file is always compressed.

Input: jpg only, output as well.

I also change the settings for compressing IF I want that to change the compression; otherwise I have mainly used the default for BIMP, which is 85.

With GIMP itself I don´t have any problems!

BIMP is not a mind reader, default setting is "generally works for most situations"

Use BIMP Change Format and Compression as a final procedure.



Set the values as you require. If worried about the quality setting bump-up to 100
Consider saving the settings to a .bimp file for future use. The individual procedures can be edited if/when required.

FWIW, Different cameras have different jpeg quality settings for "best quality". Checking on some of mine, anything from 90 to 98

How to determine the existing quality setting.
One way is command line ImageMagick (gives reams of info) http://i.imgur.com/y19fCVx.jpg
Another with a GUI is XnViewMP (which can also batch adjust your images) http://i.imgur.com/CFI9mFB.jpg

OK, great!!!

I had started changing compression for clients, not sure if it was necessary.

One final question: You have checked progressive: what does it imply?

Thanks again a lot!!!

To ensure that the jpeg when displayed on a computer monitor scans in one line after another, usually bottom to top.

The alternative is interlaced, where alternate lines are scanned, so two complete scans make a whole image. This was done when computers were slow so the viewer sees a (partially) complete image reasonably quickly.

Still used on some TV displays, as in 720p (progressive) 1080i (interlaced)