How to limit the weight of an image to a specified amount of MB

Remco Viëtor remco.vietor at wanadoo.fr
Tue Sep 24 17:24:07 BST 2024


On mardi 24 septembre 2024 17:52:19 CEST Jonathan Kamens wrote:
> On 9/24/24 10:45 AM, Remco Viëtor wrote:
> 
> > On mardi 24 septembre 2024 16:33:37 CEST Jonathan Kamens wrote:
> > 
> >> On 9/24/24 10:30 AM, Remco Viëtor wrote:
> > 
> > (...)
> > 
> >> The GIMP JPG export dialog displays in realtime the approximate size of
> >> the exported image as you move the image compression level slider, so it
> >> is apparently possible to at least estimate how big the an image is
> >> going to be when saved at a particular level even before you do the
> >> actual export.
> >>
> >>
> >>
> >>     jik
> > 
> > But note the "approximate" in there. That means you can end up with a
> > file
> > that is larger, and some systems reject files that are too large. And
> > there is
 a point where lowering the quality settings really degrade the
> > image.>
> >
> >
> > And I've seen files which were up to twice as big as others with the same
> > quality settings (and for the same image size in pixels)...
> 
> 
> Notwithstanding all of this, it is still a useful feature that people want.
> 
> I doubt you're telling anybody here anything they don't already know.
> 
> Gilles has already replied in this thread and made it clear that he has 
> no particular objection to the proposed functionality, so I'm not really 
> sure why you feel compelled to continue to spread F.U.D. about it.
> 
>    jik
> 
> 

If you want to call it F.U.D., fine with me.

Thing is, there are enough applications where a file cannot exceed a certain 
size. Having an approximate size before saving is nice, *as long as you know 
the limitations of that approximation*. And if you know what happens when you 
keep lowering the quality level...

It's not hard to check the exact size straight after saving, it's a lot more 
time-consuming having to redo the export because a bit later you discover that 
the saved file is just a bit too large (e.g. because metadata were added).








More information about the Digikam-users mailing list