Skip to main content

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

Output Quality Discrepancies

Comments

4 comments

  • Permanently deleted user

    Hi,

    Your pict on the Google Drive is not public - one needs an access to see it.

    0
  • Patrick Chan

    Hi Claude,
    My bad. I have updated the access right to the folder.
    Just in case, I re-attach the link: https://drive.google.com/drive/folders/1mlcD-Or8WMv9zyrX-q4SNUdEdlljfb9k

    Thanks,
    Patrick

    0
  • SFA

    Patrick,

    Are sure these are no just the effects of compression?

    Have you compared the output files (in a viewer through which they are likely to be viewed) with the edit process image viewed with Output Proofing active and at the 100% scale?

     

    Bear in mind that your full size image will be making use of more of the originally captured data (pixels) then smaller ones.

    In photos like the one embedded in the post this will tend, typically, to make the darker areas darker as compression calculations can reduce details to the dominant colours and luminosity as data pixels are discarded (for size and then compression) and the smallest areas of detail that may remain become so reduced in size that the eye can hardly see them or the delivery media cannot display them.

    In some situations a different effect appears. For example a large area of grass may appear to be green but is actually mostly shades of colour closer to yellow in colour creation terms. So compressing certain types of image with a lot of greengrass may intensify the colour in yellow areas and thereby render the grass area with more yellow and less green as we "see" it after processing.

    In effect that is the opposite of your observation since the darker areas may appear lighter. (Mostly mid tones would be involved) in the areas of grass. 

    Choices made with output processing recipes (or export parameters)could well have an effect.

    If you know all of that and have already taken it into account then it may be necessary to start down the discussion path related to "end to end" colour management.  I will leave that to those who have a lifetime commitment to the subject!

    0
  • ---

    nothing unusual,  this effect is caused simply by the resizing algorithm and has nothing to do with jpg compression (could be easy checked by outputting a tiff )  and I doubt you get much better results in other applications using high quality resizing settings like bicubic. the brightest detail in the large image is already very small just 1 or 2  pixel wide so it is no surprise they disappears when downscaled !  but there is a solution for you I would suggest to try out a more simple algorithm like bilinear which does less smoothing, PS or affinity photo can do this.

    0

Post is closed for comments.