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⚠️ Please note that this topic or post has been archived.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. ⚠️

Global copy/apply when image dimensions don't match

Comments

6 comments

  • SFA
    This is not something I have noticed working with RAW files.

    I rarely use TIFF and never blended so far.

    Is this something that only applies to TIFF images as far as you have seen?

    If so, is it any TIFF image or only the blended ones?


    Grant
    0
  • NN636445455587027209UL
    I'm 100% sure it's not the file format (TIFF) that's at issue - it's the slight difference in image dimensions that can happen when PS aligns say 3 or 5 bracketed shots.

    If I go into PS and adjust the canvas size so that it's a match for the source (already edited) file, I can then perform a global copy / apply as normal. I guess C1 just throws up its hands if the image and layer mask dimensions don't match perfectly, however in this case the mismatch is typically only 1 pixel. LR used to take this in its stride, and it would be nice if C1 could do the same.

    This is about the only thing in C1 that bugs me thus far (which is seriously good going), but nevertheless it does bug me 😊

    [quote="SFA" wrote:
    This is not something I have noticed working with RAW files.

    I rarely use TIFF and never blended so far.

    Is this something that only applies to TIFF images as far as you have seen?

    If so, is it any TIFF image or only the blended ones?


    Grant
    0
  • SFA
    Are you copying from TIFF to TIFF?

    All adjustment types or just some of them?

    Are you making the adjustments on the background "layer" or through a higher layer?


    Grant
    0
  • NN636445455587027209UL
    TIFF to TIFF, multiple adjustment layers covering a wide range of adjustments.
    0
  • Edward Jenner
    I'm with you. Fortunately for me I have not had to do this with a lot of adjustment layers, but I agree it is annoying.

    From a development POV I'm not sure how they would get around this - maybe allow up to 5 px difference and just shave off the outside if the destination is smaller? What if the destination is bigger and you have a complex mask? I guess you could extrapolate - there are certainly algorithms to do this and 5-10 px should work pretty easily, but you would still have to decide where the two images match up. So getting something both bulletproof, 'correct' and intuitive (so the user is not surprised) is not trivial.

    Personally I would be fine if the layers copied, but the masks were all set to 'black'. That is usually what I want - then I can just re-draw the masks.
    0
  • NN636445455587027209UL
    Going on memory LR seems to use a kind of ratio mapping algorithm, so it's not pixel accurate but is usually close enough.

    That said, your suggestion:

    [quote="NNN636460165578985672" wrote:
    Personally I would be fine if the layers copied, but the masks were all set to 'black'. That is usually what I want - then I can just re-draw the masks.


    would be a really good compromise, and I should think would be much easier for Phase One''s developers to implement.
    0

Post is closed for comments.