Skip navigation
Currently Being Moderated

Strange distribution problem

Mar 6, 2013 10:11 AM

Picture 1.png

Here I have several concentric squares in an uninterrupted stack.

I moved the topmost (small orange) one vertically.

I then went to the Alignment panel, selected all the squares and clicked on the one I moved.

Then I clicked on Vertical Distribute Center.

The right hand picture shows what happened.

One of the squares (that big orange one) near the bottom of the stack shot up so that its centre was above the centre of the topmost one in the stack.

 

The centres are evenly distributed but why is the stacking order not controlling the order in which the squares are arranged?

 

Fortunately the positioning of that big orange square was easy to correct, but something tells me this is not the way things are supposed to happen.

Any ideas?

 

Seemingly this does not always happen. I have tried the method on several stacks. Some were o.k., others showed similar behaviour to the above.

 
Replies
  • Currently Being Moderated
    Mar 6, 2013 11:12 AM   in reply to Steve Fairbairn

    It looks like it thinks the small orange square and the back one are somehow connected.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 6, 2013 11:25 AM   in reply to Larry G. Schneider

    You mean as in quantum entanglement, Larry?

     

    Illy may have developed further (out) than anyone suspected.

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 6, 2013 5:21 PM   in reply to Steve Fairbairn

    Stacking order has no bearing on such an alignment. Nor should it.

     

    JET

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 7, 2013 4:59 AM   in reply to Steve Fairbairn

    The stack was created in the reverse order, back square first, the oddball was second and the smallest one last.

    Again; stacking order has nothing to do with the distribution. You moved one of the objects, thereby moving the position of its center relative to the other objects. Then you performed the distribution, based on the vertical positions, not based on the stacking order.

     

    JET

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 7, 2013 6:26 PM   in reply to Steve Fairbairn

    ...so are you saying that I should deliberately pull all the squares a little bit out of alignment manually before...

    For the kind of distribution you are trying to do, yes (or else use a Blend or a Transform Effect).

     

    I've never done C++, but it would probably be a reasonable assumption that, programmatically, the routine adds the selected paths to an array, sorts the array by their center values, determines the range by subtracting the first center value from the last, determines the spacing by dividing the range by the number of paths, and then increments new positions to the paths in their sorted order.

     

    So in some situations, it may seem driven by stacking order when the paths have exactly the same center values, just because the stacking order is the order in which the paths are added to the array, and the resort doesn't disturb that arrangement.

     

    But a Bezier path is drawn from point to point, not  from its center outward. So the centers are calculated values. We don't know at what decimal precision that calculation is made, so mere rounding error could cause a sort of apparent "randomness" if the sort is done on those center values.

     

    JET

     
    |
    Mark as:
  • Currently Being Moderated
    Mar 7, 2013 8:10 PM   in reply to Steve Fairbairn

    I could not reproduce the misalignment, can you share your file?

     
    |
    Mark as:

More Like This

  • Retrieving data ...

Bookmarked By (0)

Answers + Points = Status

  • 10 points awarded for Correct Answers
  • 5 points awarded for Helpful Answers
  • 10,000+ points
  • 1,001-10,000 points
  • 501-1,000 points
  • 5-500 points