Nomenclature for Pinning

[roy]

Mike - while you're writing documentation....

When you wrote

Double pinning ..... is an older version of the term pinning, which refers to any automated way of creating curve control points
(in "ctrl-z and curve rotation" reply #1)

realisation suddenly dawned for me.  I'd been very confused about the term pinning up to that point, and had the sense of really not understanding curvemeister. 

So I've been wondering if it might help people learn more easily to distinguish explicitly between the older, more general use of the term pin, and the newer, more specific?  As the pins you find in the pin-pane at the bottom of the screen attract the curve to them, perhaps a term like "magnetic pin" would help?  Or even just labelling the pin-pane "Color Pins" to explicitly distinguish them from "normal pins"?  Perhaps my mistake came from the fact that only "color pins" have the pin icon?

(As I write this I realise I'm still learning, so not sure, but perhaps sharing the fruit of my confusion will help you help others learn more easily?)

[-default]

I think there needs to be something in the Wiki about this.  I'm thinking of adding a glossary, and perhaps it would contain a more detailed description of pinning, along these lines:

1) In Curvmeister, the noun "pin" always refers to a "color pin".  Shadow, highlight, neutral, and the pins in the pin palette are all examples of pins.

2) The verb "pin" always refers to a process that results in automatically adding control points to a curve.  No control points = no pinning.  The curve control points can be positional, as with the "Pin Upper Right" and similar commands, or based on a color pin, as in pinning a neutral.

[ggroess]

Mike how about Contrast Pinning?  on the menu instead of double pin or "Worm Trap"?  You are trapping the worm after all..

Greg 

[-default]

Trapping the worm - LOL.

[ggroess]

sorry no double entandre entended..

[-default]

Sorry - the phrase just wiped out any chance that I could consider an otherwise excellent idea.

[ggroess]

I had my wife look at it with fresh eyes having little understanding of what she was seeing.  I asked her for a name for the process..  she suggested..contrast pinning..or worm lock...

If that helps...

Greg

[-default]

I have no problem with contrast pin - I'll add it to the manual.  I still like double pin too, though, and will probably keep it that way on the menu.

I'm still interested in the problem you found with not always being able to rotate the pins after you create them.  Thinking a little about it, is it possible that there were other control points on the curve beforehand, that were preventing some of the pins from being created?

I also noticed a problem with double pinning aka contrast pinning that could cause the problem you are describing.  The curve is currently pinned to the original image values, instead of the current curved values.  So if you had a curve in place, it gets distorted by the double pinning.  If that distortion would create negative slope areas, one or more of the new control points will get vetoed, and you would not be able to do the rotate.

[ggroess]

I know this cross two topics and I'm sorry for that but here is more info on the contrast pin function problem...

Seems to be pretty consistant...

1) Create a mask with Skin.
2) Switch to L in Lab and select an area to work on with a double pin, no other pins in place.
3) Try to adjust the rotation by clicking at the bottom of the curve window.

Moving to the right the curve rotates. It will return to the center if you move left but not past the center.
Move the mouse to the top of the curve window and the rotation is bi-directional. Crosses the center in both directions

So..From the bottom of the frame shown the image will rotate with a mouse move to the right but not left...
Bi-directional from the top of the frame.

Greg

[-default]

Once you click, the motion has been changed to be a rotation instead of a linear motion, similar to the crop tool in Elements.  So you can move the mouse around a circle, and change the slope of the curve by large amounts as you move.

Does this address what you're seeing?