Shape Bending/Curved Scaling one side of an object
-
wouldn't it be better to use an intersect with model for this?
-
you might also try the control points plugin i just bumped too, but its not working for me right now.
-
@xrok1 said:
you might also try the control points plugin i just bumped too, but its not working for me right now.
Thanks. Which? Where? "Bumped too"?
-
@xrok1 said:
wouldn't it be better to use an intersect with model for this?
Ahh... I was afraid my example would be too simple.
Here's another example, this time with Sketchy FFD 2-09-10, with the control points moved to the curve manually. I used 24 segments for FFD and selected and moved groups of points symmetrically, from the ends in toward the center, so there were only 11 moves to do. The effect is not bad, except for the places where it looks as if SketchyFFD lost track of some edges and didn't move them, breaking up the shape.
I discovered that you can group control points, manipulate the group, and then when you ungroup the deformation is applied. But you cannot use FredoScale on the group because Fredo only works with edges and you cannot use Shape Bender on the group because Shape Bender makes a copy and that loses the connection (however it is maintained internally).
I hope this is clearer,
August
-
-
@august said:
"Bumped too"?
I grok "bumped". Makes sense if you are paying attention to that context. I usually don't.
-
there's gotta be an easy way c'mon people, i'm pullin my hair out here.
anyway, once thomthom releases his vertex editor (with soft selection! ) all will be well with this and many other challenges.
-
@xrok1 said:
there's gotta be an easy way c'mon people, i'm pullin my hair out here.
photoshop?
in SU, smoove can almost do it with it's soft selection and whatnot..
-
this would be a good one for TIG's extrusion toolset. extrude shape to (by
) rails. it would be like shape bender on steroids! OR it could also be a nice option added to the shape bender tool. -
Thanks for the enthusiasm Rocky,
@xrok1 said:
... c'mon people, i'm pullin my hair out here. ...
but dont' hurt yourself over it.I woundn't even be thinking about it as a possibility if all these other wonderful tools hadn't set my expectations so high.
And I'm certainly not the programmer to do it myself.
I have not tried "TIG's extrusion toolset" but your phrasing suggests it's another one that does not quite go this far.
I'm just trying to work out a design idea. I have manual ways to try it out in rough form. If I get hooked on one that takes some tedium to execute in finer detail, I'll just put on some music and get into a "Tower of Hanoi" rhythm.
It looks from the PhotoShop screen shot like Adobe calls that particular morph "Fisheye". I had thought about using that term myself, but I didn't because I thought it would unnecessarily predispose people to thinking in terms of 2D-like solutions.
I like your notion of "Shape Bender on steroids". SB is still at Beta 0.55, so I might imagine this for SB 2.0. Where SB 0.55 has a single Guide line and single Target Curve, I could see SB 2.0 allowing two or four (or N?) matching Guide Lines and Target Curves. (1. Select Source Group. 2. Activate SB. 3. Select single lines or group of N lines. 4. Select Target Curve or group of N Target Curves. -- How do you specify the order of the lines and the Target Curves? Like most such development projects, if you can design a UI that lets the user intuitively and unambiguously specify their intent, you're part way there.)
That would be a lot like being able to apply SB or FredoScale to the guide points of SFFD. From a programming standpoint, extending SB to multiple Guide Lines seems like a smaller conceptual jump than extending FS to be able to move SFFD's guide points, but maybe "all" it would take is getting FS to handle nested groups. Maybe extending SFFD to align guide points to a target curve, like SB, would be the easiest of all.
But of course, I don't know the internals of any of these, nor how much interest any of these developers has in going out into the ionosphere here. Comparing relative "ease" of these tasks is irrelevant when there are three different volunteers involved. (I can see it now, this year's Ruby extension challenge, submissions due 12/31.
)Thanks again for jumping in, Rocky. At least I know I'm not missing the obvious (as I too often do).
August
Advertisement