it looks like a curve description by Bezier. I think Moi uses Nurbs. Nurbs can describe Bezier-Curves, but you need access to the underlying math (knot-vector, weights ...). The incredible Max Smirnov has written a Script for the DeCastellieu-algorithm. I've never tried it, but you can find it here: http://moi.maxsm.net/startpos/24.
In general every change on the control points of a Beziercurve changes the complete curve - means no local manipulation possible - only start and end stands still if not moved.
At the end I didn't understood your problem already, could you explain it a little bit more please? Especially the values 0.75 and 0.576.
You can do Q1 by drawing a perpendicular line above that apex, the you can use the tan/tan perp/perp snap to find the apex. Then on a curve, if you turn on control points, when they are NOT selected, you can move parts of the curve where there is no control point.
Q2 can be acheived, by trimming your curves ends back, finding and placing Q1, then using rotate and a custom pivot point to get Q2. Then manually finishing the ends off again.
The better bet is to draw your curve with "through points" and just map out the dims first.
Hi tupu, maybe another couple of possibilities would be to use the size editor, which will activate if you click on the size area of the properties panel here:
Or you could also use the edit frame sizing grip to snap on to a reference line like this (click once on the corner grip to toggle between "size from center" vs opposite corner mode):
I would have wanted a fast way to find the point to transform on the curve but I am sure I can do it by one your suggestions.
I do have to reposition the other points which should be locked in their positions if I scale to snap the entire curve, but I'll check more tomorrow.
Hi tupu, probably for the case you show there I would just drag that one point over to the left and snap it directly onto the boundary, and probably move the neighboring 2 points to that boundary as well. With 3 colinear points the shape should hug to the boundary.
If you can post the 3dm file instead of only a screenshot, it might be possible to give better advice.
But even if I do find the exact point how do I pick that exact point from where to start moving the curve, as I can see it transform tool only moves whole objects not portions of the curve so I still have to handpick the point even if I can get it more precisely now thanks to you guys.
Lets say the guideline box is 156mm wide and the curve with it's mirror side combined needs to be that width after snapping the curve to it, I am still getting a width of 155,876mm for example. I can live with that but would be nice to get the exact width if possible.
On second thought maybe it's not possible right now since moving portion of curve even if I could get the exact point from where to move it would still alter position of nearby points so then I would need to readjust them and then that would give the curve a new apex point and so on :)
Basically the curve would need to have points on the curve itself to do this kind of precision work?
""""""""""Basically the curve would need to have points on the curve itself to do this kind of precision work?""""""""""""
So you have to really think about what you are asking.
The beautiful smooth curve has points and a definition. You "HAVE" to move points around that point, or you will either just flatten, or kink the curve.
You cant alter Q1 (or Q2 or both) without altering the points that came before it.... to illustrate that, run Rebuild on your curve, at say .001, then try to move just a small set of points near Q1....
If you want to precisely pick the Q1 you just found, break the curve right there, then alter the end points which are pickable. You'll have to also alter the points just next to the end points to keep tangency to the break.
tupu_size1.jpg
