Whenever I divide this curve using the shatter tool the curvature graph then gets messed up. Any idea how I can divide this thing up at specific locations without losing the constant curvature shown on the curvature graph?
Here are is the why:
I’m working with a curve that rotates vertically as part of a stair system. The curve was made with grasshopper curves, blends and the pufferfish rebuild curve component. (I do wish there was a way to get G3 or G4 blends within GH, but couldn’t find any)
The next step is dividing it up so that pieces can be manipulated for extrusion/lofting and then…hopefully down the line…fabrication. After shattering the curve, it doesn’t look so good anymore on the zebra test when extruding and offsetting. However, before shattering it looks fine.
What is “messed up” for you? The graph having denser hairs near the cuts? That is expected!
Your pre-shatter curve is uniform, with uniformly spaced control points.
But by shattering, as that function is lossless (in matter of position and tangency of the curve), more points must be added, and that’s why you see more hairs where more points are.
Extruding should have perfect results. I tried with your shattered curves and I can confirm so.
Offsetting would be wiser to do with original pre-shatter shapes.
But it is difficult to tell without seeing the particular case you are having.
But what you attached attached so far is totally expected, working as intended. Perfect result.
Note:
Your pre-shatter curve is already NOT a perfect circle, if seen from above.
If you expect that curve to re-gain perfect circular shape after an offset, you are making some mistake, then.
Thanks for clearing that up for me! I misunderstood the graph.
Also, I think I also misunderstood the zebra check. I extruded the curves and offset the resulting surfaces, and it seems that the zebra check is fine, right? It looks like there’s a hump when rotating the the surface, but it’s probably just my inexperience and a little but of optical illusion.
So if I’m trying to get a smooth surface (close to a cylinder but not a perfect cylinder), is this about the right way of getting there? Getting a smooth curve graph and then shattering the curve to work further with it.
Not sure… is it a bit “wavy”.
That could be because of low mesh quality of zebra settings, or because too many control points in the shape.
Few but well-placed points are much better than too many points.
I am no really understanding what shape are you trying to make and how are you creating that.
Are you taking you shatter results, then extruding them in Z and then offsetting (solid) the surface?
To achieve better/smoother surfaces I would do the split/shatter step as the last one. Splitting the solid, not the curves.
OK @maje90 working with the curve segments is ‘OK’ for the most part. I can see why you would say to work with the curve and then shatter/trim/whatever at the end.
Offsetting the curves presented some issues, especially offsetting in the negative direction for some reason.
To keep the curve graph smooth after a shatter and offset I ran the segments through ‘Simplify Curve’, ‘Fit Curve Smooth’, and then ‘Rebuild Curve’ (all from pufferfish)
