What about the “Pipe” tool, where the arc segments don’t use a proper revolved circle profile? Instead, it builds an extrusion with variable diameter and a ton of unwanted control points. It’s super annoying to delete and manually build revolved surfaces where the “Pipe” tool fails to do so.
Another huge problem with the “Pipe” tool is that it does not make simplistic pipes with the minimum amount of control points. Those have to be removed additionally, breaking the History…
Here is one video showing how I fix that bug by using a custom macro to set the axis for the subsequent “Revolve” operation:
Macro 1: Axis from arcs and circles: ! _CPlane _Object _Pause _ProjectOsnap _Toggle Enter _Pause _Line _Vertical 0 _BothSides _Pause _Pause _Cplane _Undo _ProjectOsnap _Toggle Enter
Creates a line which represents the aligned axis of a true arc or circle. Made to work no matter what’s the orientation of the input arc or curve. Starts from the center of the arc or circle and goes in both directions.
Useful if you need to create a true radius pipe via “Revolve” from the end of an existing straight pipe.
Note that the current implementation of the “Pipe” tool creates cylindrical extrusions with 4 rows of control points along the length, whereas the desired behaviour is to generate extrusions with just two rows of control points at the start and end of each extrusion.
Ideally, a Command line option to switch between the current and the cleaner output extrusions would suit the needs of all users.
Extrusions were known to cause some troubles with the modeling in Rhino, this is why I keep them turned off since years. I see no reason for the extruded cylinders to have two extra rows of control points. That forces the user to delete those manually, hence the History recording gets broken.
Alias and NX had a go at those kinds of surfaces, but they never really work in terms of quality. If you surface model properly, you don’t need those “crutch surfaces” anyway : )
Knowing these kinds of stuff and building it properly you can make some really smooth clean quality surfaces I have seen
So you won’t be using it much the patch in rhino 9 I take it? Does it produce bad results for the cause?
I don’t take much into account these rules when I do things but they do help sometimes understanding it to avoid or fix any issues, helps with problem solving.
im always interested hearing from other fields and views as one can always learn from it
I just fail to see where the “botch” surface would be useful (once you found a suitable patch layout for the product you want to surface model) other for some last-minute hole/gap fixing: “Get that model out to 3D printing/CNC milling, dude!”. In such cases, it’s a good solution, because it’s the only solution : )
See this forum for typical problems. Here, the user was probably a novice without industrial design education, who thought that one can draw “shape outlines” like in Adobe Illustrator with a “fill”. Here, the user was also falling into the “shape-outline-‘n’-fill” trap. Or here, another one, where the user very likely did not patch layout his object with primary and secondary surfaces first, before starting to model.
Well, let’s wait until 2026 or 2027, when V9 is officially launched. I would not deploy an unstable WIP to users. From the little I saw of V9 so far, I fail to see any practical use case for industrial designers, apart from hot-fixing a model, as before.
3D printing still suffers from a low surface finish quality and problems of isotropy due to layer-based build-processes, and much post-processing is necessary. You can easily correct surface modelling glitches in post-processing, as with rough CNC-milled pre-prototypes. Researchers in Sweden have written a very comprehensive book about the current state of affairs of 3D printing as part of Springer’s series on advanced manufacturing.
Another much needed option for the “Pipe” tool is to be able to set the seam orientation, especially when using the “FitRail” option.
Obviously, in cases where the input geometry is made by simple lines and arcs randomly in the 3d space (non-planar ones), the seam of each pipe could be oriented locally. For example, the seam of each arc pipe could be set at the minimum inner side (the best solution from a manufacturing perspective, such like laser cutting or chassis design) or the maximum outer side, whereas the seam of the straight pipes could match that of the adjacent arc pipe.
However, this is impossible if a straight pipe is adjacent to two arc pipes rotated at a different angle each.
Here is an example of the latter case where the middle straight pipe can’t inherit the same seam like the two adjacent arc pipes due to their different relative angle of rotation.
One possible solution is to force each straight pipe’s seam to match the seam of the adjacent arc pipe’s seam in a consecutive way.
