My question should be regarded in the category of quick concept modeling. I think I understand why it acts like that, but should there be also another quicker way?
On the screenshot, you can see a simple box with two filleted edges which produce creases. Creases prevent me from successfully move one edge.
In the video, you can see Rhino behavior when using gumball (which is usually great). But how one can make such modification to the shape on the fly?
To achieve this very simple effect I used ExtractSrf , modified top surface and then used Sweep2 commands to fill the gaps and Join to join everything. Which is a really long path if I would change my mind and want to modify angle or filet again.
Even after I tried something different and removed surface with the fillets and replaced it with one made with Sweep2 command, so there are no creases, it still produces strange results.
I would do it the following way. First, draw a line from side view where you want the new top shape to be (make sure that the “Project” snap is turned on). Then, turn on the solid control points and move each of them vertically using the Tab key to lock the direction of movement and snap them to the line. That’s it.
An alternative to this is to create a surface that will be the target shape, then use “Scale 1-D” to scale the model vertically until it fully goes through the surface, then use the latter to split the extrusion or use “Boolean difference”.
“Extrude face to a boundary” (part of the “Solid tools”) also may help in this situation, though it will create a separate polysurface stacked above the original model. Also, it requires a preliminary built surface above the model that will be used as a target shape.
Yes another way is to use “Shear” (part of the “Transform” tools) and edit the model from side view, but the problem with that tool is that it will modify the entire model, so after you do that you have to show the solid control points, select the bottom ones and move them manually or use “Set XYZ coordinates” to move them to their original location before the “Shear” command.
PS: The best solution (direct solid modification with the Gumball) is easier and quicker, but you have to use Rhino 8 where this functionality will be already implemented.
Splitting with a surface is ok, but not that cool as a quick method.
Shear works surprisingly good! It can be done with only top surface selected, so there is no need to readjust bottom. Pity, that with that method it is impossible to move edge for a certain distance.
Also, Shear is a little inconvenient command, especially in quick modeling when preferably you would like to stay only in perspective mode.
I wonder, who I will be, when the Rhino 8 arrives.
Hello - modifying polysurfaces ‘on the fly’, especially by manipulating faces is not an easy or particularly powerful thing in Rhino. In this particular case I’d pull the whole thing taller and then WireCut or BooleanSplit the top.
I’d be inclined to cut it with a plane, cap it, and move a face to set a specific height/length.
You might want to use ShrinkTrimmedSrf when you’re done to keep it tidy.
Thanks, I wonder why version with Sweep2 face instead of Fillets made bad results.
I would like to see a simple, quick command where I can specify which edge to stay on place (red) and which selected edges or points pull every other face borders (blue).
Is it naive vision?
In that simple geometry, selecting top face and rotating it, produces far more dramatic modifications to the geometry, (shown in the video) So I guess locking one edge shouldn’t be that complicated?
No, but, given how Rhino operates, it is difficult. Rhino knows about surfaces and breps and does not think about defining underlying, visually ‘obvious’ shapes - so when modifying as you ask, surfaces are refit and deformed, there is no reference to an underlying shape which has to maintain its integrity. In addition, surfaces are not planar - a lot of this push-pull modeling users ask for comes from SketchUp where everything is planar. There are of course sophistcated softwares that allow this type of thing on surface models but they have a different approach to how things are defined and kept track of in the file. It is not impossible, but it would require quite a re-think as to how objects are handled in Rhino. I’m not at all sure it would be compatible with Rhino’s very free approach - it does not care of you explode, extract, join, delete, point-edit, rejoin etc etc.
I know it is far more complicated and I just believe you, but at the same time, I see that in this particular example, geometry handles all crazy surface modifications with ease.
Probably, you think about a lot of possible troubles in the other cases. Maybe command like that would not work always (just as every other command does not) but in some simple cases, such as this quick concept modeling, it would (maybe).
I am not a fan of SketchUp, used it once, I jumped straight into Rhino.
