Hi,
yep, you were right.

I think that ctrl-z to cancel the operation would be very nice, a lot of people are used to that option. Also, when moving objects and right clicking it would be nice to have the object return to its original position without activating the previously used tool.

there is another problem that I am having that I can't find another way to do, and its not a bug I don't think.

when I have a rail, made up of three joined parts, and then that rail is duplicated and moved up, and slightly changed, I then create a three part joined piece between the two rails, and I try to sweep that part along those two rails. It does the sweep but then it creates three separate parts that are not joined to each other at the points where the rail's pieces are. I can do in Rhino, but MoI does not like it for some reason.

Do you want me to upload the file? There is also another problem with the two rail sweep on the same part that I was making.

> Also, when moving objects and right clicking it would be nice to have
> the object return to its original position without activating the previously used tool.

Hi Val, do you mean if you right-click during the middle of object dragging, to cancel the dragging? I'll see if I can tune that up. For the moment use the Esc key instead, it will cancel the drag without repeating the last command.


> Do you want me to upload the file?

Yes, please - that would help clarify things.

But in general it is a limitation right now in MoI that a sweep along a multi-segment rail will process the segments individually.

For v2 I would like to make the one-rail sweep handle mitered corners, but I'm not sure about handling corners for 2-rail sweep. If I can see your example and how it behaves in Rhino it might help me understand what you would like it to do.

- Michael

Ok, I remade it. Also, in Rhino I have to use simple sweep to get the desired result if I open MoI file.

Hi Val, Actually I don't think that the simple sweep option in Rhino is intended to be used for this kind of thing.

At the first glance the shape in Rhino looks ok, but if you look closer you can kind of see that the sweep in Rhino is messed up, there is bulging and a kind of slanting happening in the sweep, causing some non-constant thickness. Even though it makes corners there, it is not an accurate sweep (below is the sweep with simple option in rhino):



When you compare to the 2-rail sweep in MoI, I think MoI gives much more proper results. The shape of the sweep is correct, it hasn't slanted or compressed it in an extra way at all, but there are no corners. Below is the sweep in MoI for comparison to the above:



One of the problems with this sweep is that the rail is not actually perpendicular to the plane of the start shape, this kind of gives it a slight slant throughout:



With rails that are exact copies of one another, I'd actually recommend using one-rail sweep for that instead, with the selected curve below as the profile, and the long one as the rail. Switch the "Twist" option to say Twist:Flat - that will cause the shape to only rotate around the zaxis and not rotate in other ways around the rail curve.



Then the limitation is that MoI won't miter the corners for you automatically, you need to miter them manually right now. This is done by extruding the ends so they punch through each other:



Then select the 2 extensions, run Edit / Trim, push Done since you want to cut them with each other, and then pick the parts to throw away. That will leave you with the mitered corners which can be joined together:



For those results in MoI I also edited the rail curve so that its tangent direction was exactly perpendicular to the profile's plane.

- Michael

thanks! That gives acceptable results. I will give it a try and continue working in MoI, until/if I run into something else, I will post here.