Rhino modelling question

[iamcapitano]

Hey guys,
I'm giving learning rhino a go and I have one problem that i've tried to search an answer for ages! It's not a vray question but i know alot of you are rhino modelling pros, so someone hopefully will know!

To simplify my problem i'll give an example: When you make a normal sphere and then with a cutting plan you split it in half and then delete the cutting plane. There is obviously a new line (isocurve?) on the sphere, if i join the 2 halves of the sphere with the join tool, the sphere is joined but still has that dividing new line. Is there a way to make that line disappear, in other words, merge the 2 parts of the sphere and make it have the same isocurves as before splitting it? Thanks!

[frip51]

Re: Rhino modelling question

Hey guy,
yea, Rhino is a powerful tool - go for it!

To simplify my answer: nope.
I'm just wondering why you'd like to make it disappear at all, as it shouldn't disturb or distract in any case.
Isocurves are there to describe the surface - for the user's eye to comprehend.

A sphere's UV coordinates are defined by and longitude and latitude.
You'd need one longitudinal line to start with - this is the line you see. But, depending on your viewport settings, you'd also see it on a whole sphere.


If you were referring to the split line: it's just two halves and as long as you deal with different surfaces, you'll see the seams.
In case you want to undo your action, try the ...undo command!
Nah, just joking.
Unless you did use the split by isocurve command (or chose to shrink trimmed surfaces in the meantime) you could untrim the hemisphere using the untrim command.


Hope that brings a little light into the Rhino cave.

Cheers,
Matt

[iamcapitano]

Re: Rhino modelling question

heh thanks for the response!

And i see. The reason i asked was because i was using a tube and wanted a curvey end, so i crossed it through a sphere and with boolean-difference to make the end of the cylinder curved and cut it off where the sphere was. It worked but the created face of the curved part of the cylinder came out with 3 faces and multiple isocurves.

And the reason i had to split the sphere was to invert the normal directions so that boolean-difference worked like i wanted it to. But i found a workaround, by splitting the sphere at a part that didn't touch the cylinder.

It all worked but that made me wonder....... surely there must be a better way to invert the normals in a sphere than splitting it open, and inverting each half.

I'm loving rhino so far btw!

[frip51]

Re: Rhino modelling question


It's a pity, you can't just invert the normals on a sphere. Neither on any other closed volume. As soon as you join the surfaces, Rhino flips everything to the outside again.
That is a bit of a pain when it comes to displacement mapping. So, when dealing with an object that needs to be perfect from every angle, do the following:

- use the ExtractSrf command to loosen one (smallsurface
- select all and group
- run the Dir command and choose FlipAll to invert the normals direction
- when using displacement mapping, uncheck the Keep Continuity option

The latter is a little odd in my opinion and it took a while for me to figure out why displaced objects often featured broken seams. If V-Ray doesn't attempt to keep the surface continuity it works just fine. Maybe there's even a logical explanation for it...


btw: tubes with rounded ends are done best using the Pipe command (simply draw a line or curve and run Pipe)! There, you have an option to set the cap either to None, Flat or Round

And: Splitting cylinders with spheres or cylinders that have the same diameter results an unsolvable mathematical problem. If you're into cross vaults or crossed pipings you'd need to split the cylinder shapes with lines in top view.


Cheers,
Matt