New MOI user
All  1-3  4-6

Previous
Next
 From:  paulrus
4783.4 In reply to 4783.3 
Thanks guys.

I must be a little confused. Are you saying to name each of my objects, then take all those name objects, select them and then give them all a name to create a group?

What I'm trying to do is set up some rudimentary hierarchical structure to my objects so I can more easily navigate through them all. Basically I wanted to have each key as a named object, then put them into a parent folder called "keys".

As far as the fillet problem, it's exactly as you're describing. I'll attach an image to show you what I mean. I'm assuming it's just a matter of me coming from polygonal modeling and so I'm probably doing wrong.



That's just a quick example I threw together, but I see this effect often.

Thanks!

Paul
Attachments:

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
4783.5 In reply to 4783.4 
Hi Paul - currently there is not any way to group things using nested hierarchies, that is something that I want to add in v3.

But you can group things into different labeled sets - that's what doing the object naming can give to you.

So for example if you have some particular set of objects that you know you want to hide and show repeatedly in the future, assign that set of objects a name and then you will be able to go to the scene browser and hide/show/lock/select that object set using the scene browser controls to do it rather than clicking on the items in the viewport window.


> Basically I wanted to have each key as a named object, then put
> them into a parent folder called "keys".

You can do the "each key as a named object" part right now, but not the parent folder part.


> As far as the fillet problem, it's exactly as you're describing. I'll
> attach an image to show you what I mean. I'm assuming it's just
> a matter of me coming from polygonal modeling and so I'm
> probably doing wrong.

Could you please attach the 3DM model file instead of just the screenshot? It's really hard to analyze the geometric properties of just a screenshot - in order to figure out what might be wrong I really need to examine the model more closely than I can with a screenshot.

The particular thing I'd probably be looking for first in a case like you show there though is whether the edges you are filleting actually meet up smoothly with each other at those places where the fillets are not intersecting properly. They probably are not meeting smoothly there, and when 2 edges do not meet smoothly the fillet system will try to extend the fillets and intersect them with each other. If they meet at a really shallow angle like around 5 degrees or so it tends to be a difficult case to intersect well.

The solution for cases like that is to make sure your initial curves that you used to create the objects were actually smooth to one another instead of just eyeballed to be close to smooth but actually having a slight shallow corner between them.

So if you have the original curves you used to create that model it would also help if you could include those in the 3DM file as well so I could take a look at them too.

But my best guess that I could make just from the screenshot is that it looks like probably not enough accuracy in the initial curve framework - something like curves meeting at a shallow 5 degree angle to one another instead of meeting with a shared tangent. That kind of shallow angle kink can tend to make filleting a lot more difficult for the filleting engine and get that kind of results.

You basically want to spend some time on your initial curve framework to make sure it is high quality - filleting is a particularly sensitive area and it does not tolerate problems in the geometry very well. Among the many kinds of things that it does not like are when 2 fillet sections collide into one another at a shallow angle - it's better for things to either meet up smoothly or to meet up at a more well defined sharp angle rather than something like a shallow 5 degree angle which is close to being smooth but not actually smooth.

- Michael

EDITED: 12 Dec 2011 by MICHAEL GIBSON

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
 From:  Mike K4ICY (MAJIKMIKE)
4783.6 
Hi Paul,

If the Fillet command fails to work on parts of your geometry - you can try this (more manual) work around:



http://moi3d.com/forum/index.php?webtag=MOI&msg=4607.5


Basically, you would create a gap at the edge location by which you would construct the rounded-over corner surface with the Blend tool and other construction methods.
The gap can be created by using Boolean Difference to subtract a the result of a Swept operation. The resultant surface edges are (matched) and then Blended.

It's not a straight-forward and easy procedure since one: some surface edges on your geometry simply need to be addressed for error, and two: some edge curves on the gaps have to be Merged or Trimmed to be matched, and three: the Sweep object used to Difference your geometry to make the gap may have overlapping kinks and can't readily be used unless rebuilt themselves.

This is like using MoI brand duct tape to fix an inherent problem in your geometry, but can keep you from pulling out all of your hair when your stuck trying to get that Filleted edge you desire.
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged
 

Reply to All Reply to All

 

 
 
Show messages: All  1-3  4-6