Cant loft

Next
 From:  niko (NICKP100)
3592.1 
I'm starting to think that NURBS are a lot smarter than I am, and if it wasn't for MOI I would have gone back to SUB'Ds.
I have two problems.
1. The attached fuselage from the resulting loft, upon close inspection, shows a lot of creases along the aft end on the bottom.(yes even on high settings those creases do appear on the final render in Modo)
Why is that??? I rebuild all of my curves to have equal points. How can I get a smooth result?

2. On my new laptop(i7 2.66 Macbook pro with 330 gt) the edit frame graphic appears to be broken. I'm using paralells and it worked like a charm under paralells on my old Macbook pro. Is this fixed in the final release maybe?
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:  niko (NICKP100)
3592.2 In reply to 3592.1 
I think the divide larger than field helped with resulting mesh. I wasn't using that option at all.
I set it to 10 and the mesh was fine although very heavy...
Is there anything else that I'm missing maybe?
  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
3592.3 In reply to 3592.1 
Hi niko,

> 1. The attached fuselage from the resulting loft, upon close
> inspection, shows a lot of creases along the aft end on
> the bottom.

Could you please indicate where you are seeing the problem?

My guess is that you are seeing the result of too coarse of a tessellation in those areas, you will probably need to use the "Divide larger than" setting to make your object be divided up into some smaller polygons. When I use "Divide larger than" of 10 with your model and have the angle cranked up all the way, I get a result that looks like this:



There are some kind of subtle waves in the shape there and if you do not get enough polygons in those areas it can get a slightly angular appearance, you end up seeing some of the underlying faceted nature of the polygons.


> 2. On my new laptop(i7 2.66 Macbook pro with 330 gt) the
> edit frame graphic appears to be broken. I'm using paralells
> and it worked like a charm under paralells on my old Macbook
> pro. Is this fixed in the final release maybe?

No, not likely - as far as I know there isn't any bug in MoI for me to fix with the edit frame graphic, you are most likely running into some problem with Parallel's Direct3D interface there.

Until the recent versions of Parallels, their Direct3D interface was really very buggy and problematic. It has improved a lot in recent versions but it looks like you are running into some residual problem with it.

Make sure you are running the latest version of Parallels.

- Michael
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
3592.4 In reply to 3592.2 
Hi niko, it looks like you figured it out before I even posted!

> Is there anything else that I'm missing maybe?

No - just try to keep in mind that polygons are actually little flat faceted things and it is only through some kinds of shading tricks that they give a kind of emulation of a smooth appearance (by blending colors and/or shading normals between their vertices).

If there is some subtle variation in the vertex normals, but a relatively sparse set of polygons in that particular area (or sparse in just one direction with some skinnier polygons which is more what you were running into), the fake illusion of smoothness can be more easily broken. The solution for such things is to produce more polygons so that the illusion of smoothness holds up better.

Long and skinny polygons will have things blending between their vertices across a larger portion of the model, so they can have a tendency to make some more visible artifacts.

- Michael
  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
3592.5 In reply to 3592.2 
Hi niko, also is there some reason why you are particularly concerned about making a low polygon result?

You write:
> I set it to 10 and the mesh was fine although very heavy...

It's actually not really very heavy - using angle = 3, and Divide larger than = 10, over here produces a mesh with 44,000 faces in it, which is actually not what I'd call in a "very heavy "category at all:




You're not running on an old 486 computer or anything, right? :) That's pretty light duty for any vaguely modern machine to handle.

If you're going to have something special like a whole fleet of thousands of different ones of these, then certainly you might want to think about the density a bit. But otherwise, you may be worrying a lot about something that you don't really need to worry about for just 1 object.

- Michael
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
3592.6 In reply to 3592.2 
Hi niko, also if you're getting things that look like they're artifacts from long and skinny polygons, you could try the aspect ratio control instead of "Divide larger than".

That will force divisions on polygons that are more than a certain width/height ratio. Like if you put in 10 for "Aspect ratio limit" any UV quad that has one side more than 10 times longer than the short side will get subdivided.

It's pretty helpful for your case for the problem areas, with making a lighter mesh, like this:






But really I would not hesitate so much to just use "Divide larger than" and dice everything up pretty finely.

- Michael

  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:  niko (NICKP100)
3592.7 In reply to 3592.5 
Thanks for your responses Michael.
Yes you are correct!
  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