Flip normals (MoI to modo)
All  1-4  5-9

Previous
Next
 From:  MAT13
9271.5 
Nice to see familiar faces here :-)
Pilou tu continues à écumer les forums, ça fait plaisir
Oh hi Mark #theroom

(Mat = Matelot for those who know)

I thought I could tell the flipped faces on export, but actually no.. The LWO export - although better than OBJ - doesn't fix the issue..

Thanks for the offer Michael, i'll send you the file later today..
  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:  Frenchy Pilou (PILOU)
9271.6 In reply to 9271.5 
Le pompon rouge! :)

  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
9271.7 In reply to 9271.5 
Hi Mat, I got your model and sent you a reply.

There are a lot of issues with the model - it appears to have gaps of up to 0.02 units between many pieces so it's not going to be able to be cleanly joined.

The main way to avoid flipped polys is to have a well formed model that has every edge joined between 2 surfaces. If you have 2 surfaces just sitting next to each other but not actually having shared joined edges then that's what can easily result in different polygon normal directions between them.

When you've got a clean model with all edges joined between 2 surfaces, the whole piece will then have a consistent normal direction.

Unfortunately in your case here the model data being so inaccurate it's going to be hard to make it into a well formed result.

I also tested loading your file into Rhino and got the same problems there, with 3796 naked edges after attempting to join it all together.

Probably the first thing I'd recommend is to try and get the model in STEP (.stp) format instead of IGES. STEP format can contain edge joining information inside of it and so that may come across better.

The model quality seems very poor in general though, I just found another area with a gap of 0.07 units between surfaces.

You're not going to be able to get nice clean high qualiity output on messy inaccurate input basically.

- 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
9271.8 In reply to 9271.1 
Hi MAT, I examined the model some more and found some areas with very large gaps between surfaces like around 0.07 units. You just aren't going to be able to get a high quality mesh automatically generated from such a messy pile of individual surfaces.

But using STEP format instead of IGES is worth a try.

- 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:  MAT13
9271.9 
Thanks Michael, I brought up the issue to my client
  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  1-4  5-9