When doing DOPs simulations with an RBD fractured object, what is the general consensus on how to disguise the cracks in prefractured geometry before the geometry is meant to break apart? I have been told that adding a fuse SOP to the object merge which merges your simulated geometry into SOPs gives the illusion that your geometry isn't cracked. While this seems to work in some situations, in many it does not.
For example I have prefractured a grid into a number of triangles and then dropped the grid on the floor in DOPs. It smashes apart nicely. Now I want to render it out I slapped a VEX Glass on it and set up some GI. The trouble is I can see all the triangles as the glass sheet is falling, even when I try adding a fuse SOP. Is there any other methods of disguising these prefractured elements?
Disguising prefracturing when rendering out fractured object
11208 9 3-
- Dean_19
- Member
- 322 posts
- Joined:
- Offline
-
- djorzgul
- Member
- 235 posts
- Joined:
- Offline
-
- Dean_19
- Member
- 322 posts
- Joined:
- Offline
-
- keyframe
- Member
- 1533 posts
- Joined: July 2005
- Offline
Heya Dean,
A trick I sometimes find usefull, is to calculate the normals on the solid geometry, and store then in an attirbute (Nconsolidated) – then “shatter” the geometry using some method, and simply transfer the Nconsolidated attribute back to the shattered normals.
This will render the object as if it weren't shattered (assuming of course that you tranfer Nconsolidated to N).. and now all you have to do is figure out how to blend between the fractured normals, and the solid ones… based on time, or possibly based on the distance of the centroids of each piece to each other relative to the whole “unexploded” mesh.
Hope this helps…
Cheers,
G
A trick I sometimes find usefull, is to calculate the normals on the solid geometry, and store then in an attirbute (Nconsolidated) – then “shatter” the geometry using some method, and simply transfer the Nconsolidated attribute back to the shattered normals.
This will render the object as if it weren't shattered (assuming of course that you tranfer Nconsolidated to N).. and now all you have to do is figure out how to blend between the fractured normals, and the solid ones… based on time, or possibly based on the distance of the centroids of each piece to each other relative to the whole “unexploded” mesh.
Hope this helps…
Cheers,
G
-
- keyframe
- Member
- 1533 posts
- Joined: July 2005
- Offline
Heya Dean,
A trick I sometimes find usefull, is to calculate the normals on the solid geometry, and store then in an attirbute (Nconsolidated) – then “shatter” the geometry using some method, and simply transfer the Nconsolidated attribute back to the shattered normals.
This will render the object as if it weren't shattered (assuming of course that you tranfer Nconsolidated to N).. and now all you have to do is figure out how to blend between the fractured normals, and the solid ones… based on time, or possibly based on the distance of the centroids of each piece to each other relative to the whole “unexploded” mesh.
Hope this helps…
Cheers,
G
A trick I sometimes find usefull, is to calculate the normals on the solid geometry, and store then in an attirbute (Nconsolidated) – then “shatter” the geometry using some method, and simply transfer the Nconsolidated attribute back to the shattered normals.
This will render the object as if it weren't shattered (assuming of course that you tranfer Nconsolidated to N).. and now all you have to do is figure out how to blend between the fractured normals, and the solid ones… based on time, or possibly based on the distance of the centroids of each piece to each other relative to the whole “unexploded” mesh.
Hope this helps…
Cheers,
G
-
- Dean_19
- Member
- 322 posts
- Joined:
- Offline
-
- pbowmar
- Member
- 7046 posts
- Joined: July 2005
- Offline
-
- Dean_19
- Member
- 322 posts
- Joined:
- Offline
-
- craig
- Member
- 557 posts
- Joined: July 2005
- Offline
-
- Dean_19
- Member
- 322 posts
- Joined:
- Offline
-
- Quick Links



ops: 