odd behavior for UV's and SubD?

["DeColling, James" <James.DeColling(at)atari.com>]

Howdy, its been brought to my attention that the way 5.0 is handling merged UV's and Sub-d is a touch odd.

 

example -

 

CreatePrim "Sphere", "MeshSurface"

CreateProjection "sphere", siTxtSpherical, , , "Texture_Projection"

FreezeObj

ApplyTopoOp "DeleteComponent", "sphere.poly[16-47]", siUnspecified, siPersistentOperation

Duplicate "sphere", , 2, 1, 1, 0, 0, 1, 0, 1, , , , , , , , , , , 0

Scale , -1, 1, 1, siAbsolute, siPivot, siObj, siX, , , , , , , , 0

SelectObj "sphere,sphere1", , True

ApplyGenOp "MeshMerge", , "sphere,sphere1", 3, siPersistentOperation, siKeepGenOpInputs

TransferAllPropertiesAcrossGenOp "polymsh.polymsh.mergemesh", "polymsh", , True

DeleteGeneratorInputs "polymsh.polymsh.mergemesh"

 

(all it does it cut a sphere in half, merge tighter with overlapped uv's)

 

now when subdivided there is a stretching across the seam..is this how it should be? It seems to be culling the last few pixels of the texture map

 

now...local sub-divide the object and everything works as expected

 

 

any clues? Only way to currently preview multiple lod's with subd is to bake it down to local....and that kinda blows.

 

 

Cheers

 

James,

 

---

James De Colling, Senior Artist
Atari Melbourne House, 
Level 11, 14 Queens Rd, Melbourne, VIC 3004, Australia
  Fax: +61 3 9867-0800     Phone : +61 3 9867 0700
  melbournehouse.com       Direct: +61 3 9867 0726
james.decolling(at)atari.com     Mobile: +61 400 145 808

 

Attachment: ScreenShot006.jpg
Description: JPEG image