Blender Git Commits

Blender Git "fracture_modifier" branch commits.

Page: 121 / 130

November 25, 2013, 19:09 (GMT)
hole problem solved by 98 % or so, only 2 wrong triangles at whole car !!
November 25, 2013, 19:09 (GMT)
hole problem seems to be solved, need more tests still
November 25, 2013, 19:09 (GMT)
creating hole filling shapes separately from mesh
November 25, 2013, 19:09 (GMT)
successful test with clipping and car
November 25, 2013, 19:09 (GMT)
hole problem almost solved, need to deal with a special case still (wrong vert selected..)
November 25, 2013, 19:09 (GMT)
clipping works mostly, mesh has some leftover holes, hmm
November 25, 2013, 19:09 (GMT)
first steps towards mesh clipping
November 25, 2013, 19:09 (GMT)
fix for compound creation, was throwing unnecessary error there
November 25, 2013, 19:09 (GMT)
fix for wrong compound rotation
November 25, 2013, 19:09 (GMT)
fix for crash while changing contact dist, changing back collision condition
November 25, 2013, 19:09 (GMT)
improved refresh logic, can do updates with rigidbody constraint refresh mostly, which is called automatically on value changes as well
November 25, 2013, 19:09 (GMT)
small collision tweak for compounds (no implosions)
November 25, 2013, 19:09 (GMT)
added compound shapes and disabled collision between compound/compound and compound/shard, reduces the "explosions"
November 25, 2013, 19:09 (GMT)
windows NAN fix, better do this elsewhere...
November 25, 2013, 19:09 (GMT)
cache should work now mostly, but still need full refresh to do it properly, hmm, maybe reset the frame data...
November 25, 2013, 19:09 (GMT)
compound sim need still refresh constraints, cache works only partially because compound parents disappear from simulation (better deactivate them only or so)
November 25, 2013, 19:09 (GMT)
compound needs some work still, but becomes usable step by step
November 25, 2013, 19:09 (GMT)
still working on proper compound / constraint cooperation
November 25, 2013, 19:09 (GMT)
compound basic principle working, but very uncomfortable to control still, needs to be rebuilt once broken correctly
November 25, 2013, 19:09 (GMT)
first attempt for cell based simulation, not very good yet, needs improvement
By: Miika HämäläinenLast update: Nov-07-2014 14:18MiikaHweb | 2003-2021