Blender Git Commits

Blender Git commits from all branches.

Page: 2553 / 2888

September 3, 2014, 10:27 (GMT)
Merge branch 'master' into terrible_consequencer
September 3, 2014, 09:45 (GMT)
Cleanup: No point in passing all the implicit solver arguments
individually.
September 3, 2014, 09:34 (GMT)
Cleanup: removed the unused olddV vectors from implicit solver data.
September 3, 2014, 09:21 (GMT)
Revert "Gooseberry feature request, allow select left-right strips
operator to select all"

This change has been in master all the time after all, exposed the
operation better in master.

This reverts commit 799f974668004b3d2b7873f51951eb22b5fcf2f3.
September 3, 2014, 09:20 (GMT)
Some more debug elements for hair collisions.
September 3, 2014, 09:17 (GMT)
Extended line/face collision near-check, to allow for distance margins.

The original BLI method for line/triangle intersection returns false
in case the line does not actually intersect, but in order to generate
repulsion forces we need to also handle contacts inside the margin.
September 2, 2014, 20:46 (GMT)
added weighted percentage, distance, angle
September 2, 2014, 17:31 (GMT)
Revert "Terrible consequencer: commit to test requested feature by gooseberryteam."

This reverts commit 982d045e624201b4028186971189caa17cacff74.
September 2, 2014, 17:28 (GMT)
Terrible consequencer: commit to test requested feature by gooseberryteam.
September 2, 2014, 17:23 (GMT)
Gooseberry feature request, allow select left-right strips operator to
select all strips independently of channel of the active strip.
September 2, 2014, 16:20 (GMT)
Fix crash when setting parent.
September 2, 2014, 16:03 (GMT)
Merge branch 'master' into terrible_consequencer
September 2, 2014, 13:03 (GMT)
Fix bounce/repulse calculation.
September 2, 2014, 12:13 (GMT)
Merge branch 'master' into temp_custom_loop_normals

Conflicts:
source/blender/makesrna/intern/rna_mesh_api.c
September 2, 2014, 12:10 (GMT)
Hair debugging: use "categories" (strings) for grouping debug elements
and support clearing for categories.
September 2, 2014, 10:49 (GMT)
Fix for hair collision detection: need to use the second point of the
timestep segment.

This ensures the distance for a collision pair is the one of the current
point position, and the response gets calculated accordingly.
September 2, 2014, 09:54 (GMT)
Filepath should include view even when rendering one view (if multiview is on)

When multiview is not used the filepath is as usual (e.g., Image.jpg).
When multiview is on, even if only one view is enabled the view is incorporated
into the file name (e.g., Image_L.jpg). That allows for the user to re-render
individual images.
September 2, 2014, 09:54 (GMT)
Refactor to use view name instead of view id

This was is partly from the original code review round. At that time
Brecht Van Lommel suggested using names for the API instead of integers
(as we do for RenderLayers). I went a bit further and changed *most* of
the code to use strings.

The only parts that still use view_id are the ones closely related to UserImage.

That simplify the code and should make the code easier to understand,
review and maintain.

I tested many (if not all) of the supported areas, and apparently things
are working. But given the size of the intervention I wouldn't be
surprised if something stopped working, so please report.
September 1, 2014, 16:49 (GMT)
Use repulsion forces in combination with the one-time penalty forces
in collision.
September 1, 2014, 16:47 (GMT)
Fix for crash when going into particle edit mode with baked hair
simulation.

Note that this currently generates an extreme amount of points, by
making a edit pathcache curve for each hair in every frame! But at least
doesn't simply crash now.
By: Miika HämäläinenLast update: Nov-07-2014 14:18MiikaHweb | 2003-2021