Blender Git Loki

Blenderin Git "master"-kehityshaaran kommitit.

Page: 2894 / 5574

April 21, 2014, 03:13 (GMT)
Skin mark/clear: replace GHash with GSet
April 20, 2014, 21:12 (GMT)
Fix stupid typo that broke split normal shading in edit mode, when there was some deform modifiers.
April 20, 2014, 18:00 (GMT)
Fix incorrect ghash free
April 20, 2014, 16:50 (GMT)
GHash, EdgeHash: hint on unlikely branch

also avoid searching buckets for empty hashes
April 20, 2014, 16:48 (GMT)
Correct cast from recent bool changes
April 20, 2014, 16:02 (GMT)
Code cleanup: correct abs use

also minor cleanup to rotation code
April 20, 2014, 15:01 (GMT)
Code cleanup: style
April 20, 2014, 12:47 (GMT)
Nodes: make default group nodes slighly wider, to match other node width defaults.
April 20, 2014, 12:47 (GMT)
Fix error making datablock with fake user local, the user count would be wrong afterwards.
April 20, 2014, 10:56 (GMT)
Math Lib: faster version of rotation_between_vecs_to_mat3

Use axis/angle calculation but avoid calls to sin/cos
April 19, 2014, 21:51 (GMT)
BGE: Fixing memory leaks when deactivating Sensors, Controllers, or actuators.

Found by Coverity and confirmed with Valgrind.
April 19, 2014, 12:17 (GMT)
Math Lib: add shell_v3v3_normalized_to_dist and v2 version

bypass angle calculation to avoids (asin, sqrt, cos).
April 19, 2014, 12:10 (GMT)
Bullet: making bullet friction workaround more furureproof by using the __apple_build_version__ macro, TODO: check if problem persists with newer clang
April 19, 2014, 09:48 (GMT)
Minor change to recent commit (avoid transpose)
April 19, 2014, 05:52 (GMT)
Math Lib: minor optimization to barycentric_transform

replace rotation_between_vecs_to_quat with axis_dominant_v3_to_m3.
April 19, 2014, 05:36 (GMT)
Math Lib: add rotation_between_vecs_to_mat3

- behaves like rotation_between_vecs_to_quat
- avoids calling sin,cos calls (approx 1.6x faster).
April 19, 2014, 04:10 (GMT)
Math Lib: minor optimization for axis_angle_normalized_to_mat3
April 18, 2014, 13:22 (GMT)
Disable quicktime when in cmake's lite configuration
April 18, 2014, 12:40 (GMT)
Cycles: SVM optimization for mix shaders, to skip more code when the mix weight
for one of the input shaders is zero.

This gives about 5% speedup for koro_final.blend. In general this is important
so you can design shaders that run faster for shadows, diffuse bounces, etc, for
example by skipping procedural textures or even using a single fixed color.
April 18, 2014, 11:51 (GMT)
Fix cycles using acosf in panorama render when it should use safe_acosf.
Tehnyt: Miika HämäläinenViimeksi päivitetty: 07.11.2014 14:18MiikaH:n Sivut a.k.a. MiikaHweb | 2003-2021