March 31, 2016, 15:54 (GMT) |
Cycles: Fix wrong initialization order of mesh flags and object transform |
March 31, 2016, 15:46 (GMT) |
Cycles: Fix too many objects being considered an instance |
March 31, 2016, 14:47 (GMT) |
Cycles: Fix compilation on Win32 after bitscan commit Need to revisit utility headers a bit more carefully and perhaps move such utilities outside of simd-related headers. |
March 31, 2016, 14:41 (GMT) |
Fix memory leak when assigning driver Also added NULL-checks for RNA paths, though they might be a bit paranoid. |
March 31, 2016, 14:39 (GMT) |
Partial revert: Use only BVMFunction stub type as a handle for all backends. API functions are still distinguished by backend. |
March 31, 2016, 14:35 (GMT) |
Use the BVM postfix to distinguish 'Function' in bvm backend from future LLVM functions. |
March 31, 2016, 14:27 (GMT) |
Fix memory leak when calling new driver eyedropper Would return OPERATOR_RUNNING_MODAL for non-modal operator. |
March 31, 2016, 14:14 (GMT) |
Fix incorrect stub type name for texnodes. |
March 31, 2016, 12:07 (GMT) |
Remove the obsolete windows launcher |
March 31, 2016, 12:05 (GMT) |
CMAKE disable building the windows launcher, fix type in delayloading of debug dll |
March 31, 2016, 11:52 (GMT) |
Cycles: Name cleanup and some comments in BVH code |
March 31, 2016, 11:46 (GMT) |
CMAKE / msvc openmp, delay loading of openmp dll so we can set environment variable before it is loaded |
Revision b26e1ad by Martin Felke (fracture_modifier, fracture_modifier-master, temp-fracture-modifier-2.8) March 31, 2016, 11:19 (GMT) |
fix for occasional crashes with dynamic fracture while freeing the modifier crash happened with scene objects which had a freed rigidbody world but didnt set the pointer to NULL. |
March 31, 2016, 11:00 (GMT) |
Added llvm_engine files for managing global LLVM initialization and ExecutionEngine. |
March 31, 2016, 09:44 (GMT) |
Cycles: Commit hack which allows to render less samples on the poles This is more an experiment for now to see how it could improve render times without loosing too much quality. Implementation is quite hacky due to timing constraints here. Now it is possible to control how much samples will be rendered on the pole, so now regular Samples are controlling samples in the middle of the frame, and those samples are falling off to the pole samples using sine interpolation. Limitations: - Only works for final renders without progressive refine option. - It is per-tile samples, so tile size should be small enough to see an effect. This is not something even nearly considered for master, more like a research and test project. |
Revision 7befec2 by Martin Felke (fracture_modifier, fracture_modifier-master, temp-fracture-modifier-2.8) March 31, 2016, 09:27 (GMT) |
fix for loading splitshards to island blends and clusters properly was necessary to skip "empty shards" properly and keep cluster index |
March 31, 2016, 08:24 (GMT) |
Merge branch 'master' into cycles_panorama_experiments |
March 31, 2016, 08:22 (GMT) |
Cycles: Add easy to use spin lock primitive Currently unused, but will be handy for an upcoming changes. It'll also be nice to be able to do scoped_lock() for both Mutex and Spin, but currently it's not really easy to do, need some changes in typedefs and such, will happen as a separate commit. |
March 31, 2016, 08:06 (GMT) |
Cycles: Expose array's capacity via getter function This way it's possible to query capacity of an array, which then could be used for some smart re-allocation and reserve policies. |
March 31, 2016, 08:06 (GMT) |
Cycles: Avoid per-split memory allocation for the new references list |
|
|
|


Master Commits
MiikaH:n Sivut a.k.a. MiikaHweb | 2003-2021