Revision 351f10c by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadow: Fix, improve cascade shadow map UI. |
Revision 90efcd6 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadows: Add cubemap filtering and adaptive sample count. Filter size is constant in world space and not dependant of shadow resolution. The filter size is limited to the number of precomputed samples. |
Revision 314739b by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadow: Fix Incorrect shadowing after sun lamp far clip plane. Reject cascade sample if not in shadowmap volume. |
Revision 9fdf094 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadows: Filtering improvement. - Replace poisson by concentric samples: Less variance. They are sorted by radius then by angle. - Separate filtering into 2 blur. First blur is 3x3 box blur. Second is user dependant. - Group fetches by group of 4. |
Revision adeaf37 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Add Cascaded Shadow Map options. |
Revision cc6e97d by Clément Foucault September 10, 2017, 01:09 (GMT) |
Lamps: Remove uper limit of the buffer bias parameter. |
Revision 716f785 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Add Cascaded Shadow Map support with filtering. This brings some data structure changes. Shared shadow data are stored in ShadowData (in glsl) (aka EEVEE_Shadow in C). This structure contains the array indices of the first shadow element of this shadow "object". It also contains how many shadow to evaluate (to be used for Multiple shadow maps). The filtering is noisy and needs improvement. |
Revision 6c17348 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadow: Add high bitdepth option. This option is here for reducing the memory usage of shadow maps. Also lower bitdepth are quicker to process. |
Revision f46b908 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Expose Shadow filter size. |
Revision 32e9644 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Add Variance Shadow Mapping This is an alternative to ESM. It does not suffer the same bleeding artifacts. |
Revision e2603a6 by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Shadows: Add UI buttons for size and method Only one method is available right now. VSM and PCF are comming. |
Revision 8b7a83a by Clément Foucault September 10, 2017, 01:09 (GMT) |
Eevee: Refactor Shadow System - Use only one 2d texture array to store all shadowmaps. - Allow to change shadow maps resolution. - Do not output radial distance when rendering shadowmaps. This will allow fast rendering of shadowmaps when we will drop the use of geometry shaders. |
Revision 11a9434 by Campbell Barton September 9, 2017, 12:56 (GMT) |
Resolve T52687: Add node label shows as 'Unknown' Add type access method, need to extend to other types for now just get node UI working properly again. |
Revision 3c3d089 by Campbell Barton September 9, 2017, 01:05 (GMT) |
PyAPI: Fix mathutils freeze allowing owned data |
Revision 30d8829 by Campbell Barton September 9, 2017, 01:02 (GMT) |
Docs: mathutils docstrings |
Revision 90eb937 by Campbell Barton September 9, 2017, 00:44 (GMT) |
Cleanup: mathutils vector comments Use doxy markup & correct outdated info. |
Revision 3930e46 by Campbell Barton September 9, 2017, 00:10 (GMT) |
Correct last commit |
Revision 002cc6a by Campbell Barton September 9, 2017, 00:03 (GMT) |
Cleanup: Simplify SWIZZLE macro - Use indices instead of character args. - Use numbered macros instead of variadic args. Parsing using rtags used over 11gb of memory. While this should be resolved upstream (report as #1053), the extra complexity didn't give any real advantage. |
Revision 4265d85 by Ray molenkamp September 8, 2017, 20:24 (GMT) |
[msvc/make.bat] mention the msvc-2017 option in the make.bat help text. |
Revision ae41a08 by Sergey Sharybin September 8, 2017, 16:37 (GMT) |
Cycles: Attempt to work around compilation of sm_20 and sm_21 Disabled forceinline for those architectures, which seems to be compiling successfully more often. There might be ~3% slowdown based on quick tests, but better be rendering something rather than failing to compile kernels again and again. Those architectures will be doomed for abandon once we'll switch to toolkit 9. |
|
|
|


Master Commits
MiikaHweb | 2003-2021