git.blender.org/blender.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Collapse)	Author
2009-09-28	Raycounters fix	Andre Susano Pinto

2009-09-25	*Added VlakPrimitive (this rayobject rimitive only stores ↵	Andre Susano Pinto
	ObjectRenderInstance and VlakRen pointers) - it difers from RayFace that localy stored the vertex coordinates. - basicaly this reduces memory usage
2009-09-14	*Ray counters (number of BB/primitive tests/hits and other raytrace ↵	Andre Susano Pinto
	counters) can now be enabled/disabled at compile-time. #define RE_RAYCOUNTER (/source/blender/render/extern/include/RE_raytrace.h) *Some other small organization on code (will be disable as this only matters for testing and developping)
2009-09-06	*introduced new method for packing/optimizing trees after building	Andre Susano Pinto
	(this is a generalization of some of the experimental stuff i tried during SoC, but only had time to improve a few days ago) - it should yield slightly better results - the cost model can somehow be tweaked to optimize for diferent trees. cleaned up some code added counters for number of SIMD BB tests *added GPL license block on missing files
2009-08-29	Code reorganization	Andre Susano Pinto
	-separated vbvh, svbvh, qbvh in diferent files (before the only way to switch between them was at compile time)
2009-08-26	*Changed RayObject_ calls to RE_rayobject to keep consistency on calls	Andre Susano Pinto
	*Moved part of counters code to a separated file (rayobject_raycounter.c)
2009-08-14	Addition of some fake nodes to use SIMD even when theres only 3 nodes	Andre Susano Pinto

2009-08-13	Another tree pass during build to increase the number of nodes that have ↵	Andre Susano Pinto
	multipe of 4childs
2009-08-11	fix instance support when using SIMD	Andre Susano Pinto

2009-08-11	*Added a tree structure with a variable number of childs per node, but with ↵	Andre Susano Pinto
	groupped childs (for SIMD) SIMD support for the first 4N childs of each node *Some bvh code organized
2009-08-11	Ability to disable hints at compile time	Andre Susano Pinto

2009-08-07	Made -O3 and -msse2 default flags for bf_render_raytrace	Andre Susano Pinto

2009-08-07	*BLI_memarena support for any power of two alignment	Andre Susano Pinto
	*some simd stuff on bvh
2009-08-07	Fix _MM_SHUFFLE_ order	Andre Susano Pinto

2009-08-07	improved SIMD on raytrace (up to the moment support of SIMD is done at dfs ↵	Andre Susano Pinto
	and on any type of tree) (it only shows worth on -O3 -msse2) because it seems gcc makes horrible asm code on -O2
2009-08-07	usage of simd on bb tests	Andre Susano Pinto

2009-08-06	*Process leafs as soon as found instead of pushing them on stack for later ↵	Andre Susano Pinto
	evaluation (leads to early exits) (this is mixed with some simd code commit, althouth no simd is being used up to the moment)
2009-08-06	no need to calculate the exact nearest distance if we are not using any ↵	Andre Susano Pinto
	heuristic based on that
2009-08-05	generic hints for raytracer	Andre Susano Pinto
	for now only BB hint (i am hoping to find a decent frustum-test)
2009-08-04	Fix point-hint	Andre Susano Pinto

2009-08-04	Skip BB tests on primitives	Andre Susano Pinto
	the efficiency of this depends on ray-bb and ray-triangle functions efficiency
2009-08-04	single tree (by default)	Andre Susano Pinto
	now that build is nlogn (it should be worth to the tree of trees)
2009-08-03	NlogN building:	Andre Susano Pinto
	sort once select subsets and kept the order (on X, Y and Z)
2009-08-02	changed to STL sort	Andre Susano Pinto

2009-07-17	Another try with building better trees (this should never make worst trees)	Andre Susano Pinto
	Expected number of BB tests should reduce a bit (depending on the scene)
2009-07-15	*increased stack size (later this should be prepared for dealing with stack ↵	Andre Susano Pinto
	size in runtime) *put cost model back to normal
2009-07-15	*Added support to "BB hints" (which works like a BB version of LCTS - ↵	Andre Susano Pinto
	longest common transversing subtree) It creates a tree cut after knowing that a given point will pass on a BB. This tree cut is used to accelarate the rays casted from a given BB, eliminating unnecessary BB tests from root till the tree cut.
2009-07-15	*fix (was losing childs)	Andre Susano Pinto

2009-07-15	Just another experimental stuff to optimize the expected number of BB test ↵	Andre Susano Pinto
	on bvh trees *tree pushdowns after the pushsups :P (its still not local optimum)
2009-07-12	*Moved rtbuild to bf_render_raytrace	Andre Susano Pinto
	*Added vbvh - Just a experimental tree type :) Variable Way BVH - there is no hardcoded number of childs per each Tree Node - idea is to optimize a tree to reduced the expected number of BB tests even after applying SAH (for that an hardcoded n-way is not enough) - for now childs are stored on a linked list
2009-07-12	I had applied a patch twice.. code was duplicated	Andre Susano Pinto

2009-07-12	Added module bf_render_raytrace (source/blender/render/intern/raytrace)	Andre Susano Pinto
	to be able to use C++ at raytrace code C++ used in here is basicly C with templates and function overloads, to make it easier to reuse code between structures. For now BVH was converted in C++ and moved to this module