Commit Graph

1186 Commits

Author SHA1 Message Date
barry
d84666c829 Populate Far::PatchTable with triangular patches for Loop scheme:
- extended patch construction in base PatchBuilder to support triangles
    - added extraction of regular triangular patches to base PatchBuilder
    - converted irregular patches to supported patch types in LoopPatchBuilder
2018-10-11 17:51:56 -07:00
David G Yu
c5c9704b90
Merge pull request #1003 from barfowl/tri_patch_support
Added low-level support in Far for triangular patches
2018-10-03 17:44:01 -07:00
David G Yu
2fe36033d5
Merge pull request #1002 from barfowl/vtr_level_tag_bits
Simplify Vtr bitfield operations to avoid issues with older versions of GCC
2018-10-01 13:05:17 -07:00
David G Yu
521fc52e70
Merge pull request #1001 from barfowl/osd_cpu_patch_table
Corrected initialization of patch arrays in Osd::CpuPatchTable
2018-10-01 13:04:20 -07:00
barry
d6572ab226 Updated Osd evaluators to deal with changes to PatchParam boundary:
- updated evaluators to extract the extended 5-bit mask from PatchParam
2018-09-27 20:13:01 -07:00
barry
e96bf355e0 Added low-level support in Far for triangular patches:
- fixed missing PatchDescriptor functionality for Loop triangle
    - add new PatchDescriptor::Type for Gregory triangle
    - extended PatchParam boundary mask from 4 to 5 bits (!)
    - added new PatchParam methods for parameterization of triangles
    - extended PatchMap to support triangular patches
    - added basis evaluation for triangular patch types
2018-09-27 20:09:59 -07:00
David G Yu
54346afaa9
Merge pull request #999 from barfowl/tri_adaptive_refinement
Remove Catmark-only restriction on adpative refinement in Far
2018-09-26 14:39:59 -07:00
barry
99f6a27604 Simplified access to Vtr::Level tag bits for integer operations:
- added alternate/explicit constructor to initialize from integer bits
    - added accessor to retrieve collections of bits as an integer
    - replaced pointer aliasing in bitwise combinations with use of new methods
2018-09-25 15:51:01 -07:00
barry
89c4ce78b7 Corrected initialization of patch arrays in Osd::CpuPatchTable
- assigned appropriate patch count and base offsets to varying arrays
    - assigned appropriate patch count and base offsets to face-varying arrays
    - fixed append to face-varying PatchParam buffer for each array
2018-09-25 15:13:55 -07:00
barry
fe0b397e2f Removed assertion restricting adaptive refinement to Catmark scheme:
- split face tag analysis into functions for face and individual corner
    - avoided forced boundary isolation of triangles, preserved for quads
2018-09-25 14:29:35 -07:00
David G Yu
91478f149a
Merge pull request #995 from barfowl/fvar_patch_array_dev
Extended face-varying patch arrays to properly support two types
2018-09-19 17:29:01 -07:00
David G Yu
c732ff346a
Merge pull request #994 from barfowl/compute_local_points
Reverted PatchTable local point methods to address template issues
2018-09-17 17:54:47 -07:00
barry
b542bc78c3 Added missing include of <algorithm> for std::max with stricter compilers 2018-09-17 11:46:41 -07:00
barry
86e3712564 Fixed oversight deriving Far::LimitStencil from new template class:
- added the "public" keyword that was inadvertantly omitted
2018-09-17 09:33:08 -07:00
barry
45928c1b81 Extended face-varying patch arrays to properly support two types:
- extended Far::PatchTable's FVar channels with two descriptors and stride
    - updated Far::PatchTableFactory to properly construct FVar patch arrays
    - extended Osd::PatchArray with two descriptors and common stride
    - updated Osd::CpuPatchTable and CpuEvaluator to use PatchArray extensions
2018-09-13 16:51:53 -07:00
barry
5c9f77d83e Reverted PatchTable::ComputeLocalPointVaues...() to avoid template issues:
- reverted original methods to operate in single precision instead of both
    - removed overloads in favor of direct stencil table access and update
    - updated precision related assertions with clearer condition
    - updated far/tutorial_6 to use template stencil table access
    - updated far/tutorial_9 to use conventional stencil table access
2018-09-13 12:50:00 -07:00
David G Yu
79476c87cb
Merge pull request #991 from barfowl/val2_gregory_simplify
Simplify construction of Gregory patches with val-2 interior corners
2018-09-13 12:29:06 -07:00
barry
725d3c585b Simplified Gregory point initialization with more direct assignment:
- replaced use of Append() with Assign(index) now that row size is clear
    - added asserts for all row sizes to ensure assignment matches allocation
    - simplified utility class for SparseMatrix point/row down to bare minimum
2018-09-12 16:01:03 -07:00
barry
e0457241e8 Simplified Catmark->Gregory conversion with val-2 interior corner:
- replaced the invasive valence-2 conditions with a post-process
    - removed inefficient and now unnecessary Point::AddOrAppend()
2018-09-12 15:52:59 -07:00
barry
f58330fdee Minor extensions to Far classes to help manage large PatchTables:
- overloaded PatchTableFactory::Create() to take subset of base faces
    - overloaded TopologyRefiner::RefineAdaptive() to take subset of base faces
    - added TopologyRefinerFactory::Create() to create new TopologyRefiner
      that shares base level with another, allowing independent refinement
    - overloaded PatchTable::ComputeLocalPointValues...() to accept split
      buffers with separate base-level primvar values
    - overloaded StencilTable::UpdateValues() to accept split buffers as above
2018-09-11 16:13:06 -07:00
David G Yu
c566e8f2ab
Merge pull request #988 from barfowl/sparse_matrix_additions
Minor improvements to internal Far::SparseMatrix class
2018-08-29 18:22:35 -07:00
David G Yu
bb826b0f7d
Merge pull request #987 from barfowl/vtr_indexing_bug
Fix rare non-manifold topology indexing issue in Vtr::Level
2018-08-29 18:22:18 -07:00
barry
86dd8cc912 Bug fix and method addition to Far::internal::SparseMatrix:
- fixed error in Copy() causing compilation failure on instantiation
    - added new Swap() method analogous to std::vector::swap()
2018-08-27 11:02:36 -07:00
barry
a79a12269b Fixed rare non-manifold topology construction bug in Vtr::Level:
- use address arithmetic instead of std::vector[] to determine vector end
2018-08-27 11:00:37 -07:00
barry
a154426d33 Added new flags to Far::PatchTableFactory::Options to:
- address indexing inconsistencies with Uniform PatchTables (#737)
    - suppress generation of Varying patches and/or their local point stencils
    - added suppression of Varying patches to far/tutorial_6
2018-08-27 10:58:46 -07:00
barry
0f63aa997d Added bilinear end-cap option to Osd::MeshBits 2018-08-06 16:29:07 -07:00
David G Yu
148d3005d3
Merge pull request #982 from barfowl/stencil_table_append
fix Far::StencilTable append when base StencilTable empty
2018-08-06 12:10:48 -07:00
barry
af4b1e3fec Fixed StencilTableFactory::append() when base stencils empty:
- duplicate the local point StencilTable when no base stencils
2018-08-05 22:06:56 -07:00
barry
5b9e986d62 Fix a couple oversights in recent changes supporting double precision:
- corrected Far::LimitStencilTableFactory wrapper to use StencilTable
    - promoted constants in Far::CatmarkPatchBuilder to double precision
2018-08-05 21:31:45 -07:00
barry
647039a8ee Major public addition of class templates for multiple precision:
- converted classes for stencils, tables and factories to templates
    - added backward compatible class wrappers around new templates
    - extended PatchTable to support variable precision stencil tables
    - extended PatchTableBuilder to manage PatchTable's new stencil tables
    - templated PatchTableBuilder conversion from SparseMatrix to stencils
    - extended public PatchTableFactory interface to specify precision
    - converted PrimvarRefiner class to template and added wrapper class
2018-07-28 14:56:43 -07:00
barry
f6172f9608 Minor public extensions to support multi-precision evaluation of patches:
- templated PatchParam normalization methods for multiple precision
    - instantiated double precision basis evaluation templates
    - converted PatchTable::EvaluateBasis...() methods to templates
        - added overloaded versions to assist template parameter resolution
    - extended PatchMap::FindPatch() to double precision
2018-07-28 14:43:32 -07:00
David G Yu
cdcec73088 Fixed Far::PrimvarRefiner::limitFVar prototype
Fixed the method signature of the internal limitFVar method to make
the dst argument a reference rather than a pointer to be more consistent
with earlier changes for other primvar classes. This also makes it
easier to use the public template LimitFaceVarying method to operate
on a wider range of primvar container types.
2018-07-28 13:37:03 -07:00
barry
a0fcfb54f9 Reordered #includes in far/catmarkPatchBuilder.cpp to match convention 2018-07-27 12:07:31 -07:00
barry
9b2d161c4e Internal numerical extensions to retain double precision:
- retained precision of constants in Sdc Catmark scheme
    - improved precision of constants & intermediate values in Sdc Loop scheme
    - improved precision of constants & intermediate values CatmarkPatchBuilder
    - converted all internal Far patch basis evaluation to templates
2018-07-23 11:10:56 -07:00
barry
f15d4ae02f Internal refactoring to support multiple precision:
- overloaded PatchBuilder to populate SparseMatrix as float or double
    - simplified PatchTableBuilder handling of matrices and Varying stencils
2018-07-23 11:00:35 -07:00
barfowl
7a4fe81351
Merge pull request #975 from Nazg-Gul/fix_topology_refiner_stack_memory
Thanks again!
2018-07-20 13:35:18 -07:00
David G Yu
3b5c3d99d2 Release 3.3.3 2018-07-17 19:17:49 -07:00
barry
13b36490a7 Fixed PatchTable's ordering of varying patches when multiple arrays present:
- added new member to PatchArrayBuilder for varying patch points
    - assign varying patch points similar to vertex and fvar patches
2018-07-17 09:29:45 -07:00
Sergey Sharybin
d6f68b17e8 Fix uninitialized stack memory used in topology refiner
The issue is that FeatureMask structure has size of an integer, but only
first 11 bytes are initialized, the rest is left non-initialized to a
stack memory of RefineAdaptive(). This will cause issues with IsEmpty()
which check whether any of integer's bits are set, effectively causing
jumps dependent on a non-initialized memory.
2018-07-14 15:57:16 +02:00
David G Yu
74601558e3
Merge pull request #972 from barfowl/fvar_patch_fix
Fixed regression in PatchTable construction with face-varying patches
2018-07-03 17:46:41 -07:00
barry
365ba36d6f Fixed regression in PatchTable construction for face-varying patches:
- fixed stencil table size estimate when no adaptive refinement occurs
2018-07-03 12:06:23 -07:00
Sergey Sharybin
80cee207b6 Fix wrong argument order for memset 2018-07-03 14:40:33 +02:00
David G Yu
70e3c5f300 Release 3.3.2 2018-06-27 14:29:08 -07:00
barry
02602e7abf Fixed misuse of vector::capacity() in Far::SparseMatrix:
- replaced use of capacity() with size()
    - added missing initializer
2018-06-26 17:42:13 -07:00
barry
5e8e123d73 Major refactor of the construction of Far::PatchTable:
- new PatchTableBuilder class now implements PatchTableFactory::Create()
        - population of tables now independent of scheme or patch type
        - nested LocalPointHelper handles local point sharing and stencils
    - new PatchBuilder class encapsulates patch queries and conversions:
        - provides patch types associated with schemes and client options
        - handles all topological queries about patches in the base class
        - subclasses for the three subdivision schemes to support conversions
        - conversions encapsulated in new SparseMatrix class for re-use
    - removed Far end-cap factory classes and Far::GregoryBasis
    - extended adaptive refinement to support Bilinear scheme (not enabled)
    - extended Vtr::StackBuffer to avoid compiler inefficiencies
2018-05-08 17:24:14 -07:00
David G Yu
171629fbd8 Release 3.3.1 2018-02-07 15:09:46 -08:00
David G Yu
fe47db273c Fixed computation of PatchParam for tri refinement
Previously, Far::PatchTableFactory::computePatchParam() was
considering all faces with a face vertex count != 4 as
irregular.  This was incorrect for triangle refinement as
was the calculation of the UV subface parametric offsets for
triangle subfaces.

This is foundational work for implementing full support
for Loop scheme patch tables. This fix also allows correct
decoding of the parameter mapping for triangles resulting
from uniform refinement of Loop scheme meshes.
2018-02-06 16:51:17 -08:00
barfowl
9c9108c3db
Merge pull request #959 from davidgyu/hlsl_patch_normals
Thanks David
2018-01-16 15:31:33 -08:00
barfowl
dcd82f4137
Merge pull request #960 from davidgyu/mtl_patch_normals
Metal patch shader changes for degenerate normals
2018-01-16 15:31:11 -08:00
David G Yu
dc4948fa75
Merge pull request #957 from barfowl/reserve_overflow_bugs
Fixed integer overflow bugs for large meshes in PatchTable factories
2017-12-15 14:59:47 -08:00