fix(aBSDF): Improved appearance/accuracy in exclusion zone during peak extraction. Thanks to David Geisler-Moroder for initial idea, implementation and testing.
Increased number of sampling directions for better peak evaluation (thanks to David G-M for implementation and testing)
Removed small source sampling bias in direct_specular_OK()
Improved cthru coefficient calculation for aBSDF by summing more inclusive peak
Added max() and min() as .cal library functions
Added -orRxX options to rtrace for VR rendering
Added balancing calls to SDfreeCache()
Removed removal of cosine factor, which was actually in error(!)
Switched to minimum solid angle rather than local SA for over-counting test
Decided to go with "aBSDF" for "aperture BSDF" rather than "sBSDF"
Fixed bug in direct calculation for BSDF introduced in last change
Added sBSDF material type for explicit control over peak extraction
Some paranoia to avoid infinite loops
Avoid loss of specular sampling for low-weight daughter rays
Reverted previous change and added 1.5 fudge factor to avoid over-counting
Fixed issue with under-counting specular
Tweaked peak extraction code to improve results
Made it so weighted average only kicks in for small sources
Added cosine correction for through component and improved accuracy of direct specular component
Fixed occasional issue with zero BSDF regions
Minor change should not affect behavior
Implemented view component for BTDFs
Fixed over-counting error introduced in revision 2.31
Fixed issue with indirect under-estimation in proxied BSDF transmitters
Further tweak to make sure we never double-count
Fixed issue with variable-resolution BSDF sampling of direct component
Fixed comments and changed threshold to 0.7% for "through" shadow testing
Made shadow testing work through BSDF's with through component (Thanks to David G-M)
Renamed direct sampling routine for clarity
Corrected double-counting of diffuse contributions
Simplified handling of diffuse components
Had to reinstate ambRayInPmap() macro to avoid over-counting bug
Fixed various compiler warnings (mostly harmless)
Removed deprecated ambRayInPmap() macro from code
Initial check-in of photon map addition by Roland Schregle
Enabled back face invisibility (-bv0) for transparent/translucent types
Added missing call to set function context
Added extra checks where projected solid angle is undefined
Fixed bug for BSDF materials included via octree or mesh instances
Fixes to reciprocity for tensor tree representation
Created rcontrib program (eventually to replace rtcontrib)
Further refinements in error testing/reporting
Improved error reporting for illegal orientation vector
Fixed serious bug in transmitting proxy material ambient sampling
Minor optimizations
Improved oversampling check
Further improved sampling method
Minor improvements to direct specular sampling
Vastly improved direct specular sampling
Minor fix to through component calculation
Stratified specular sampling with -ss > 1
Partial implementation of variable-resolution BSDFs
Fixed interface for determining BSDF solid angles
Improved output efficiency by reducing number of calls to fwrite()
Fixed problem with double-counting through component in indirect
Fixed bug in source sampling for proxies
Made BSDF proxy behavior consistent for shadowing, reflection and transmission
Fixes to BSDF including blurring of angle boundaries
Minor changes and fixes -- first working version of BSDF material
Minor fixes and adjustments
Major code reorg, moving mgflib to common and introducing BSDF material
This form allows you to request diffs between any two revisions of this file. For each of the two "sides" of the diff, select a symbolic revision name using the selection box, or choose 'Use Text Field' and enter a numeric revision.