perf(rtrace): minor optimization
fix(rtrace): Fixed spectral output for -om option
fix(rtrace): bug for float spectral output
syle: fixed compiler warning
fix(rtrace): added check for -f?c output and 1-channel sensitivity function
feat(rpict,rtrace,rcontrib,rtpict): Hyperspectral rendering (except photon map)
refactor: Changed some char* args to const char* to avoid warnings
refactor: made "castonly" variable accessible outside rtrace
fix(rtrace): Restored original silence during virtual source pretests
fix(rtrace): Fixed segmentation violation introduced in 2.93 (Apr 2020)
fix(rtrace): restored tracing to sources with -oT option
fix: minor issue with rayorigin() called before direction set
fix: better consistency with infinite rays returning origin instead of (0,0,0)
fix(mkillum, rcontrib, rtrace, ranimove, rsensor): stall under macOS 10.15 due to broken flockfile() implementation -- workaround is better, anyway
Backed out last change, which actually made things worse
Changed so batch optimization only happens if no flush interval
Optimize rtrace for throughput, especially with ambient cache
Removed superfluous declaration
Fixes and improvements from last change, making ray-casting faster
Made start-up much faster when only doing ray-casting
Added better check to see if input is from a pipe or not
Added NCOMP= to header and better checking of -o* options
Turn off multiprocessing with -oT* and -ot* options
Added input ray queue for more efficient batch processing
Added precautionary testing when checking for FHUGE
Undid last change, which introduced a bug!
Minor simplification to last change - should not affect results
Slightly improved flushing logic
Took out redundant setting
Avoid bug where -onN and -oNn output different information
Keep track of ray flips so rayreorient() can put it back as it was
More warnings and correction because -oW and -oV are useless options
Made -f?c RGBE output compatible with any of 'vVWrx' options and added warning
Check that number of processes does not exceed flush interval
Removed unnecessary clear + made multiple zero directions -> single flush
Made rtrace a little more robust when reporting bogus ray values
For consistency, reflected ray distance should be inf for diffuse surface
Added missing initialization for rmt and rxt
Made rtrace a little more efficient under Windows if -n > 1
Added -orRxX options to rtrace for VR rendering
Moved findmaterial() from source.c to initotypes.c
Improved error message
Switched over to more efficient fread/fwrite replacements getbinary/putbinary
Switched to more efficient i/o for small vectors
Made reading of vectors more consistent with FVFORMAT macro
Added check for blocked shadow rays
Created rcontrib program (eventually to replace rtcontrib)
Improved output efficiency by reducing number of calls to fwrite()
Eliminated flushing with -x >1 if -y also set to improve -n performance
Bug fix for zero-direction rays with multiprocessing
Precautionary change to rayirrad()
Important bug fix in -I calculation
Minor cosmetic change
Eliminated restriction with -n and -I in rtrace
Bug fixes and performance improvements to rtrace -n option
Added -n option to rtrace and moved quit() funciton out of raypcalls
Fixed -oo output of rtrace -I
Increased limits to number of modifiers for trace and ambient include/exclude
Changed default in rtrace to -u+ (random sampling)
Made -u+ option truly random
Increased trace modifier limit to 8192 in rtrace
Added -V option to rtcontrib and -oV option to rtrace to report contributions
Added final write error check
Removed SPEED from SCons. Changes to build with mingw, as suggested by Francesco Anselmo.
Fixed bugs in black glass transmission and source weight computation
Added -R option for pure Monte Carlo rendering, replacing -DMC compile option
Fixe in rtcontrib -M option
Fixed bug where rtrace would quit after 2^31 primary rays
Changed rtrace to default to Russian roulette ray termination
Added binary i/o options to total and piped output to rtcontrib
Got rtcontrib working and wrote basic man page
Created rtcontrib program for computing ray contributions and coefficients
Extensive changes to enable rtrace -oTW option for tracking ray contributions
Added -oM option to rtrace to output material
Increased defaults and allowed for compiler override related to -a{E,I} options
Continued ANSIfication. There are only bits and pieces left now.
Increased -aa default settings
More adjustments and corrections to default parameters
Increased default ambient options in rpict, rtrace, rview, rad & mkillum
Eliminated copystruct() macro, which is unnecessary in ANSI. Reduced ambiguity warnings for nested if/if/else clauses.
Minor compile fixes
Replaced most outdated BSD function calls with their posix equivalents, and cleaned up a few other platform dependencies.
Macros for setting binary file mode. Replacing MSDOS by _WIN32.
Added rtrace -oc option to output local (u,v) coordinates
Replaced inline copyright notice with #include "copyright.h"
Changes and check-in for 3.5 release Includes new source files and modifications not recorded for many years See ray/doc/notes/ReleaseNotes for notes between 3.1 and 3.5 release
modified code so zero record flushing happens unless -x > 0 && -y > 0
added -ld boolean option to rtrace for limiting ray distance
changed rtrace behavior to send bogus record when ray dir==0
fixed bug in printing void materials.
changed albedo to 3-color parameter
added -aw option for ambient value weight
added M_MIST (mist) material and global participating medium
increase default parameters for more accurate calculations
added -bv option for back face visibility (normally on)
added -vo and -va (fore and aft clipping plane) view options
consistency improvements related to -I option
added -ti, -te, -tI and -tE options to rtrace for exclude/include traces
made sure that -I option completely overrides -o option
portability fixes (removed gcc warnings)
bug fixes and improvements on stdio flushing
added -P and -PP persist options
changed -di to ! -dv
added -oN for unperturbed surface normal and made -on perturbed
added setmode(fd,O_BINARY) where needed for MSDOS
fixed bug introduced in last change with -h- option of rtrace
added ability to print COLRFMT output of values
added -oL output option for single-ray length
changed default values and handling of specthresh
added -sj and -st options and changed old -s options to -p
removed atof() declarations for NeXT
updated revision number for release 2.0
added source sampling (-ds option)
got rid of scanf() call
added -di option for direct source invisibility
increased defaults for -dp parameter
added virtual source visibility pretesting
added virtual sources
added stratified random sampling with urand
added -i option for irradiance calculation
added -oI option for irradiance after intersection
added conditional ray casting to rtrace for efficiency
Nit-picking for Stardent C-compiler
Added rt field to RAY structure for more accurate z-buffering
improved portability and speed of bcopy()
added flushing for zero input direction vector
Reduced default ambient resolution
Added direct certainty variable (compromise btwn. thresh. & tol.)
Changed direct threshold testing so that it is certain
Efficient approximation to direct component with many sources Glow type changed Spot type eliminated
eliminated unused variable
Initial revision
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.