ViewVC Help
View File | Revision Log | Show Annotations | Download File | Root Listing
root/radiance/ray/doc/notes/ReleaseNotes
(Generate patch)

Comparing ray/doc/notes/ReleaseNotes (file contents):
Revision 1.1 by greg, Sat Mar 15 17:32:55 2003 UTC vs.
Revision 1.90 by greg, Sat Jul 30 16:41:46 2005 UTC

# Line 921 | Line 921 | Fixed bug in Type B photometry angle calculation in so
921   by Erich Phillips.
922  
923   ==============================
924 < Version 3.5 (Released 3/14/03)
924 > Version 3.5 (Released 3/21/03)
925  
926   Fixed bug in replmarks pointed out by Georg Mischler, which caused
927   garbage to be output under specific conditions (unscaled, unrotated
# Line 1044 | Line 1044 | to create a rendering library called "libradiance.a" a
1044   to install it in the src/lib/ directory along with the new
1045   raycalls.o module, which should facilitate the building
1046   of new applications outside the src/rt/ directory.  The
1047 < raypcalls.o module, also included in -lradiance, manages
1047 > raypcalls.o module, also installed in src/lib, manages
1048   parallel rendering on multiprocessors with shared memories.
1049  
1050   Wrote new ranimove animation rendering program, which is an
# Line 1138 | Line 1138 | models where many faces come together at a point.
1138   This required a little reprogramming in for couple
1139   of recursive routines using sets to avoid overrunning
1140   the stack limit.
1141 +
1142 + ==============================
1143 + Version 3.6
1144 +
1145 + Modified 4-byte RGBE encoding routine to avoid problems
1146 + when one or two primary values is negative and the other(s)
1147 + positive.  This used to generate some strange color values.
1148 +
1149 + Added -oc option to rtrace to output local (u,v) coordinates.
1150 +
1151 + Fixed bugs in local uv and j unit vector variables in src/rt/func.c.
1152 +
1153 + Added warning message to OpenGL rendering routines about new
1154 + mesh primitive, which we cannot currently render.
1155 +
1156 + Added -if, -id, -of, and -od options to rcalc to provide for
1157 + binary float and double i/o.
1158 +
1159 + ****************** COMPATIBILITY CHANGE ***********************
1160 + Changed holodeck format to used off_t instead of long for beam
1161 + positions, which allows holodecks to grow to system file size limit.
1162 +
1163 + Fixed wrapping long in source hit test code, which would cause
1164 + odd horizontal lines in some very long renderings.
1165 +
1166 + Added -f and -w options to ra_tiff to create IEEE float and 16-bit/primary
1167 + TIFF output, respectively.  These formats are also understood on reverse
1168 + conversion.  (16-bit/primary has been understood for some time, but not
1169 + IEEE float.)
1170 +
1171 + Added -N option to rad to start multiple rendering processes in
1172 + parallel.
1173 +
1174 + Fixed bug where mesh instances were not being freed, so subsequent calls
1175 + in programs like ranimove would refer to bogus data.
1176 +
1177 + Added support in normtiff for floating-point and 16-bit/sample TIFF input
1178 + files.
1179 +
1180 + Changed marksources() routine (src/rt/source.c) to check all
1181 + primitives in the modifier chain for emitting types, not only
1182 + the first.  This permits sources to be described with intervening
1183 + modifiers before the light/illum/glow primitive and allows aliases
1184 + as well.
1185 +
1186 + Fixed bug in xform where it was neglecting to add transform arguments
1187 + to new mesh primitive.
1188 +
1189 + Added -dF and -dD options to pvalue for byte-swapped floats and
1190 + doubles, respectively.
1191 +
1192 + Changed meta2tga to output correct colors.
1193 +
1194 + Changed local coordinate index from 32-bits to 16-bits, which requires
1195 + that Radiance mesh files be recompiled.  (Later decided against this
1196 + change, as it caused problems with some models.)
1197 +
1198 + Changed integer tone-mapping routines to better deal with images
1199 + containing large, constant regions.
1200 +
1201 + Improved interaction with rholo -o ogl driver under OS X, which was
1202 + returning bad values from the depth buffer.
1203 +
1204 + Minor bug fix in ranimove where it was restarting processes when it
1205 + didn't need to.
1206 +
1207 + Increased accuracy of default ambient parameters in rpict, rview, and
1208 + rtrace.  In rad program, doubled -ar value, -ad value, and -as value,
1209 + and decreased -aa value a bit.  Also adjusted -lr upwards and -lw
1210 + downwards.  I should have done these things long ago, as processors
1211 + have been getting faster and faster over the years.
1212 +
1213 + Fixed bug discovered by Marcus Jacobs where oconv was dumping core
1214 + when it was given too many modifiers.
1215 +
1216 + Fixed problem with rpict where it was drawing illum sources badly
1217 + even when they weren't all that small in the view.
1218 +
1219 + Added in(i) function to rcalc for more flexible channel evaluation.
1220 +
1221 + Replaced all calls to vfork() with regular fork() calls.
1222 +
1223 + Fixed bug in glarendx for dgr calculation with sources behind view.
1224 +
1225 + Added -p option to histo to print percentages/percentiles r.t. counts.
1226 +
1227 + Added a source occluder cache to accelerate shadow testing.  This
1228 + can save up to 50% of the calculation time in scenes with many
1229 + light sources, especially ones containing large occluders.  This
1230 + optimization may be tuned or switched off with the -DSHADCACHE=N
1231 + compile switch, where N is the resolution of the shadow cache, or
1232 + 0 to disable it.  Since this is purely a time vs. memory trade-off,
1233 + I decided not to make it a new command-line option.  (We already
1234 + have plenty.)
1235 +
1236 + ****** COMPATIBILITY CHANGE ********
1237 + Changed the following command names to avoid conflicts:
1238 +        rview -> rvu (with symbolic link back to rview)
1239 +        calc -> icalc
1240 +        lam -> rlam
1241 +        gencat -> gencatenary
1242 +
1243 + Improved -f option of normpat, which caused odd color changes before.
1244 +
1245 + Created pgblur script which does fast Gaussian blur of image without changing
1246 + its dimensions.  This is very useful in certain types of image processing
1247 + calculations, such as finding or removing low frequencies.
1248 +
1249 + ****** COMPATIBILITY CHANGE ********
1250 + Changed little-used objpict script so that it takes scene files rather than
1251 + rpict options on the command line, and puts everything into the right
1252 + coordinate position automatically rather than requiring the user to do it.
1253 +
1254 + Added -I option to replmarks for mesh instancing.
1255 +
1256 + Made similar change for virtual sources to allow modifiers and aliases
1257 + on the mirror and prism types to work properly.
1258 +
1259 + Rob Guglielmetti added -lw and -lh options to falsecolor to control the
1260 + legend dimensions.  Made it so -lw 0 and/or -lh 0 turns legend off.
1261 +
1262 + Made it so pcompos will read everything from a stream even if it doesn't
1263 + need it, so input commands will not receive a hangup signal.
1264 +
1265 + Wrote ra_bmp program to convert to and from Windows BMP files (finally).
1266 + It's the most horrible image format I've ever seen, and I've seen a lot.
1267 +
1268 + Increased -aa default value for all renderers (and also -ad and -as in
1269 + rview) at Schorsch's suggestion.  I think this undid my earlier decrease
1270 + of -aa last time, which I guess was ill-considered.
1271 +
1272 + Modified oconv to keep subdividing the octree if there are too many
1273 + objects in a voxel even when the -r resolution limit has been
1274 + reached.  This was not done before due to problems with recursion,
1275 + and still, it will only continue to 256 times the original -r limit.
1276 + If that isn't enough, then there's probably no hope, anyway.
1277 +
1278 + Added -l option to obj2mesh to search for material files in library
1279 + locations.
1280 +
1281 + Increased polygon vertex limit to 512 in obj2mesh and obj2rad and added
1282 + warning if this limit is exceeded.
1283 +
1284 + Added tone-mapping and monitor primary options to ra_bmp.  This gives
1285 + new functionality similar to normtiff and ximage.  It would be nice to
1286 + add this feature to the other ra_* programs as well...  someday.
1287 +
1288 + Fixed long-standing bug in reported pixel ordering out of certain
1289 + protate commands.
1290 +
1291 + Added -p option to rcalc for "passive mode" that transmits
1292 + unrecognized input -- useful for substituting data in the
1293 + middle of a stream.
1294 +
1295 + Fixed a very obscure bug in rpict that caused a dark horizontal
1296 + line in certain cylindrical projections due to improper drawing
1297 + of the sky dome source in rt/srcdraw.c.
1298 +
1299 + Fixed problems with effective ray lengths computed for dielectric
1300 + and BRTDfunc materials.  This was most evident in breakup of holodeck
1301 + views when VDISTANCE=True.
1302 +
1303 + Fixed rad so that 'scene' variable may consist only of inline commands.
1304 +
1305 + Fixed bug introduced in shadow caching routines where sources in the
1306 + middle of spheres were not illuminating properly.  Thanks to Guilio
1307 + Antonutto for spotting this problem.
1308 +
1309 + Modified gensky behavior so if time is given with standard time zone
1310 + designation, any -m option is ignored.
1311 +
1312 + Added -n option to mkillum for (shared memory) parallel processing.
1313 +
1314 + Fixed a long-standing bug in parallel processing code that only
1315 + seemed to manifest with the new mkillum -n option, where rtrace
1316 + would deadlock upon termination.  The reason it never showed up
1317 + before is because I always closed processes in the reverse order
1318 + of opening them in other programs.  I went through and found similar
1319 + (potential) bugs in the rview device subprocess code and util/netproc.c.
1320 +
1321 + Corrected the normalization of the Gaussian reflectance model.  This will
1322 + cause an increase in specularity for low-lying light sources.
1323 + Thanks to Don Walters of Boeing and Arne Duer of the Univ. of Innsbruck
1324 + for pointing out this error.  The square root used for the two cosines
1325 + in the denominator of the normalization factor should not be present.
1326 + This was a serious mistake I made in the original model, which went
1327 + undiscovered (at least to my knowledge) for over a decade.  I think
1328 + Bruce Walter of Cornell may have known about it, but I never heard
1329 + the details of his discovery.
1330 +
1331 + Enabled the -p option for primary conversion without tone-mapping
1332 + in ra_bmp.  Also provides for XYZE input conversion to RGB.
1333 +
1334 + Fixed what I consider a bug in pvalue -o when taking XYZE files on input,
1335 + where it was inappropriately undoing the exposure value so that it always
1336 + reported radiance rather than XYZ values with Y in cd/m^2.
1337 +
1338 + Made it so ra_bmp only produces run-length encoded 8-bit output for
1339 + grayscale images that are tone-mapped.  This minimizes circumstances
1340 + under which output scanline order is reversed.
1341 +
1342 + Fixed bug in rcalc where it sometimes wasn't parsing the input format
1343 + correctly.
1344 +
1345 + =====================
1346 + Version 3.7
1347 +
1348 + Fixed bug in source occluder cache which caused consistency error
1349 + for distant sources and mist materials.  (Thanks to Mark Stock for
1350 + finding the problem.)
1351 +
1352 + Fixed serious bug in source drawing routines for rpict, where the edges
1353 + of light sources were appearing through obstructions.  (Thanks to Veronica
1354 + Sundstedt for pointing this out.)
1355 +
1356 + (The above were included in the 3.6.1 patch release.)
1357 +
1358 + Changed default temporary directory from /usr/tmp, which doesn't exist
1359 + on a lot of systems these days, to /tmp.
1360 +
1361 + Eliminated many unused files and programs from the distribution, including
1362 + the following programs that were being built but no longer used:
1363 +        oki20   - printer driver for OkiMate 20 printer (B&W)
1364 +        oki20c  - printer driver for OkiMate 20 color printer
1365 +        pcwarp  - intended as more general substitute for macbetcal,
1366 +                        but never quite worked
1367 +        ra_avs  - convert to & from AVS image format
1368 +        ra_bn   - convert to & from Barneyscann format
1369 +        ra_pr   - convert to & from 8-bit Sun rasterfile format
1370 +        ra_pr24 - convert to & from 24-bit Sun rasterfile format
1371 +        thf2rad - convert GDS "things file" to Radiance scene format
1372 + If anyone is still wanting any of the above programs, let us know and we'll
1373 + reinstate it.
1374 +
1375 + Fixed bug in pcond's acuity simulation for small view angles (less than
1376 + 4 degrees horizontally or vertically).  Thanks to James Lee for figuring
1377 + this out.
1378 +
1379 + Added script to convert to/from ICT's Poskanzer Float Map, which I don't
1380 + care to support officially due to its machine-dependent byte order.
1381 +
1382 + Fixed bug that caused incorrect reporting of non-surface in set when
1383 + mixing meshes and instances.  (Thanks to Jack de Valpine for discovery.)
1384 +
1385 + Fixed inaccuracy in calculation of diffuse transmission in the presence
1386 + of pure specular (Fresnel reflection) using "trans" material.  Thanks
1387 + to Christoph Reinhart for pointing out the error.
1388 +
1389 + Made internal change to tone-mapping library to use structure pointer
1390 + rather than stack, making code re-entrant and more up to modern standards.
1391 +
1392 + Added -pd option to handle depth-of-field blur in rpict.  This requires
1393 + that the view directon (-vd option) include focal distance information,
1394 + which is now preserved by the various view file routines, and reported
1395 + by the vwright command.
1396 +
1397 + Added DBLUR variable to ranimate and pmdblur script to handle simultaneous
1398 + motion and depth-of-field blurring.  Thanks to Visarc for funding this work.
1399 +
1400 + Added setting of focus distance in rvu ("focus" command) and rholo
1401 + ('f'/"frame" command).
1402 +
1403 + Changed default remote shell in ranimate to "ssh" rather than "rsh",
1404 + which has been abandoned on most Unix systems due to security issues.
1405 +
1406 + Fixed behavior of source primitive using illum with no alternate
1407 + material.  These now become transparent so one can see the sky or
1408 + light-probe background for use in image-based lighting.  Previously,
1409 + the user needed to specify the background glow as the alternate
1410 + material for a distant illum source to behave properly.  (This still
1411 + works.)
1412 +
1413 + Fixed some minor problems and inconsistencies with 16-bit/channel
1414 + TIFF i/o in ra_tiff.
1415 +
1416 + Created mksource program to generate distant sources for improved
1417 + image-based lighting with light probes.
1418 +
1419 + Changed ambient calculation so that -ad will not limit recursion
1420 + depth, and a minimum of 27 samples will be sent out for each stored
1421 + ambient value, or 3 samples if -aa is 0 (caching off).
1422 +
1423 + Added -oM option to rtrace to report material (rather than modifier)
1424 + for cases where actual material is aliased or further up modifier chain.
1425 +
1426 + Rearranged the way ray contributions are computed so that it might be
1427 + possible to use rtrace to derive daylight coefficients and optical
1428 + transfer functions.  The new -oTW option of rtrace produces color
1429 + contribution weights for each ray, which combined with the -ti (or -tI)
1430 + option permits the contribution of selected surfaces or sources to
1431 + be determined.  Using the -oTW option, one should also specify
1432 + options "-dt 0 -aa 0 -as 0" for optimal results.  In particular,
1433 + one cannot expect reasonable tallies with -ab >= 1 and irradiance
1434 + caching on (-aa > 0).
1435 +
1436 + The same code modifications included a change to the indirect computation
1437 + with caching switched off (-aa 0).  Whereas before, the number of secondary
1438 + rays after the first bounce were half of the previous generation, they are
1439 + now determined by the surface reflectance, making for a better and faster
1440 + calculation.  It should now be possible to compute many bounces with -aa 0,
1441 + which used to be prohibitively expensive.
1442 +
1443 + Added -T option to xshowtrace to trace rays to light sources.
1444 +
1445 + Added meta2bmp program to produce BMP files from metafile graphics.
1446 +
1447 + Added cct_x(t) and cct_y(t) to src/cal/cal/blackbody.cal to interpolate
1448 + CIE chromaticities for particular black-body temperatures.
1449 +
1450 + Increased maximum number of input pictures in pcomb and pcompos to 512.
1451 +
1452 + Created rtcontrib program for computing ray contribution coefficients.
1453 + This tool may be used to do standard optical ray tracing or daylight
1454 + coefficient calculations for annual simulations, among other uses.
1455 +
1456 + Added tilde ('~') output option as part of the -o* specification of
1457 + rtrace.  This permits ray trees to be parsed more easily in binary
1458 + output mode, as needed by rtcontrib.
1459 +
1460 + Implemented Russian roulette ray termination with -lr <= 0.
1461 + Made -lr -10 the default in rtcontrib.
1462 +
1463 + Made rtcontrib -f option search RAYPATH directories.
1464 +
1465 + Made it so -o option of rtcontrib can pipe -o output to a command.
1466 +
1467 + Added -if and -of options to total for binary i/o (following rcalc mold).
1468 +
1469 + Switched default options in rtrace to use Russian roulette.  (Rpict and rvu
1470 + still use biased ray termination.)
1471 +
1472 + Added rtcontrib -r option to recover partial output.
1473 +
1474 + Added -u rendering option for pure Monte Carlo sampling.
1475 + (Got rid of -DMC compile option, which this makes superfluous.)
1476 +
1477 + Changed -t option of rlam to accept strings as well as single characters.
1478 +
1479 + Fixed bug in rcalc with -l option where it would sometimes go into an
1480 + infinite loop.
1481 +
1482 + Fixed truly ancient bug in bounding box computation for cones, which
1483 + could (and did) cause missing segments.
1484 +
1485 + Added -z option to normtiff to output LZW-compressed files.
1486 +
1487 + Last change to total causes it to produce results when it sees an empty
1488 + line on the input.  This was not intentional, but I decided I like this
1489 + behavior so I documented it rather than changing it.
1490 +
1491 + =====================
1492 + Version 3.8
1493 +
1494 + Fixed Russian roulette ray termination, which was doing the opposite of
1495 + what it should have been doing, resulting in overestimated values below
1496 + minimum ray sampling weight (serious).

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines