Wenzel Jakob
03632fa11c
attempt to fix a photon mapper crash
2013-03-01 21:05:58 -05:00
Wenzel Jakob
e2efbfcfe3
fix an issue where rendered .pfm files got the wrong extension
2013-03-01 17:15:51 -05:00
Wenzel Jakob
41cad0c23e
Added tag v0.4.4 for changeset bd6ddacdf795
2013-02-28 19:52:10 -05:00
Wenzel Jakob
9d8d59280d
vector class compilation fix for windows
2013-02-28 15:00:37 -08:00
Wenzel Jakob
55ae1126a6
one more version string updated
2013-02-28 13:22:58 -05:00
Wenzel Jakob
766741da85
version bump
2013-02-28 12:51:22 -05:00
Wenzel Jakob
f177f80916
merge with ev-cmake
2013-02-28 11:41:47 -05:00
Edgar Velazquez-Armendariz
dc0fa6a26b
Added some initial configuration flags in CMake for gcc/clang
2013-02-26 17:43:51 -05:00
Edgar Velazquez-Armendariz
f514bc4a53
Mark the strings for the metadata and label commands as used.
2013-02-25 15:35:17 -05:00
Wenzel Jakob
5c1c2fe533
removed a duplicated function which caused problems when compiling in spectral mode
2013-02-22 11:51:38 -05:00
Wenzel Jakob
5a2b0105ad
setpath.sh: don't print a warning message on ZSH if Mitsuba has not been built yet
2013-02-21 23:53:39 -05:00
Edgar Velazquez-Armendariz
bd55638624
Make sure the return type of TVectorN<T>::length is a floating point value.
...
For example, in TVector2<int> the length will almost never be an integer. If the
return type of length were T the resoult would be implicitly truncated from
float/double (the result of std::sqrt) to T.
2013-02-21 02:20:07 -05:00
Edgar Velazquez-Armendariz
a4fb98c4ca
Fix minor double to float warnings
2013-02-21 01:32:29 -05:00
Edgar Velazquez-Armendariz
47a4119ccc
Merge with default
2013-02-21 01:31:52 -05:00
Wenzel Jakob
4d91dbc02c
minor addendum to the docs
2013-02-17 19:03:01 -05:00
Wenzel Jakob
3d93215e81
fixed quaternion slerp to always use the 'short way', updated documentation
2013-02-17 18:28:53 -05:00
Wenzel Jakob
bedef4a672
added turntable rendering info to the documentation
2013-02-17 16:58:39 -05:00
Wenzel Jakob
beeb88b3e3
moved around some code to facilitate rendering turntables and such from Python
2013-02-17 16:37:06 -05:00
Wenzel Jakob
abfc45d7f1
renamed an oddly-named matrix
2013-02-17 13:02:11 -05:00
Wenzel Jakob
2d44b16442
added a few more camera bindings, support for motion blur manipulation from python
2013-02-17 12:46:33 -05:00
Wenzel Jakob
3ed9b6f6f1
mitsuba-python: added some casts() to Scene getters
2013-02-17 11:24:52 -05:00
Wenzel Jakob
fbae5af122
fix for another OSX path issue
2013-02-17 11:11:22 -05:00
Wenzel Jakob
c2498b84a7
added a few more python bindings
2013-02-16 15:29:15 -05:00
Wenzel Jakob
de7ea8d38d
fixed some more occurrences of EPoint
2013-02-15 09:18:52 -05:00
Wenzel Jakob
270a2f8064
minor fix for ConstVolume::supportsVectorLookups
2013-02-14 21:30:01 -05:00
Wenzel Jakob
cecbae219f
documentation fix
2013-02-14 21:28:36 -05:00
Wenzel Jakob
ea20e75c38
windows libpython path bugfix
2013-02-12 14:41:49 -08:00
Wenzel Jakob
05bf6642f1
updated debian build requirements
2013-02-12 13:36:46 -05:00
Wenzel Jakob
0554e4cf74
linux bugfix for the previous commit
2013-02-12 13:24:28 -05:00
Wenzel Jakob
a8f3d76b78
moved FileResolver initialization from libpython to libcore -- yet to be tested on Windows and Linux
2013-02-12 02:00:17 -05:00
Wenzel Jakob
e97081619b
merge with ev-cmake
2013-02-12 01:41:08 -05:00
Wenzel Jakob
e3fbedff69
partial changeset
2013-02-12 01:40:38 -05:00
Wenzel Jakob
6f7e52654d
OSX build fix
2013-02-12 00:28:35 -05:00
Wenzel Jakob
d6d0203cb3
initial support for texture baking, i.e. rendering directly to irradiance textures
2013-02-11 23:52:03 -05:00
Wenzel Jakob
175da3a314
rectangle shape: store UV position in sampleArea()
2013-02-11 22:07:38 -05:00
Wenzel Jakob
979e0fa7a0
ptracer: better handling of unscattered emission -> camera paths
2013-02-11 22:07:00 -05:00
Wenzel Jakob
f374df4680
position the cube at the origin for consistency with other shapes
2013-02-11 21:30:02 -05:00
Wenzel Jakob
47319ea9d4
bugfix for an issue that prevented 'twosided' from working properly in some cases
2013-02-11 19:32:31 -05:00
Wenzel Jakob
ea3b5a220a
minor fix to cube shape doc
2013-02-09 01:40:22 -05:00
Wenzel Jakob
d235a19636
thindielectric fresnel bugfix
2013-02-06 21:40:48 -05:00
Wenzel Jakob
a279b7f638
improved reliability of specular perturbations
2013-02-06 21:17:50 -05:00
Wenzel Jakob
d806c6194e
fixed instancing code once more
2013-02-06 21:16:53 -05:00
Wenzel Jakob
30efd3c151
fixed some openmp + TLS/related crashes
2013-02-06 19:40:15 -05:00
Wenzel Jakob
403706fa62
thin dielectric classifies itself incorrectly
2013-02-06 19:39:55 -05:00
Wenzel Jakob
78c87acfea
added libxml2 dependency
2013-02-06 16:15:18 -05:00
Edgar Velazquez-Armendariz
38349fb575
Fix line endings in changeset 190ef16a6266
2013-02-04 16:32:43 -05:00
Wenzel Jakob
e15097ffae
added support for the IPT color space by Ebner and Fairchild
2013-02-02 20:23:03 -05:00
Edgar Velazquez-Armendariz
01b77d64ee
Get the automatic file resolved path from the mitsuba-core module directly.
...
This way the plugins and other resources would be found when either loading
a mitsuba executable or the Python bindings.
2013-02-02 18:11:49 -05:00
Edgar Velazquez-Armendariz
266b44e404
Initial patches for ARM support [armv7l GNU/Linux]
2013-02-02 10:02:03 -05:00
Edgar Velazquez-Armendariz
bd567511b0
Stricter compatibility with the getopt signature (returns an int, not a char)
...
gcc 4.7.2 in Ubuntu 12.10 detected an issue with assigning the result of getopt
to a char. When getopt returns -1 after exhausting the options, if it is
assigned to a char it sets all its bits to 1. When this result is then tested
for equality to the literal -1, the char was promoted to int, but the byte 0xff
turns into the signed integer 255. Aparently other compiling where assuming -1
is a char-type literal avoiding the conversions.
2013-02-02 09:37:19 -05:00