Commit Graph

855 Commits

Author SHA1 Message Date
Ignacio
84595a62f2 Add helper methods for jai bindings. 2019-02-07 15:39:32 -08:00
Ignacio
c591c5f8b4 Compute spherical harmonics from cube maps. Work in progress. 2019-01-31 18:06:02 -08:00
leper
2ac75fc932 Remove duplicate link instruction. 2019-01-09 00:55:49 +01:00
leper
542711f862 Link with -fPIC when using Clang.
Remove the need for a few workarounds by handling this in just one place.
2019-01-09 00:55:19 +01:00
Ignacio
7c68e09d77 Fix ATOC overflow error on large textures. Merge fix from The Witness. 2018-10-29 12:52:00 -07:00
Ignacio
a9a6f6968e Fix minor warnings.
Add output directories to ignore list.
2018-10-29 12:37:16 -07:00
Starnick
0f65d60602 Updated C-API to support 2D array textures + added missing enum values from nvtt.h 2018-10-18 20:29:57 -04:00
Ignacio
d2b514a3c1 Comment out hack. 2018-10-10 16:30:25 -07:00
Ignacio
a1c54bc7f7 Add fast sRGB conversion. 2018-10-10 14:04:13 -07:00
Ignacio
bc60e8c154 Use our own parallel for by default to avoid dependencies. 2018-10-09 18:11:21 -07:00
Ignacio
1722b00802 FloatImage does not need a virtual dtor. 2018-10-09 18:10:25 -07:00
Ignacio
95f1f60490 Enable stb libraries for image loading. This was disabled by error. 2018-10-09 18:01:15 -07:00
Ignacio
5c70ffef0b
Merge pull request #287 from madscientist159/master
Fix build on OpenPOWER (ppc64le) platforms
2018-08-19 14:04:22 -07:00
Ignacio
005b2518c9 Install nvtt_wrapper.h. Fix issue #288. 2018-08-19 14:02:34 -07:00
Timothy Pearson
95454e9024 Fix build on OpenPOWER (ppc64le) platforms 2018-08-15 21:53:49 -05:00
Ignacio
95bd6193cc Fix bug 281. 2018-05-29 23:28:08 -07:00
Stefan Brüns
ff3bd55892 Fix compilation on Linux - missing include, semicolon
According to http://man7.org/linux/man-pages/man3/memalign.3.html ,
memalign() is defined in malloc.h.
Also, the line should be terminated with a semicolon.

Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
2018-02-06 08:38:37 +01:00
Ignacio
9489aed825 Merge changes from The Witness. 2018-02-05 18:55:07 -08:00
Stefan Brüns
00d7e249e2 Add nvimgdiff to TOOLS target, so it is installed by default
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
2018-02-02 19:46:58 +01:00
Stefan Brüns
389f7582e5 Fix linking of bc7, needs symbols from nvmath
If nvtt is build with -Wl,--as-needed, the linker will not resolve bc7
symbols from any library listed earlier, make sure it (also) appears
after libbc7.a
2018-02-02 19:46:54 +01:00
Remi Papillie
b03d5178fc Fix .hdr images being loaded as LDR before compression 2018-01-28 20:18:13 +01:00
Ignacio
befe79d103 Merge changes from thekla-atlas. 2017-12-15 11:39:21 -08:00
Ignacio
e442d6d390 Remove incorrect assert. Fixes issue #261. 2017-09-04 12:34:00 -07:00
Ignacio
ed735ac2cb Fix alpha coverage subsampling. Fixes issue #266. 2017-09-04 12:25:22 -07:00
Justin Meiners
d47de35893 added noalpha option to assemble and fixed RGB write 2017-06-28 15:53:50 -06:00
Ignacio
4fb0070b7e Do not use undeclared format. 2017-06-28 11:26:18 -07:00
Ignacio
fa03dfc291 Add option to control prefered swizzle. 2017-06-28 11:19:47 -07:00
Ignacio
923b57db45 Merge pull request #255 from AMDmi3/patch-3
Use libexecinfo on FreeBSD as well
2017-06-06 17:58:17 -07:00
Ignacio
ef30f69c9d Merge pull request #256 from AMDmi3/patch-4
Use HW_NCPU on FreeBSD as well
2017-06-06 17:57:50 -07:00
Ignacio
7849ee9e71 Merge pull request #260 from leper/master
Add virtual destructor to TaskDispatcher.
2017-06-06 17:57:01 -07:00
Ken Cooke
01597842fe Update CMakeLists.txt and vc8/9/10 projects 2017-06-04 11:10:15 -07:00
Ken Cooke
96b73af196 Fast implementations of toLinear() and toGamma() for default gamma=2.2 2017-06-01 13:44:28 -07:00
leper
78054e977b Add virtual destructor to TaskDispatcher.
This removes a GCC warning about that missing while virtual functions exist.
2017-05-04 03:22:19 +02:00
Dmitry Marakasov
a64cc24169 Use HW_NCPU on FreeBSD as well 2017-02-16 18:21:39 +04:00
Dmitry Marakasov
6b24b1f45c Use libexecinfo on FreeBSD as well 2017-02-16 18:20:07 +04:00
Lunkhound
0535ab3414 fix a bug where created normal maps were not packed prior to compression 2016-12-26 12:34:06 -08:00
Lectem
24e63456c6 Fix compression for non multiple of 4 texture size
If the width or the height was not a multiple of 4, we would read data outside of the bitmap array
2016-10-19 17:47:00 +02:00
Ignacio
6fb57d3478 Fix order of kaiser parameters. 2016-10-05 11:14:11 -07:00
Ignacio
383331bb42 Fix issue #240. 2016-09-19 11:29:40 -07:00
Ignacio
1a75484520 Minor fixes. 2016-09-19 11:20:25 -07:00
Ignacio
a11a6739c2 Fix comments. Minor fixes. 2016-09-19 11:12:19 -07:00
Lectem
1770ad7280 Fix build on VC2015
On VC2015 the project wouldn't build because of the missing typedefs.
This was tested on both 2013 and 2015.
2016-09-19 10:28:03 +02:00
Ignacio
b0afc82d7d Minor osx fixes.
Remove FindCUDA and use cmake version instead.
2016-08-16 12:31:31 -07:00
Kamil Rytarowski
3c4ccd41fa Add basic NetBSD support 2016-07-30 11:49:33 +02:00
Timothy Pearson
f8790a536a Fix remaining issues on PPC64EL
One AltiVec-related build issue remained
Self tests now pass on LE POWER platforms

Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com>
2016-07-12 06:04:04 +00:00
Timothy Pearson
203eda1d47 Fix build on PPC64LE
Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com>
2016-06-26 20:09:39 -05:00
Timothy Pearson
47b06d3255 Fix build failure due to PPM save incorrect data types
Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com>
2016-06-26 20:02:28 -05:00
sanderd17
65b3dfa4a6 Fix compilation on GCC 6
GCC 6 apparently doesn't want to cast booleans to pointers anymore. So this fixes it for us. See also our changeset: http://trac.wildfiregames.com/changeset/18364
2016-06-12 15:09:19 +02:00
Ignacio
875a78a124 Merge pull request #236 from mpersano/master
Add support for compressing texture arrays.
2016-02-29 12:50:39 -08:00
mpersano
46e5c6a8ee Fix nvassemble command-line options parsing. 2016-02-28 08:17:35 -03:00
mpersano
2bd367a4dc Add support for compressing texture arrays.
Also add support for reading/writing DX10 uncompressed (RGB/lumi) textures, so
now it's possible to compress a texture array assembled with nvcompress.
2016-02-28 08:05:59 -03:00
Ignacio
49677da371 Remove executable property. 2016-02-22 11:54:43 -08:00
mpersano
db8fc561c3 Support assembling texture arrays. 2016-02-11 18:43:01 -02:00
Ignacio
9b79638f79 Disable test code from nvdecompress. Update vc12 projects and fix vc12 build. Fixes bug #232. 2015-12-11 12:15:22 -08:00
Ignacio
7596650a2d Be more clear that Qt4 is not used. From cfcohen. 2015-12-05 15:23:00 -08:00
Ignacio
9d6c7d6a69 Integrate more changes from cfcohen. 2015-12-05 15:16:14 -08:00
Ignacio
a20044bccc Integrate changes from cfcohen to supress warning. 2015-12-05 15:09:57 -08:00
Ignacio
268e6abd65 Integrate cfcohen changes to fix linker problems with gcc. 2015-12-05 15:06:37 -08:00
Ignacio
87ca69bd14 Do not output messages for disabled libraries. 2015-12-05 14:58:05 -08:00
Ignacio
63d29a508e Fix compiler warning. 2015-12-05 14:56:36 -08:00
Ignacio
e9a8630a3f Increase array size to supress compiler warning. 2015-12-05 14:50:52 -08:00
Ignacio
c0ad0f4d31 Merge changes from The Witness. 2015-10-28 23:53:08 -07:00
Ignacio Castaño
a382ea5b21 Fix compiler warning. 2015-09-02 12:25:47 -07:00
Ignacio Castaño
26723de502 Fix memory leak reported by Sylvain Rochette in issue #229. 2015-08-21 21:53:16 -07:00
Ignacio
58617584d4 Apply patch provided in issue #224. Add support for ARM64. 2015-07-20 23:47:01 +02:00
Ignacio
4ef408b591 Remove debug code. 2015-07-20 14:32:16 +02:00
Ignacio
d32f0e0173 Merge pull request #223 from cmdrf/no-include-library-h
Do not include Library.h
2015-06-12 10:27:02 -07:00
cmdrf
7b031aa4bf Do not include removed Library.h 2015-06-12 14:00:37 +02:00
cmdrf
d01d72f8e7 Update NV_FOREACH for C++11 2015-06-12 12:38:44 +02:00
Ignacio
0d205015ae Merge pull request #221 from cmdrf/snprintf-vs2015
Do not redefine snprintf on VS2015
2015-06-10 10:51:07 -07:00
Ignacio
7aa0390f32 Merge pull request #220 from cmdrf/cmakecuda
Fix CMake CUDA introspection
2015-06-10 10:49:35 -07:00
cmdrf
fa062aa298 Do not redefine snprintf on VS2015 2015-06-10 18:46:26 +02:00
cmdrf
205f492e2e Fix CMake CUDA introspection 2015-06-10 18:14:12 +02:00
cmdrf
899a99a106 Fix link order for Linux 2015-06-10 17:59:57 +02:00
cmdrf
766edcb1a6 Fix more link order for Linux 2015-06-10 17:59:56 +02:00
Thomas Roß
9ad63e7f3a Fixed compiler flag being set on wrong target. 2015-04-21 22:18:25 +02:00
Ignacio
81336cc3e9 Remove unnecessary files. 2015-03-25 12:36:16 -07:00
Ignacio
e9d44813d2 Merge pull request #216 from esok/master
Added support for gcc and c++11 features.
2015-03-24 17:08:37 -07:00
Ignacio
a083337473 Merge changes from The Witness. 2015-03-24 12:14:49 -07:00
esok
6bc3224fe7 Added support for gcc and c++11 features. 2015-03-23 22:05:33 -04:00
castano
7e2a9d1adb osx fixes. Fix issue 211. 2014-12-02 20:23:21 +00:00
castano
2d6fc0e304 DXT1 compressor tweaks. 2014-12-02 05:43:13 +00:00
castano
d019cd7080 Merge changes from the witness. 2014-11-04 17:49:29 +00:00
castano@gmail.com
4cb60cc5ba Fix vc12 build. 2014-09-23 17:49:18 +00:00
castano
ac87eefa6d Apply patch to update C# wraper. Fixes issue 184. 2014-06-11 21:02:04 +00:00
castano
f705b2eaa1 Fixes issue 93. 2014-06-11 19:46:17 +00:00
castano
6ade0e27b7 Fixes issue 202. 2014-05-21 18:09:47 +00:00
castano
7e74261f92 Fixes issue 204. 2014-05-21 17:47:06 +00:00
castano
57298c71c8 Fix issue 194. 2014-05-21 17:23:05 +00:00
castano
b06a6f46a5 Fixes issue 207. 2014-04-18 16:25:24 +00:00
castano
63d6036e3a Fix issue 206. 2014-04-03 22:10:12 +00:00
nathaniel.reed@gmail.com
ab316deeaa Add BC7 support. It's incredibly slow - ~60 seconds to compress a 512x512 image, on a Core i7 - but it works.
- Added AVPCL compressor to projects and got it building with VC9 and VC10.
- Removed unused command line interface & file read/write code from AVPCL.
- Convert AVPCL to use NV vector math lib, asserts, etc.
- Convert AVPCL to use double instead of float.
- Added 4x4 symmetric eigensolver, for AVPCL; it's based on the existing 3x3 one, but I had to rewrite the Householder reduction stage.  As with ZOH, using the eigensolver (instead of SVD) gives a ~25% speedup without significantly affecting RMSE.
- Encapsulate ZOH and AVPCL stuff into their own namespaces to keep everything separate.
- Added some missing vector operators to the nvmath lib.
2013-12-07 02:17:08 +00:00
nathaniel.reed@gmail.com
f2fa0517b5 Ported over James Arvo's SVD code from ZOH. It's a good deal slower than the eigensolver, and no better in RMSE - so didn't use it, but left the code in place for future reference.
Also replaced doubles with floats in the eigensolver code.  Speeds up BC6 another 5%.  No change to RMSE in test suite.
2013-12-05 02:13:17 +00:00
nathaniel.reed@gmail.com
474239c784 Add BC6 support to nvtt lib and utils.
- Use 3x3 eigensolver for initial fit in ZOH.  Slightly better perf and RMSE than power method.
- Remove use of double precision in ZOH - speeds up by 12%.
- Fixed RGBM encoding that was broken for HDR images.
- Use gamma-2.0 space for RGBM for HDR images (improves precision in darks).
- Use UNORM instead of TYPELESS formats when saving a DX10 .dds file.  The TYPELESS formats break most viewers.
- Cleaned up warnings in ZOH code.
- Command-line utils will warn if you give them an unrecognized parameter.
- Added VS2010 profiling results.
2013-10-25 17:30:55 +00:00
nathaniel.reed@gmail.com
77188a42ac Fix VC10 project files. Fix code to build with CUDA 5.5 (the cudaMemcpyToSymbol overload that takes the symbol name as a string was removed). 2013-10-16 21:18:16 +00:00
castano@gmail.com
57b99a1595 Apply CubeSurface fold/unfold patch. Fixes issue 199. 2013-09-04 01:10:52 +00:00
castano@gmail.com
b2b367ef5f Fix xyzToCieLab bug and add CIE Lab DeltaE 1994 color differencing.
Fixes issue 197.
Fixes issue 198.
2013-09-04 01:08:39 +00:00
castano@gmail.com
022944e49b Apply patch to fix issue 192. 2013-07-20 00:08:06 +00:00
castano@gmail.com
27024253ce Handle 3d textures in nvcompress. 2013-07-09 17:16:41 +00:00
castano
22bc0b949d Fix assert. Fixes issue 190. 2013-06-25 17:03:08 +00:00