Commit Graph
-
1ffa994fcore: add boolean operators on real numbers, add unit tests for that, and simplify the Remez code accordingly. by 2011-09-29 19:10:00 +0000 -
6d85192etest: minor updates to the Pi program (now almost deprecated) and the Remez exchange program. by 2011-09-28 17:05:00 +0000 -
8f400b63core: create global real number variables such as R_E, R_PI, R_LN2 etc. by 2011-09-28 17:04:52 +0000 -
4822f02dcore: implement shift operators for reals; they're useful for fast multiplications by integers, especially powers of two. by 2011-09-28 17:04:44 +0000 -
ab03cdb1core: improve exp() on reals: faster (constant time) and a lot more accurate. by 2011-09-28 17:04:37 +0000 -
f62946fccore: implement log() for real numbers, and start documenting our next improved implementation of exp(), which relies on log(). by 2011-09-28 17:04:29 +0000 -
93162ee1test: the Remez algorithm is now almost functional. by 2011-09-28 00:59:42 +0000 -
593aa3afcore: implement sin(), cos() and abs() for reals and fix a crash in the addition and subtraction operators occurring when exponents were too different. by 2011-09-28 00:59:39 +0000 -
afd2cfd6test: more work on the Remez exchange algorithm. by 2011-09-27 17:09:35 +0000 -
c0edb49ecore: add exp() for real numbers and fix the == operator. by 2011-09-27 17:09:28 +0000 -
908b2b3breal: fix a bug in the real number multiplication when one of the arguments is zero. by 2011-09-27 17:09:21 +0000 -
02bcc443core: add sqrt() for real numbers. by 2011-09-27 17:09:13 +0000 -
b7a1270btest: start working on a Remez exchange algorithm implementation so that we can create our own high-precision polynomial approximations. by 2011-09-26 23:20:53 +0000 -
ff45d2actest: add a 10-line example program that computes Pi to the 150th digit. by 2011-09-26 23:20:50 +0000 -
58ee1d5bcore: add operators +=, *= etc. to the real class, and refactor the print() method so that it displays decimal values. by 2011-09-26 23:20:47 +0000 -
058d8823core: add rounding to real->double conversion, fix a bug in the addition code shortcut, fix bugs in the addition and subtraction, another one in the multiplication code, and add new unit tests for most of these. by 2011-09-26 00:00:52 +0000 -
fbdec834image: add a GDI+ loader. by 2011-09-24 22:21:54 +0000 -
56511457test: give more explicit names to unit test cases. by 2011-09-24 22:21:44 +0000 -
48bde6fatest: add simple unit tests for image loading. by 2011-09-24 22:21:40 +0000 -
fde54491core: new benchmark snippet for operations on reals. by 2011-09-22 16:16:58 +0000 -
cdc155c4core: allow to cast reals to doubles in addition to floats. by 2011-09-22 16:16:51 +0000 -
325ea948core: implement division of reals and change their default precision to 32 bigits. by 2011-09-22 16:16:42 +0000 -
93479c28core: implement real subtraction. by 2011-09-22 07:16:05 +0000 -
a173e49dcore: implement comparison operators for reals. by 2011-09-22 07:16:02 +0000 -
48f00defcore: fix real unary minus and handle negative numbers in additions. by 2011-09-22 07:15:59 +0000 -
884978c7core: handle zero, negative zero and infinite in the real constructor, and add a test suite check for unary minus. by 2011-09-22 07:15:56 +0000 -
dbb0463fcore: start working on the extended floating point addition. by 2011-09-21 17:10:31 +0000 -
202ac6aacore: switch real mantissa to uint16_t instead of uint32_t to ease the multiplication. by 2011-09-21 17:10:21 +0000 -
5521b9fdbuild: start adding Xbox 360 configuration settings to the VS projects. by 2011-09-20 16:07:31 +0000 -
5aedae49core: add missing real.cpp unit test. by 2011-09-20 12:16:05 +0000 -
1d9507bacore: remove spurious dependency between androidapp.cpp and monsterz. by 2011-09-20 12:13:18 +0000 -
5fc29896core: start working on a "real" class for arbitrarily sized floats. by 2011-09-20 12:03:55 +0000 -
cc2bdf0btest: fix the benchmark suite compilation on SNC. by 2011-09-20 12:03:46 +0000 -
fe0d76cebuild: fix PS3 and Android build; it was broken since the ImageData refactor. by 2011-09-19 00:07:41 +0000 -
a59f2efbbuild: bring the iOS port up to date by synchronising new source file names with the project file. by 2011-09-19 00:01:03 +0000 -
abbf9ba4build: fix a shitload of build errors introduced by the Image refactoring and the OS X port. by 2011-09-18 23:44:14 +0000 -
35691729osx: link with the proper OpenGL libraries on OS X. by 2011-09-18 23:32:18 +0000 -
2435bd47osx: automake does not support .mm files yet, so rename ios-image.mm back to ios-image.cpp, but disable it on OS X because UIKit isn't available. by 2011-09-18 21:01:26 +0000 -
11dce0c6ios: rename iOS image loader to .mm. by 2011-09-18 19:18:49 +0000 -
bcea65d2core: remove calls to exit() in image loaders since they can now report on errors. by 2011-09-18 19:13:59 +0000 -
c61ae1ffcore: implement dynamic registration of image loaders. by 2011-09-18 19:13:56 +0000 -
73e9bfddbuild: fix GLEW usage to use <glew.h>, not <GL/glew.h>. by 2011-09-14 18:54:06 +0000 -
b9e2be19build: disable "make check" on win64 because the version of Wine installed on the autobuilders doesn't support it yet. by 2011-09-14 18:53:59 +0000 -
70e3ae82build: do not set --build if not really cross-compiling (eg. building for mingw64 on mingw32). by 2011-09-14 14:36:43 +0000 -
489389abbuild: better clean rule for the test directory. by 2011-09-14 14:36:36 +0000 -
beb141ddbuild: fix Win64 compilation with mingw64, using an updated SDL build. by 2011-09-13 17:19:11 +0000 -
7a3709abbuild: move SDL_image and SDL_mixer binaries to an i686-only directory. by 2011-09-13 16:09:44 +0000 -
53347d96build: fix a typo in the mingw configuration sequence in lol-build and add set -e to the build-mingw scripts to catch possible regressions. by 2011-09-12 16:59:07 +0000 -
247af5b0build: add a Win64 build of libSDL. Useless for now, because SDL_Image and SDL_Mixer are needed and are a lot harder to build. by 2011-09-10 22:37:15 +0000 -
b9e28aa9build: add debug information to lol-build output. by 2011-09-10 19:47:45 +0000 -
160102babuild: new lol-build script for easier autobuilds. by 2011-09-10 19:33:10 +0000 -
301bc29dbuild: add a Win64 cross-build script and upgrade glew to 1.7.0. by 2011-09-10 17:57:28 +0000 -
ba10e59eios: fix iOS build by adding missing files to XCode project. by 2011-09-10 15:49:11 +0000 -
5081f703test: reorganise test suite and benchmark code. by 2011-09-10 14:55:51 +0000 -
17005050test: add unit tests to check whether the build environment behaved as expected. by 2011-09-10 14:55:48 +0000 -
04c4a959core: add our custom isnan() macro to the core headers. by 2011-09-10 14:55:44 +0000 -
7b7d4fe0build: stop defining ANDROID_NDK and check for __ANDROID__ instead. by 2011-09-10 12:42:13 +0000 -
69648f38build: add LolUnit to the build rules. by 2011-09-10 12:42:09 +0000 -
5a5b8112lolunit: rename a few methods for consistency. by 2011-09-09 14:47:56 +0000 -
2e314ddclolunit: protect against multiple registration of the same fixture type. by 2011-09-09 14:39:00 +0000 -
8341fd4fbuild: disable RTTI, we don’t need it and won’t use it. by 2011-09-09 14:38:57 +0000 -
75d8811clolunit: rename test runner’s run() method to Run() for consistency. by 2011-09-09 14:38:54 +0000 -
8fba1a07lolunit: add some comments to document the framework internals. by 2011-09-08 21:39:08 +0000 -
00b8d5c2test: improve matrix and vector uni tests. by 2011-09-08 21:02:19 +0000 -
bcc62b54lolunit: add several helper macros for various comparison checks. by 2011-09-08 21:02:16 +0000 -
70fdf1cacore: add std::ostream operators for vector and matrix classes. by 2011-09-08 20:53:45 +0000 -
c38732adlolunit: tweak output formatting so that it more closely matches CppUnit. by 2011-09-08 17:12:16 +0000 -
2290b7b9lolunit: add a LOLUNIT_SET_CONTEXT / LOLUNIT_UNSET_CONTEXT macro pair. by 2011-09-08 12:17:00 +0000 -
e34856fbcore: handle denormal halfs properly on the PS3; it's denormal floats that we do not care about. by 2011-09-08 12:16:52 +0000 -
62b5838flolunit: reverse the expected/actual logic. by 2011-09-08 07:18:31 +0000 -
4d99340dlolunit: fix an uninitialised member in TestCase. by 2011-09-08 06:58:31 +0000 -
916d6081core: define FP_USE on exotic architectures. by 2011-09-08 01:47:03 +0000 -
b64d051elolunit: free all the memory we allocated to avoid a minor memory leak, and slightly tweak the API syntax. by 2011-09-08 01:35:55 +0000 -
5bd1a0f8build: add -j6 to our build scripts' "make" invocations. by 2011-09-08 01:35:52 +0000 -
1569f162core: mediocre workaround for mingw32 who ignores our FP_USE macro. by 2011-09-07 22:21:32 +0000 -
5ab9eb3ctest: get rid of the need for LOLUNIT_SETUP_FIXTURE(). by 2011-09-07 22:21:29 +0000 -
579f77bdtest: make LolUnit's output match CppUnit's more closely. by 2011-09-07 22:21:26 +0000 -
a374127etest: replace CppUnit with our LolUnit test suite. by 2011-09-07 17:17:07 +0000 -
0e622822test: add a few Visual Studio 2010 project files and fix testsuite code so that it builds properly. by 2011-09-07 17:16:58 +0000 -
14a3e1f7core: fix a bug on the PS3 version of lol_sin(). by 2011-09-07 17:16:47 +0000 -
e9296e79core: implement the FP_USE() macro on x86, amd64 and powerpc. by 2011-09-07 17:16:15 +0000 -
b3e2429bcore: avoid returning to memory when giving GCC floating point hints. by 2011-09-06 06:43:06 +0000 -
5454005ddebug: various improvements to the test programs. by 2011-09-06 06:43:03 +0000 -
e0067142build: new Visual Studio build files, for the engine core and the debug quad application. by 2011-09-05 17:18:15 +0000 -
320e70d8core: a few compilation fixes for non-GCC compilers. by 2011-09-05 17:18:06 +0000 -
b5cb0dffdebug: fix PS3 and Android compilation. by 2011-09-04 21:02:36 +0000 -
9eeef663debug: add point sprite examples. by 2011-09-04 20:11:43 +0000 -
5feadd8bcore: fix PS3 compilation; the lol_fdiv implementation was missing. by 2011-09-04 13:46:16 +0000 -
bf164defcore: fix an accuracy error in the cos() part of lol_sincos(). by 2011-09-04 12:28:59 +0000 -
2489233acore: improve tan() accuracy by tweaking higher order Taylor coefficients. by 2011-09-04 12:28:56 +0000 -
82d81256core: implement accelerated lol_sincos() and lol_tan(). by 2011-09-04 12:28:52 +0000 -
5be195a8core: slightly reorganise Taylor series in sin() and cos(). by 2011-09-03 22:06:00 +0000 -
49f9c59fcore: implement accelerated cos(). by 2011-09-03 22:05:57 +0000 -
a65c472fcore: add a code shortcut for sin() on platforms that have cheap branches. by 2011-09-03 17:45:16 +0000 -
148f6838test: replace lol-bench with lol-bench$(EXEEXT) in the Makefile to allow "make bench" to work when cross-compiling for Windows. by 2011-09-03 17:45:13 +0000 -
6ed53d98test: do the trigonometry tests on doubles instead of floats. by 2011-09-03 17:45:10 +0000 -
35454b8ftest: add an empty sandbox program and update .gitignore. by 2011-09-03 07:47:52 +0000 -
3402db36optim: better isnan() reimplementation. by 2011-09-02 11:38:54 +0000 -
824551e4optim: split the Taylor series calculation into two separate values. by 2011-09-02 11:36:42 +0000 -
de496b15core: remove one multiplication and one selection from the optimised version of sinf(). by 2011-09-02 00:28:41 +0000