

@1036

8 years 
sam 
build: allow to build without the PS3specific files.



@1035

8 years 
sam 
core: disallow casting eg. a vec3 into a vec2. Use the xy() getter instead.



@1034

8 years 
sam 
build: remove cruft from the repository.



@1028

8 years 
sam 
test: minor fixes to the test suite: disable optimisation tests in …



@1027

8 years 
sam 
test: minor Remez algorithm tests.



@1020

8 years 
sam 
core: improve sin() and exp() performance by replacing divisions with …



@1019

8 years 
sam 
core: implement real methods cbrt(), log2(), exp2(), and copysign().



@1017

8 years 
sam 
core: fix an accuracy error in real::re() and real::sqrt() introduced …



@1015

8 years 
sam 
test: random useless crap.



@1014

8 years 
sam 
test: various improvements to the Remez exchange solver.



@1012

8 years 
sam 
test: more Remez exchange experimentations.



@1011

8 years 
sam 
test: more Remez exchange experimentations.



@1010

8 years 
sam 
test: allow to perform Remez solving on an arbitrary range.



@1009

8 years 
sam 
test: some refactoring in the Remez solver to prepare multiple …



@1008

8 years 
sam 
test: add missing <cstdio> include in pi test program.



@1007

8 years 
sam 
test: use namespace "std" to avoid PS3 build errors.



@1005

8 years 
sam 
test: add support for relative error in the Remez exchange test program.



@1003

8 years 
sam 
core: implement round() and fmod() for real numbers.



@1002

8 years 
sam 
core: implement floor() and ceil() for real numbers.



@1001

8 years 
sam 
core: implement asin() and acos() for real numbers and add unit tests …



@998

8 years 
sam 
core: add real constants for 0, 1, 2 and 10, and reorder static …



@997

9 years 
sam 
core: add boolean operators on real numbers, add unit tests for that, …



@996

9 years 
sam 
test: minor updates to the Pi program (now almost deprecated) and the …



@994

9 years 
sam 
core: implement shift operators for reals; they're useful for fast …



@991

9 years 
sam 
test: the Remez algorithm is now almost functional.



@989

9 years 
sam 
test: more work on the Remez exchange algorithm.



@985

9 years 
sam 
test: start working on a Remez exchange algorithm implementation so …



@984

9 years 
sam 
test: add a 10line example program that computes Pi to the 150th digit.



@982

9 years 
sam 
core: add rounding to real>double conversion, fix a bug in the …



@979

9 years 
sam 
test: give more explicit names to unit test cases.



@978

9 years 
sam 
test: add simple unit tests for image loading.



@977

9 years 
sam 
core: new benchmark snippet for operations on reals.



@976

9 years 
sam 
core: allow to cast reals to doubles in addition to floats.



@975

9 years 
sam 
core: implement division of reals and change their default precision …



@974

9 years 
sam 
core: implement real subtraction.



@973

9 years 
sam 
core: implement comparison operators for reals.



@972

9 years 
sam 
core: fix real unary minus and handle negative numbers in additions.



@971

9 years 
sam 
core: handle zero, negative zero and infinite in the real constructor, …



@970

9 years 
sam 
core: start working on the extended floating point addition.



@969

9 years 
sam 
core: switch real mantissa to uint16_t instead of uint32_t to ease the …



@967

9 years 
sam 
core: add missing real.cpp unit test.



@965

9 years 
sam 
core: start working on a "real" class for arbitrarily sized floats.



@964

9 years 
sam 
test: fix the benchmark suite compilation on SNC.



@961

9 years 
sam 
build: fix a shitload of build errors introduced by the Image …



@959

9 years 
sam 
build: add svn:ignore properties where needed.



@958

9 years 
sam 
osx: automake does not support .mm files yet, so rename iosimage.mm …



@951

9 years 
sam 
build: better clean rule for the test directory.



@945

9 years 
sam 
build: new lolbuild script for easier autobuilds.



@942

9 years 
sam 
test: reorganise test suite and benchmark code.



@941

9 years 
sam 
test: add unit tests to check whether the build environment behaved as …



@940

9 years 
sam 
core: add our custom isnan() macro to the core headers.



@937

9 years 
sam 
lolunit: rename a few methods for consistency.



@934

9 years 
sam 
lolunit: rename test runner’s run() method to Run() for consistency.



@931

9 years 
sam 
test: improve matrix and vector uni tests.



@927

9 years 
sam 
lolunit: add a LOLUNIT_SET_CONTEXT / LOLUNIT_UNSET_CONTEXT macro pair. …



@919

9 years 
sam 
lolunit: free all the memory we allocated to avoid a minor memory …



@914

9 years 
sam 
test: get rid of the need for LOLUNIT_SETUP_FIXTURE().



@913

9 years 
sam 
test: make LolUnit's output match CppUnit's more closely.



@912

9 years 
sam 
test: replace CppUnit with our LolUnit test suite.
CppUnit is really …



@911

9 years 
sam 
test: add a few Visual Studio 2010 project files and fix testsuite …



@907

9 years 
sam 
debug: various improvements to the test programs.



@903

9 years 
sam 
debug: add point sprite examples.



@900

9 years 
sam 
core: improve tan() accuracy by tweaking higher order Taylor coefficients.



@899

9 years 
sam 
core: implement accelerated lol_sincos() and lol_tan().



@897

9 years 
sam 
core: implement accelerated cos().



@895

9 years 
sam 
test: replace lolbench with lolbench$(EXEEXT) in the Makefile to …



@894

9 years 
sam 
test: do the trigonometry tests on doubles instead of floats.



@893

9 years 
sam 
test: add an empty sandbox program and update .gitignore.



@892

9 years 
sam 
optim: better isnan() reimplementation.



@890

9 years 
sam 
core: remove one multiplication and one selection from the optimised …



@889

9 years 
sam 
test: new test for trigonometric functions.



@887

9 years 
sam 
test: implement a custom isnan()
This version is faster (about 25% on …



@886

9 years 
sam 
test: add benchmark tests for trigonometry functions.



@881

9 years 
sam 
contrib: rebuild the Win32 CppUnit under Linux and fix the includes.
…



@880

9 years 
sam 
test: add matrix code to the benchmark program.



@879

9 years 
sam 
core: tune the half precision code so that the best variants are being …



@878

9 years 
sam 
build: clean up .self files and allow to build the benchmark on the PS3.



@877

9 years 
sam 
test: improve the benchmark code to measure a lot more half precision …



@875

9 years 
sam 
test: add a fallback for isnan() when not defined by the system includes.



@874

9 years 
sam 
core: rename half::isnan() etc. to half::is_nan() because "isnan" can …



@872

9 years 
sam 
core: minor refactoring in the float / half conversions to accomodate …



@871

9 years 
sam 
core: implement new half / float operations and the associated tests.



@870

9 years 
sam 
test: augment the half precision numbers testsuite.



@868

9 years 
sam 
test: add first test routines for the half class.



@865

9 years 
sam 
core: move platformspecific code to specific directories.



@863

9 years 
sam 
core: rename vec2i to ivec2 etc. to better match GLSL.



@817

9 years 
sam 
test: add a unit test for the recent vector promotion problem.



@770

9 years 
sam 
test: start writing really minimalist GL rendering tests.



@769

9 years 
sam 
test: add a debug quad test application.



@768

9 years 
sam 
core: add the unary minus operator to vectors.



@757

9 years 
sam 
math: implement vector comparisons and add minimal unit tests for this …



@686

9 years 
sam 
Put everything in the "lol" namespace. Better late than never.



@664

9 years 
sam 
Get rid of float3, float4, int3 etc. in favour of GLSL types.



@651

9 years 
sam 
Make the float4x4 constructor fill only the matrix's diagonal. Thus …



@648

9 years 
sam 
Add matrix determinant and inverse methods.



@642

9 years 
sam 
Add an optional cppunit dependency for unit tests.
