magnum

Commit Graph

Author	SHA1	Message	Date
Vladimír Vondruš	263d8351b6	Math: vector overloads for normalize() and denormalize().	13 years ago
Vladimír Vondruš	ec23a3d13c	Minor code cleanup.	13 years ago
Vladimír Vondruš	ad9a55a867	Doc++	13 years ago
Vladimír Vondruš	4d6d3fde88	Math: scalar/vector implementation of min(), max() and clamp(). This is gonna be some ugly template hell.	13 years ago
Vladimír Vondruš	bdba6b7221	Math: no need to have two versions of denormalize(), they are the same. Overlooked in `cade175304`.	13 years ago
Vladimír Vondruš	3daefd70a9	Math: simplified internal pow<n>() implementation.	13 years ago
Vladimír Vondruš	cade175304	Math: finally implemented precise signed/unsigned normalization. (De)normalization from/to [0, 1] for unsigned types and from/to [-1, 1] for signed types. -1.0 is always denormalized to min+1. Also updated some tests for Color, as they depended on previous imprecise implementation.	14 years ago
Vladimír Vondruš	7dc0fc4a4c	Math: Renamed Math.h to Functions.h Math.h is reserved for forward declaration header.	14 years ago
Vladimír Vondruš	52300cf402	Math: function for base-2 logarithm. Verbatim copy from Corrade, might be better to have it here too (possibly some SIMD optimizations?).	14 years ago
Vladimír Vondruš	7183110eee	Doc++, @todo++	14 years ago
Vladimír Vondruš	c7d7fdad28	Math: removed done TODO.	14 years ago
Vladimír Vondruš	0f889369f4	No `using namespace std` anywhere. Also added missing std:: prefix to remaining cases of std::size_t, std::[u]int[0-9]+_t, std::sin() etc., std::exit().	14 years ago
Vladimír Vondruš	c1cf94fb46	Using fixed-size integer types also everywhere else. It solves issues with int/long int/long long int ambiguity, which is good.	14 years ago
Vladimír Vondruš	cdc2e5302c	Using fixed-size 32bit integer instead of size_t for log/pow. I want something like "use 32bit on x86, but 64bit on x86_64 and x86_32", but <cstdint> header has nothing like that.	14 years ago
Vladimír Vondruš	eb803200d6	Disabled denormalization test for long double altogether. Fails in Debug, but passes in Release mode on my GCC 4.7 @ x86_64. Not sure what to do about it.	14 years ago
Vladimír Vondruš	fe62f547de	Reduced #includes in headers. Magnum.h now doesn't include anything except OpenGL headers, thus changes in Math library don't trigger recompilation of everything, but only of things really depending on it. Math constants moved to separate file for similar reasons, de-inlined some functions to remove the need for some #includes.	14 years ago
Vladimír Vondruš	3558859cc6	Utility function for clamping value in given limits.	14 years ago
Vladimír Vondruš	d96cb96d60	Documented convenience functions on vectors and angles.	14 years ago
Vladimír Vondruš	eeca90c17c	Added @todos .	14 years ago
Vladimír Vondruš	51f2478b1e	More precise double Math constants. Double has 15-17 significant decimal digits precision, extended the constant to have 15 decimal digits. On the other hand, float has only 6-9 digits, so there is no need to have more than 9. Added just-to-be-sure test for sqrt* constants.	14 years ago
Vladimír Vondruš	3bcaed6322	Using round() in denormalization. Gives more precise results for smaller types, but creates overflow for large types.	14 years ago
Vladimír Vondruš	2553e50444	Functions for normalizing and denormalizing numeric values.	14 years ago
Vladimír Vondruš	45d1cb1225	Anonymized unused function parameters.	14 years ago
Vladimír Vondruš	6f65dd972c	Renamed visibility configuration headers to resemble library names.	14 years ago
Vladimír Vondruš	4af08d04c1	Renamed visibility headers to distinguish them from API headers. Also added note about that to coding style documentation.	14 years ago
Vladimír Vondruš	952f8ebb27	Compiling with -fvisibility=hidden. Reusing macros MAGNUM_EXPORT and friends already used for Windows builds. For exported classes added MAGNUM_LOCAL to private members which are not referenced from any inline function. Added explicit MAGNUM_EXPORT to private members which are referenced. CMake provides its own macro <target>_EXPORTS, using that instead of set_target_properties().	14 years ago
Vladimír Vondruš	7f6f81afc7	Math::Constants are now inline functions instead of static variables. Static variables were compiled into the library and then linked on every use, which was not good for performance.	14 years ago
Vladimír Vondruš	09acfa711f	Documented directories containing various namespaces.	14 years ago
Vladimír Vondruš	2b7aa09520	Added some TODOs.	14 years ago
Vladimír Vondruš	925cc3d2ff	Make Math::pow() templated on type.	14 years ago
Vladimír Vondruš	f0d88fbf9f	Added templated Math::Constants structure to replace PI #define. And added sqrt(2) and sqrt(3) constants while we are at it.	14 years ago
Vladimír Vondruš	1cf69616a3	Documentation updates, "markdownization". All HTML code and Doxygen shortcuts such as @c, @b and @em are now rewritten using Markdown syntax, which makes it more readable. Also updated Doxyfile. Doxygen 1.8 at least is required to generate the documentation now.	14 years ago
Vladimír Vondruš	8efb5a1e42	Added dllimport/dllexport macros for Windows builds.	14 years ago
Vladimír Vondruš	8d525be511	Using TypeTraits::epsilon and TypeTraits::equals instead of EPSILON. Improves comparison performance on integral Matrix and Vector, fixes NaN comparison bug (but infinity bug is still present).	14 years ago
Vladimír Vondruš	e0b4fca4fe	Doc++ Removed doc/namespaces.dox, moved namespace documentation to main files of given namespace.	15 years ago
Vladimír Vondruš	b945f5f68b	Happy New Year.	15 years ago
Vladimír Vondruš	e536c892c4	Added Math::pow() and Math::log() functions. Math::pow() is templated on exponent, so it should be as fast as possible (without recursion).	15 years ago
Vladimír Vondruš	be3dd97d0a	Renamed Math/{constants.h => Math.h}.	15 years ago
Vladimír Vondruš	3c1f5392e8	Useful functions for making angle entering less error-prone. Both functions convert the value to radians at compile time. For example deg(180.0f) is converted to 3.14f. Less intuitive than user-defined literals (C++11 feature), but works everywhere.	15 years ago
Vladimír Vondruš	c7b331d100	(Ahem...) Happy New Year!	15 years ago
Vladimír Vondruš	e746fc88c0	Lowered fuzzy compare precision even more Reason for that was that Matrix(m)*Matrix(m).inverse() == Matrix() shouldn't fail with overprecision.	16 years ago
Vladimír Vondruš	dc420e69fc	Even smaller precision for floating point comparison.	16 years ago
Vladimír Vondruš	c41d6ad582	New namespace Magnum::Math.	16 years ago
Vladimír Vondruš	2a2d8c3dd5	Template vector class.	16 years ago
Vladimír Vondruš	8f0bacd2f8	Function for transposing matrix.	16 years ago
Vladimír Vondruš	09d74422d9	Initial commit	16 years ago

8 Commits (263d8351b6bf2112e66ce3ea330546bd7032eb9a)