magnum

Commit Graph

Author	SHA1	Message	Date
Vladimír Vondruš	5b1463fa0c	Math: test constexpr methods on DualQuaternion.	13 years ago
Vladimír Vondruš	dccfc1f8ba	Math: test constexpr methods on Quaternion.	13 years ago
Vladimír Vondruš	184bf38fc1	Math: test constexpr methods on Dual.	13 years ago
Vladimír Vondruš	e2591253ae	Math: test constexpr swizzle().	13 years ago
Vladimír Vondruš	227517ff89	Math: added DualQuaternion::transformPoint().	13 years ago
Vladimír Vondruš	7e0d8d119f	Math: ability to normalize DualQuaternion. Also hope that this is proper implementation.	13 years ago
Vladimír Vondruš	12c786c408	Math: renamed DualQuaternion::norm() to length(), added lengthSquared(). Hopefully this is properly implemented and properly named. On the other hand, having length everywhere (vectors, quaternions, complex numbers) and norm only at one place is inconsistent.	13 years ago
Vladimír Vondruš	f1da7a08b4	Math: added sqrt() overload for Dual.	13 years ago
Vladimír Vondruš	fec6822187	Math: added sqrt() scalar/vector overload.	13 years ago
Vladimír Vondruš	83346ce023	Math: converting DualQuaternion to matrix. Also updated and simplified Quaternion to matrix test.	13 years ago
Vladimír Vondruš	07c0c4d845	Math: added tests for composed dual quaternion parts. To be sure everything is implemented properly.	13 years ago
Vladimír Vondruš	a49bb0d718	Math: convenience functions for transforming vectors with matrices. It's now possible to conveniently transform 2D vectors and points with 3x3 matrices and 3D vectors/points with 4x4 matrices. Previous most low-level solution: Matrix4 m; Vector3 v; Vector3 a = (mVector4(v, 1.0f)).xyz(); Vector4 b = (mVector4(v, 0.0f)).xyz(); Another, more generalized solution for points was with Point2D/Point3D, adding a lot of confusion (what is that class and what does .vector()?): Vector3 a = (mPoint3D(v)).vector(); And the worst solution was with generic 2D/3D code (WTF!): auto a = (mtypename DimensionTraits::PointType(v)).vector(); Now it is just this, similar for both dimensions: Vector3 a = m.transformPoint(v); Vector3 b = m.transformVector(v); Note that transformation three-component vectors with 3x3 matrices or four-component vectors with 4x4 matrices is easy enough so it doesn't need any special convenience functions whatsoever: Vector3 c = m.rotation()*v;	13 years ago
Vladimír Vondruš	0f554a49f6	Math: convenience function to transform points with DualQuaternion.	13 years ago
Vladimír Vondruš	0f130ad55d	Math: simplified Quaternion::rotateVector*() tests. Comparing to Matrix4 transformation.	13 years ago
Vladimír Vondruš	03ae82b448	Math: creating DualQuaternion from vector.	13 years ago
Vladimír Vondruš	3ad264893e	Math: creating rotation and translation DualQuaternion.	13 years ago
Vladimír Vondruš	4c1608378a	Math: document that Quaternion::rotationAxis() might return NaN vector.	13 years ago
Vladimír Vondruš	5615a4c7f6	Math: initial implementation of DualQuaternion class. Currently practically unusable.	13 years ago
Vladimír Vondruš	3f6a2d5ae8	Math: dual number implementation.	13 years ago
Vladimír Vondruš	5d6b2b4ecd	Math: properly test Quaternion comparison.	13 years ago
Vladimír Vondruš	95982e5176	Math: various code cleanup.	13 years ago
Vladimír Vondruš	a0b8dcc067	Math: rename Quaternion::fromRotation() to rotation(). Now it is similar in usage to Matrix4 functions and is now also crossreferenced in documentation. Also updated the test to also check assertion.	13 years ago
Vladimír Vondruš	93ac7f0d3c	Math: rotating vectors with Quaternion.	13 years ago
Vladimír Vondruš	730c09567d	Math: ability to explicitly create Quaternion for given vector.	13 years ago
Vladimír Vondruš	7c81755152	Math: slightly reorganize Quaternion test.	13 years ago
Vladimír Vondruš	0144c38a54	Math: moved lerp() from Vector to Functions.h. It does nothing special for vectors (unlike e.g. quaternion LERP) and scalar overload is useful too.	13 years ago
Vladimír Vondruš	ce2a07a09a	Math: slightly faster Vector::projectedOntoNormalized(). The assertion will still compute the dot product (but not in release builds).	13 years ago
Vladimír Vondruš	8a63939134	Math: added vector/scalar abs() function.	13 years ago
Vladimír Vondruš	f517a000ed	Math: follow Vector changes in BoolVector. Added constexpr "filling" constructor, updated tests, added constexpr test.	13 years ago
Vladimír Vondruš	bbea562a6c	Math: properly testing all constexpr methods in RectangularMatrix. Fixed data() method, it doesn't need to use reinterpret_cast.	13 years ago
Vladimír Vondruš	bbc98746e4	Math: properly testing all constexpr methods in Vector. Fixed conversion constructor, as pointer chasings are not allowed in constexpr functions.	13 years ago
Vladimír Vondruš	f7a8641081	Math: fixed one-argument Vector constructors. It is now possible to implicitly create one-element Vector and also explicitly fill more-element Vector with one value without any ambiguous overload conflicts: Vector<1, int> a1 = 1; // calls implicit constructor //Vector<3, int> a3 = 1; // error! Vector<1, int> b1(1); // still calls the implicit constructor, the // explicit is disabled for one-element vector Vector<3, int> b3(1); // calls the explicit "filling" constructor, // the implicit is disabled for only one argument The downside of this is that now specifying improper element count in constructor doesn't lead to static_assert with human readable error, but rather cryptic "no match" error.	13 years ago
Vladimír Vondruš	0f547d7e8e	Math: added RectangularMatrix::sum(),product(),{min,max}(),{min,max}Abs().	13 years ago
Vladimír Vondruš	60f86b8ea0	Math: added Vector::{min,max}Abs().	13 years ago
Vladimír Vondruš	7b507d0485	Math: improve Vector::min()/max() tests.	13 years ago
Vladimír Vondruš	e58962dc43	Don't use fully qualified Corrade::Utility::Debug when not needed.	13 years ago
Vladimír Vondruš	37955c468f	Math: creating matrix from diagonal, retrieving values on diagonal.	13 years ago
Vladimír Vondruš	0274ad7d56	Math: explicit one-parameter PointD constructor. Was causing improper implicit conversions, such as here (example directly from unit tests, where it was unintentionally used): Vector3 normal; Matrix4 transformation; auto transformedNormal = transformationnormal; Not only that it was possible to multiply 4x4 matrix with 3-component vector, but the resulting type was Point3D which was absolutely confusing. Currently it must be explicitly converted: transformedNormal = transformation*Point3D(normal);	13 years ago
Vladimír Vondruš	799bc629eb	Math: matrix/vector rework, part 4: using BoolVector for comparisons. The comparisons have also SIMD instructions returning bool vector, this future-proofs them. Also it isn't confusing anymore (a < b is true when all are less or just one?).	13 years ago
Vladimír Vondruš	dcb4595e6e	Math: new class BoolVector. Will handle output from various comparison operators in Vector.	13 years ago
Vladimír Vondruš	9bbc656da5	Math: creating Vector from one value is broken. Added "expect fail" test, it is not the priority now (and it is handled correctly in subclasses).	13 years ago
Vladimír Vondruš	5819dd4bd7	Math: matrix/vector rework, part 3: conversion as explicit constructor. Also updated all dependent classes to follow the change, such as Color and Rectangle. Backwards compatibility for GCC 4.6 (with lack of support for delegating constructors) will be done as non-constexpr constructor using operator=().	13 years ago
Vladimír Vondruš	baeadc9663	Math: matrix/vector rework, part 2: matrix as array of column vectors. Overall architecture is simplififed with this change and also it's not needed to use reinterpret_cast in matrix internals anymore, thus there is no need for operator() and [][] works now always as expected without any risk of GCC misoptimizations. On the other side, constructing matrix from list of elements is not possible anymore. You have to specify the elements as list of column vectors, which might be less convenient to write, but it helps to distinguish what is column and what is row: Matrix<2, int> a(1, 2, // before 3, 4); Matrix<2, int> a(Vector<2, int>(1, 2), // now Vector<2, int>(3, 4)); For some matrix specializations (i.e. Matrix3 and Matrix4) it is possible to use list-initialization instead of explicit type specification: Matrix<3, int>({1, 2, 3}, {4, 5, 6}, {7, 8, 9}); I didn't yet figure out how to properly implement the general (constexpr) constructor to also take lists, so it's a bit ugly for now. Matrix operations are now done column-wise, which should help with future SIMD implementations, documentation is also updated accordingly. I also removed forgotten remains of matrix/matrix operator*=(), which can be confusing, as the multiplication is not commutative. Why it is not present is explained in `d9c900f076`.	13 years ago
Vladimír Vondruš	0ed0f261b4	Math: matrix/vector rework, part 1: synchronized unit tests. They now test the same functionality in two places, but the functionality will be implemented differently for matrix and vector.	13 years ago
Vladimír Vondruš	b7b955289a	Math: Vector::lerp() can take `t` of arbitrary type. Usable when interpolating e.g. colors in denormalized integral form, as `t` could be only `0` or `1` in that case, which is pretty unusable.	13 years ago
Vladimír Vondruš	263d8351b6	Math: vector overloads for normalize() and denormalize().	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	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	7d0a523618	Moved projection matrix computation to Math namespace.	14 years ago

1 2 3 4 5

228 Commits (deb54fc92f2b76ae562cd2a6c707fbc844e6bbc0)