Files
FlaxEngine/Source/Engine/Core/Math/Matrix3x3.cpp

333 lines
11 KiB
C++

// Copyright (c) 2012-2024 Wojciech Figat. All rights reserved.
#include "Matrix3x3.h"
#include "Matrix.h"
#include "Quaternion.h"
#include "../Types/String.h"
#include <intrin.h>
#include <emmintrin.h>
const Matrix3x3 Matrix3x3::Zero(0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f);
const Matrix3x3 Matrix3x3::Identity(
1.0f, 0.0f, 0.0f,
0.0f, 1.0f, 0.0f,
0.0f, 0.0f, 1.0f);
Matrix3x3::Matrix3x3(const Matrix& matrix)
{
Platform::MemoryCopy(&M11, &matrix.M11, sizeof(Float3));
Platform::MemoryCopy(&M21, &matrix.M21, sizeof(Float3));
Platform::MemoryCopy(&M31, &matrix.M31, sizeof(Float3));
}
String Matrix3x3::ToString() const
{
return String::Format(TEXT("{}"), *this);
}
float Matrix3x3::GetDeterminant() const
{
return M11 * M22 * M33 + M12 * M23 * M31 + M13 * M21 * M32 - M13 * M22 * M31 - M12 * M21 * M33 - M11 * M23 * M32;
}
void Matrix3x3::NormalizeScale()
{
const float scaleX = 1.0f / Float3(M11, M21, M31).Length();
const float scaleY = 1.0f / Float3(M12, M22, M32).Length();
const float scaleZ = 1.0f / Float3(M13, M23, M33).Length();
M11 *= scaleX;
M21 *= scaleX;
M31 *= scaleX;
M12 *= scaleY;
M22 *= scaleY;
M32 *= scaleY;
M13 *= scaleZ;
M23 *= scaleZ;
M33 *= scaleZ;
}
void Matrix3x3::Invert(const Matrix3x3& value, Matrix3x3& result)
{
const float d11 = value.M22 * value.M33 + value.M23 * -value.M32;
const float d12 = value.M21 * value.M33 + value.M23 * -value.M31;
const float d13 = value.M21 * value.M32 + value.M22 * -value.M31;
float det = value.M11 * d11 - value.M12 * d12 + value.M13 * d13;
if (Math::Abs(det) < ZeroTolerance)
{
result = Zero;
return;
}
det = 1.0f / det;
const float d21 = value.M12 * value.M33 + value.M13 * -value.M32;
const float d22 = value.M11 * value.M33 + value.M13 * -value.M31;
const float d23 = value.M11 * value.M32 + value.M12 * -value.M31;
const float d31 = value.M12 * value.M23 - value.M13 * value.M22;
const float d32 = value.M11 * value.M23 - value.M13 * value.M21;
const float d33 = value.M11 * value.M22 - value.M12 * value.M21;
result = Matrix3x3(
+d11 * det,
-d21 * det,
+d31 * det,
-d12 * det,
+d22 * det,
-d32 * det,
+d13 * det,
-d23 * det,
+d33 * det
);
}
void Matrix3x3::Transpose(const Matrix3x3& value, Matrix3x3& result)
{
result = Matrix3x3(
value.M11,
value.M21,
value.M31,
value.M12,
value.M22,
value.M32,
value.M13,
value.M23,
value.M33
);
}
void Matrix3x3::Add(const Matrix3x3& left, const Matrix3x3& right, Matrix3x3& result)
{
result.M11 = left.M11 + right.M11;
result.M12 = left.M12 + right.M12;
result.M13 = left.M13 + right.M13;
result.M21 = left.M21 + right.M21;
result.M22 = left.M22 + right.M22;
result.M23 = left.M23 + right.M23;
result.M31 = left.M31 + right.M31;
result.M32 = left.M32 + right.M32;
result.M33 = left.M33 + right.M33;
}
void Matrix3x3::Subtract(const Matrix3x3& left, const Matrix3x3& right, Matrix3x3& result)
{
result.M11 = left.M11 - right.M11;
result.M12 = left.M12 - right.M12;
result.M13 = left.M13 - right.M13;
result.M21 = left.M21 - right.M21;
result.M22 = left.M22 - right.M22;
result.M23 = left.M23 - right.M23;
result.M31 = left.M31 - right.M31;
result.M32 = left.M32 - right.M32;
result.M33 = left.M33 - right.M33;
}
void Matrix3x3::Multiply(const Matrix3x3& left, float right, Matrix3x3& result)
{
result.M11 = left.M11 * right;
result.M12 = left.M12 * right;
result.M13 = left.M13 * right;
result.M21 = left.M21 * right;
result.M22 = left.M22 * right;
result.M23 = left.M23 * right;
result.M31 = left.M31 * right;
result.M32 = left.M32 * right;
result.M33 = left.M33 * right;
}
void Matrix3x3::Multiply(const Matrix3x3& left, const Matrix3x3& right, Matrix3x3& result)
{
#if true
{
// First : naive solution with but with some tricks to make compiler (MSVC) behave
//* Note that, in this case, manually unrolling the loop helps
//* as the compiler can't auto-vectorize non-contagious memory access
float* __restrict const matData = result.Raw;
//Mat matC{ matB.width, matA.height, matB.rowSpan, matData };
for (int rowC = 0; rowC < 3; ++rowC) {
for (int colC = 0; colC < 3; ++colC) {
// an independent, local accumulator.
float accumulate = 0;
int pos = 0;
// manual unrolling IS helpful in this case
for (; pos < 3 - 4; pos += 4) {
accumulate += left.Raw[rowC * 3 + pos] *
right.Raw[pos * 3 + colC] +
left.Raw[rowC * 3 + pos + 1] *
right.Raw[(pos + 1) * 3 + colC] +
left.Raw[rowC * 3 + pos + 2] *
right.Raw[(pos + 2) * 3 + colC] +
left.Raw[rowC * 3 + pos + 3] *
right.Raw[(pos + 3) * 3 + colC];
}
for (; pos < 3; ++pos) {
accumulate += left.Raw[rowC * 3 + pos] *
right.Raw[pos * 3 + colC];
}
matData[rowC * 3 + colC] = accumulate;
}
}
}
#else
/*
__m256i vec_multi_res = _mm256_setzero_si256();
__m256i vec_mat1 = _mm256_setzero_si256();
__m256i vec_mat2 = _mm256_setzero_si256();
int i, j, k;
for (i = 0; i < 3; i++)
{
for (j = 0; j < 3; ++j)
{
//Stores one element in mat1 and use it in all computations needed before proceeding
//Stores as vector to increase computations per cycle
vec_mat1 = _mm256_set1_epi32(left.Values[j][i]);
for (k = 0; k < 3; k += 8)
{
vec_mat2 = _mm256_loadu_si256((__m256i*) & right.Values[k][j]); //Stores row of second matrix (eight in each iteration)
vec_multi_res = _mm256_loadu_si256((__m256i*) & result.Values[k][i]); //Loads the result matrix row as a vector
vec_multi_res = _mm256_add_epi32(vec_multi_res, _mm256_mullo_epi32(vec_mat1, vec_mat2));//Multiplies the vectors and adds to th the result vector
_mm256_storeu_si256((__m256i*) & result.Values[k][i], vec_multi_res); //Stores the result vector into the result array
}
}
}
*/
result = Matrix3x3(
left.M11 * right.M11 + left.M12 * right.M21 + left.M13 * right.M31,
left.M11 * right.M12 + left.M12 * right.M22 + left.M13 * right.M32,
left.M11 * right.M13 + left.M12 * right.M23 + left.M13 * right.M33,
left.M21 * right.M11 + left.M22 * right.M21 + left.M23 * right.M31,
left.M21 * right.M12 + left.M22 * right.M22 + left.M23 * right.M32,
left.M21 * right.M13 + left.M22 * right.M23 + left.M23 * right.M33,
left.M31 * right.M11 + left.M32 * right.M21 + left.M33 * right.M31,
left.M31 * right.M12 + left.M32 * right.M22 + left.M33 * right.M32,
left.M31 * right.M13 + left.M32 * right.M23 + left.M33 * right.M33
);
/*Matrix3x3 result2 = Matrix3x3(
left.M11 * right.M11 + left.M12 * right.M21 + left.M13 * right.M31,
left.M11 * right.M12 + left.M12 * right.M22 + left.M13 * right.M32,
left.M11 * right.M13 + left.M12 * right.M23 + left.M13 * right.M33,
left.M21 * right.M11 + left.M22 * right.M21 + left.M23 * right.M31,
left.M21 * right.M12 + left.M22 * right.M22 + left.M23 * right.M32,
left.M21 * right.M13 + left.M22 * right.M23 + left.M23 * right.M33,
left.M31 * right.M11 + left.M32 * right.M21 + left.M33 * right.M31,
left.M31 * right.M12 + left.M32 * right.M22 + left.M33 * right.M32,
left.M31 * right.M13 + left.M32 * right.M23 + left.M33 * right.M33
);
ASSERT(result2 == result);*/
#endif
}
void Matrix3x3::Divide(const Matrix3x3& left, float right, Matrix3x3& result)
{
ASSERT(!Math::IsZero(right));
const float inv = 1.0f / right;
result.M11 = left.M11 * inv;
result.M12 = left.M12 * inv;
result.M13 = left.M13 * inv;
result.M21 = left.M21 * inv;
result.M22 = left.M22 * inv;
result.M23 = left.M23 * inv;
result.M31 = left.M31 * inv;
result.M32 = left.M32 * inv;
result.M33 = left.M33 * inv;
}
void Matrix3x3::Divide(const Matrix3x3& left, const Matrix3x3& right, Matrix3x3& result)
{
result.M11 = left.M11 / right.M11;
result.M12 = left.M12 / right.M12;
result.M13 = left.M13 / right.M13;
result.M21 = left.M21 / right.M21;
result.M22 = left.M22 / right.M22;
result.M23 = left.M23 / right.M23;
result.M31 = left.M31 / right.M31;
result.M32 = left.M32 / right.M32;
result.M33 = left.M33 / right.M33;
}
void Matrix3x3::RotationQuaternion(const Quaternion& rotation, Matrix3x3& result)
{
const float xx = rotation.X * rotation.X;
const float yy = rotation.Y * rotation.Y;
const float zz = rotation.Z * rotation.Z;
const float xy = rotation.X * rotation.Y;
const float zw = rotation.Z * rotation.W;
const float zx = rotation.Z * rotation.X;
const float yw = rotation.Y * rotation.W;
const float yz = rotation.Y * rotation.Z;
const float xw = rotation.X * rotation.W;
result.M11 = 1.0f - 2.0f * (yy + zz);
result.M12 = 2.0f * (xy + zw);
result.M13 = 2.0f * (zx - yw);
result.M21 = 2.0f * (xy - zw);
result.M22 = 1.0f - 2.0f * (zz + xx);
result.M23 = 2.0f * (yz + xw);
result.M31 = 2.0f * (zx + yw);
result.M32 = 2.0f * (yz - xw);
result.M33 = 1.0f - 2.0f * (yy + xx);
}
void Matrix3x3::Decompose(Float3& scale, Matrix3x3& rotation) const
{
// Scaling is the length of the rows
scale = Float3(
Math::Sqrt(M11 * M11 + M12 * M12 + M13 * M13),
Math::Sqrt(M21 * M21 + M22 * M22 + M23 * M23),
Math::Sqrt(M31 * M31 + M32 * M32 + M33 * M33));
// If any of the scaling factors are zero, than the rotation matrix can not exist
rotation = Identity;
if (scale.IsAnyZero())
return;
// Calculate an perfect orthonormal matrix (no reflections)
const auto at = Float3(M31 / scale.Z, M32 / scale.Z, M33 / scale.Z);
const auto up = Float3::Cross(at, Float3(M11 / scale.X, M12 / scale.X, M13 / scale.X));
const auto right = Float3::Cross(up, at);
rotation.SetRight(right);
rotation.SetUp(up);
rotation.SetBackward(at);
// In case of reflexions
scale.X = Float3::Dot(right, GetRight()) > 0.0f ? scale.X : -scale.X;
scale.Y = Float3::Dot(up, GetUp()) > 0.0f ? scale.Y : -scale.Y;
scale.Z = Float3::Dot(at, GetBackward()) > 0.0f ? scale.Z : -scale.Z;
}
void Matrix3x3::Decompose(Float3& scale, Quaternion& rotation) const
{
Matrix3x3 rotationMatrix;
Decompose(scale, rotationMatrix);
Quaternion::RotationMatrix(rotationMatrix, rotation);
}
bool Matrix3x3::operator==(const Matrix3x3& other) const
{
return
Math::NearEqual(M11, other.M11) &&
Math::NearEqual(M12, other.M12) &&
Math::NearEqual(M13, other.M13) &&
Math::NearEqual(M21, other.M21) &&
Math::NearEqual(M22, other.M22) &&
Math::NearEqual(M23, other.M23) &&
Math::NearEqual(M31, other.M31) &&
Math::NearEqual(M32, other.M32) &&
Math::NearEqual(M33, other.M33);
}