VS2012下基于Glut 矩阵变换示例程序:

也可以使用我们自己的矩阵运算来实现OpenGL下的glTranslatef相应的旋转变换。需要注意的是OpenGL下的矩阵是列优先存储的。


示例通过矩阵运算使得圆柱或者甜圈自动绕Y轴旋转,可以单击鼠标右键来弹出菜单选择是否显示坐标轴、正视图或者是透视图、是否打印变换矩阵、显示圆柱还是甜圈。程序用到math3d中的矩阵相关函数。由于绘制的坐标轴并未参加矩阵变换,在运行过程中会发现坐标轴并不会在定时器作用下不断旋转。




源代码:

GlutTransformDemo

// GlutTransformDemo.cpp : 定义控制台应用程序的入口点。
//


#include "stdafx.h"
#include <gl/glut.h>
#include <math.h>
#include "math3d.h"
//圆周率宏
#define GL_PI 3.1415f
//获取屏幕的宽度
GLint SCREEN_WIDTH=0;
GLint SCREEN_HEIGHT=0;
//设置程序的窗口大小
GLint windowWidth=400;
GLint windowHeight=300;
//绕x轴旋转角度
GLfloat xRotAngle=0.0f;
//绕y轴旋转角度
GLfloat yRotAngle=0.0f;
//受支持的点大小范围
GLfloat sizes[2];
//受支持的点大小增量
GLfloat step;
//最大的投影矩阵堆栈深度
GLint iMaxProjectionStackDepth;
//最大的模型视图矩阵堆栈深度
GLint iMaxModeviewStackDepth;
//最大的纹理矩阵堆栈深度
GLint iMaxTextureStackDepth;

GLint iCoordinateaxis=2;//是否显示坐标轴
GLint iProjectionMode=1;//投影模式
GLint iPrintMatrix=1;//是否打印变换矩阵
GLint iCylinder=1;//显示圆柱还是甜圈
void changSize(GLint w,GLint h);

void DrawTorus(M3DMatrix44f mTransform){  
    // 大圆只存在于 xy 平面,  
    // 小圆存在于 xyz 空间中,  
    // 其圆心是大圆圆周上的点。  
    // 小圆环大圆半径方向为起始旋转一周形成的。  
    // 由于 z 轴垂直于 xy 平面,  
    // 又因为大圆的半径位于 xy 平面,  
    // 因此,z 轴垂直于大圆的半径(垂直于面,垂直于线),  
    // 因此,z 轴与大圆的半径方向是正交的。  
    // 小圆位于 z 轴与大圆半径方向形成的平面,  
    // 后面计算具体点的位置是基于上面的描述。  
      
    // 大圆半径  
    GLfloat majorRadius = 55.0f;  
    // 小圆半径  
    GLfloat minorRadius = 15.0f;  
    // 大圆圆周被切分的点数  
    GLint   numMajor = 50;  
    // 小圆圆周被切分的点数  
    GLint   numMinor = 20;  
    M3DVector3f objectVertex;         // Vertex in object/eye space  
    M3DVector3f transformedVertex;    // New Transformed vertex  
    // 每个点对应的弧度数  
    double majorStep = 2.0f*M3D_PI / numMajor;  
    double minorStep = 2.0f*M3D_PI / numMinor;  
    int i, j;  
      
    // 对于大圆上的点进行迭代  
    for (i=0; i<numMajor; ++i)   
        {  
        // 第一个点对应的弧度  
        double a0 = i * majorStep;  
        // 第二个点对应的弧度  
        double a1 = a0 + majorStep;  
        // 第一个点在 x 与 y 轴上的单位长度  
        GLfloat x0 = (GLfloat) cos(a0);  
        GLfloat y0 = (GLfloat) sin(a0);  
        // 第二个点在 x 与 y 轴上的单位长度  
        GLfloat x1 = (GLfloat) cos(a1);  
        GLfloat y1 = (GLfloat) sin(a1);  
  
        glBegin(GL_TRIANGLE_STRIP);  
        // 对小圆上的点进行迭代  
        for (j=0; j<=numMinor; ++j)   
            {  
            // 小圆上点对应的弧度  
            double b = j * minorStep;  
            // 小圆上点在半径方向的单位长度  
            GLfloat c = (GLfloat) cos(b);  
            // 小圆上点,在xy 平面的分量长度  
            GLfloat r = minorRadius * c + majorRadius;  
            // 小圆上点在 z 轴上的长度  
            GLfloat z = minorRadius * (GLfloat) sin(b);  
              
            // 小圆上点坐标确认的过程:将该点分为在 z 轴 与 大圆半径方向,由于大圆半径只存在于 xy 平面,就相对容易求到 x , y 坐标。  
              
            // First point  
            objectVertex[0] = x0*r;// 小圆上点对应的 x 坐标  
            objectVertex[1] = y0*r;// 小圆上点对应的 y 坐标  
            objectVertex[2] = z; // 小圆上点对应的 z 坐标  
            m3dTransformVector3(transformedVertex, objectVertex, mTransform);  
            glVertex3fv(transformedVertex);  
  
            // Second point  
            objectVertex[0] = x1*r;  
            objectVertex[1] = y1*r;  
            objectVertex[2] = z;  
            m3dTransformVector3(transformedVertex, objectVertex, mTransform);  
            glVertex3fv(transformedVertex);  
            }  
        glEnd();  
        }  
}  

void DrawCylinder(M3DMatrix44f mTransform){  
    // 大圆半径  
    GLfloat majorRadius = 55.0f;  
    // 大圆圆周被切分的点数  
    GLint   numMajor = 100;  

    M3DVector3f objectVertex;         // Vertex in object/eye space  
    M3DVector3f transformedVertex;    // New Transformed vertex  
    // 每个点对应的弧度数  
    double majorStep = 2.0f*M3D_PI / numMajor;  
  
    glBegin(GL_TRIANGLE_STRIP);  
    // 对于大圆上的点进行迭代  
    for (int i=0; i<=numMajor; ++i)   
        {  

        // 第一个点对应的弧度  
        double a0 = i * majorStep;  
        // 第二个点对应的弧度  
        double a1 = a0 - majorStep;  
        // 第一个点在 x 与 y 轴上的单位长度  
        GLfloat x0 = (GLfloat) cos(a0);  
        GLfloat y0 = (GLfloat) sin(a0);  
        // 第二个点在 x 与 y 轴上的单位长度  
        GLfloat x1 = (GLfloat) cos(a1);  
        GLfloat y1 = (GLfloat) sin(a1);  
  
		// First point  
		objectVertex[0] = x0*majorRadius;// 小圆上点对应的 x 坐标  
		objectVertex[1] = y0*majorRadius;// 小圆上点对应的 y 坐标  
		objectVertex[2] = 50.0f; // 小圆上点对应的 z 坐标  
		m3dTransformVector3(transformedVertex, objectVertex, mTransform);  
		glVertex3fv(transformedVertex);  

		// Second point  
		objectVertex[0] = x1*majorRadius;  
		objectVertex[1] = y1*majorRadius;  
		objectVertex[2] = -50.0f;  
		m3dTransformVector3(transformedVertex, objectVertex, mTransform);  
		glVertex3fv(transformedVertex); 
		}
        glEnd();  
}  

//菜单回调函数
void processMenu(int value){
	switch(value){
		case 1:
			iCoordinateaxis=1;
			break;
		case 2:
			iCoordinateaxis=2;
			break;
		case 3:
			iProjectionMode=1;
			//强制调用窗口大小变化回调函数,更改投影模式为正交投影
			changSize(glutGet(GLUT_WINDOW_WIDTH),glutGet(GLUT_WINDOW_HEIGHT));
			break;
		case 4:
			iProjectionMode=2;
			//强制调用窗口大小变化回调函数,更改投影模式为透视投影
			changSize(glutGet(GLUT_WINDOW_WIDTH),glutGet(GLUT_WINDOW_HEIGHT));
			break;
		case 5:
			iPrintMatrix=1;
			break;
		case 6:
			iPrintMatrix=2;
			break;
		case 7:
			iCylinder=1;
			break;
		case 8:
			iCylinder=2;
			break;
		default:
			break;
	}
	//重新绘制
	glutPostRedisplay();
}

//显示回调函数
void renderScreen(void){
    M3DMatrix44f   transformationMatrix;   // Storeage for rotation matrix
    static GLfloat yRot = 0.0f;         // Rotation angle for animation
    yRot += 0.5f;
	//将窗口颜色清理为黑色
	glClearColor(0.0f, 0.0f, 0.0f, 0.0f);
    //把整个窗口清理为当前清理颜色:黑色;清除深度缓冲区。
    glClear(GL_COLOR_BUFFER_BIT|GL_DEPTH_BUFFER_BIT);

    //将当前Matrix状态入栈
    glPushMatrix();
	if(2==iProjectionMode)
		glTranslatef(0.0f, 0.0f, -250.0f);	//透视投影为便于观察整个坐标系往内移动250个单位
    //坐标系绕x轴旋转xRotAngle
    glRotatef(xRotAngle,1.0f,0.0f,0.0f);
    //坐标系绕y轴旋转yRotAngle
    glRotatef(yRotAngle,0.0f,1.0f,0.0f);
	//进行平滑处理 
	glEnable(GL_POINT_SMOOTH);
	glHint(GL_POINT_SMOOTH,GL_NICEST);
	glEnable(GL_LINE_SMOOTH);
	glHint(GL_LINE_SMOOTH,GL_NICEST);
	glEnable(GL_BLEND);
	glBlendFunc(GL_SRC_ALPHA,GL_ONE_MINUS_SRC_ALPHA);

	if(1==iCoordinateaxis){
		glColor3f(1.0f,1.0f,1.0f);
		glBegin(GL_LINES);
			glVertex3f(-90.0f,00.0f,0.0f);
			glVertex3f(90.0f,0.0f,0.0f);
			glVertex3f(0.0f,-90.0f,0.0f);
			glVertex3f(0.0f,90.0f,0.0f);
			glVertex3f(0.0f,0.0f,-90.0f);
			glVertex3f(0.0f,0.0f,90.0f);
		glEnd();

		glPushMatrix();
		glTranslatef(90.0f,0.0f,0.0f);
		glRotatef(90.0f,0.0f,1.0f,0.0f);
		glutSolidCone(3,6,10,10);
		glPopMatrix();

		glPushMatrix();
		glTranslatef(0.0f,90.0f,0.0f);
		glRotatef(-90.0f,1.0f,0.0f,0.0f);
		glutSolidCone(3,6,10,10);
		glPopMatrix();

		glPushMatrix();
		glTranslatef(0.0f,0.0f,90.0f);
		glRotatef(70.0f,0.0f,0.0f,1.0f);
		glutSolidCone(3,6,10,10);
		glPopMatrix();
	}
	glColor3f(0.5f,0.5f,1.0f);
	memset(transformationMatrix,0,sizeof(transformationMatrix));
	//打印变换矩阵
	if(2==iPrintMatrix){
		printf("--------------------------------------
");
		for(int i=0;i<4;i++){
			for(int j=0;j<4;j++){
				printf("%9.6f ",transformationMatrix[4*j+i]);
			}
			printf("
");
		}
	}
    m3dRotationMatrix44(transformationMatrix, m3dDegToRad(yRot), 0.0f, 1.0f, 0.0f);
	//transformationMatrix[12]、transformationMatrix[13] = 0.0f、transformationMatrix[14] = 0.0f是平移参数,分别代表x、y、 z轴的偏移参数。
	//transformationMatrix[15]代表缩放为原来的1/transformationMatrix[15]
	transformationMatrix[12] = 0.0f;
	transformationMatrix[13] = 0.0f;
	transformationMatrix[14] = 0.0f;     
	transformationMatrix[15] = 2.0f; 
	//打印变换矩阵
	if(2==iPrintMatrix){
		printf("--------------------------------------
");
		for(int i=0;i<4;i++){
			for(int j=0;j<4;j++){
				printf("%9.6f ",transformationMatrix[4*j+i]);
			}
			printf("
");
		}
	}
    
	if(1==iCylinder)
		DrawCylinder(transformationMatrix);
	else
		DrawTorus(transformationMatrix);

    //恢复压入栈的Matrix
    glPopMatrix();
    //交换两个缓冲区的指针
    glutSwapBuffers();
}

//设置Redering State 
void setupRederingState(void){
    glEnable(GL_DEPTH_TEST);	//使能深度测试
    glFrontFace(GL_CCW);		//多边形逆时针方向为正面
    //glEnable(GL_CULL_FACE);		//不显示背面
	glPolygonMode(GL_FRONT_AND_BACK,GL_LINE);//背面正面均使用线填充

    //设置清理颜色为黑色
    glClearColor(0.0f,0.0,0.0,1.0f);
    //设置绘画颜色为绿色
    glColor3f(1.0f,1.0f,0.0f);
	//使能深度测试
	glEnable(GL_DEPTH_TEST);
	//获取受支持的点大小范围
	glGetFloatv(GL_POINT_SIZE_RANGE,sizes);
	//获取受支持的点大小增量
	glGetFloatv(GL_POINT_SIZE_GRANULARITY,&step);
	//获取最大的投影矩阵堆栈深度
	glGetIntegerv( GL_MAX_PROJECTION_STACK_DEPTH,&iMaxProjectionStackDepth);
	//获取最大的模型视图矩阵堆栈深度
	glGetIntegerv( GL_MAX_MODELVIEW_STACK_DEPTH,&iMaxModeviewStackDepth);
	//获取最大的纹理矩阵堆栈深度
	glGetIntegerv( GL_MAX_TEXTURE_STACK_DEPTH,&iMaxTextureStackDepth);
	printf("point size range:%f-%f
",sizes[0],sizes[1]);
	printf("point step:%f
",step);
	printf("iMaxProjectionStackDepth=%d
",iMaxProjectionStackDepth);
	printf("iMaxModeviewStackDepth=%d
",iMaxModeviewStackDepth);
	printf("iMaxTextureStackDepth=%d
",iMaxTextureStackDepth);
}

//窗口大小变化回调函数
void changSize(GLint w,GLint h){
    //横宽比率
    GLfloat ratio;
    //设置坐标系为x(-100.0f,100.0f)、y(-100.0f,100.0f)、z(-100.0f,100.0f)
    GLfloat coordinatesize=100.0f;
    //窗口宽高为零直接返回
    if((w==0)||(h==0))
        return;
    //设置视口和窗口大小一致
    glViewport(0,0,w,h);
    //对投影矩阵应用随后的矩阵操作
    glMatrixMode(GL_PROJECTION);
    //重置当前指定的矩阵为单位矩阵 
    glLoadIdentity();
    ratio=(GLfloat)w/(GLfloat)h;
    //正交投影
	if(1==iProjectionMode){
		printf("glOrtho
");
		if(w<h)
			glOrtho(-coordinatesize,coordinatesize,-coordinatesize/ratio,coordinatesize/ratio,-coordinatesize,coordinatesize);
		else
			glOrtho(-coordinatesize*ratio,coordinatesize*ratio,-coordinatesize,coordinatesize,-coordinatesize,coordinatesize);
		//当前矩阵设置为模型视图矩阵
		glMatrixMode(GL_MODELVIEW);
		//重置当前指定的矩阵为单位矩阵 
		glLoadIdentity();
	}
	else{
		printf("gluPerspective
");
		gluPerspective(45,ratio,10.0f,500.0f);
		//当前矩阵设置为模型视图矩阵
		glMatrixMode(GL_MODELVIEW);
		//重置当前指定的矩阵为单位矩阵 
		glLoadIdentity();
	}

}

//按键输入处理回调函数
void specialKey(int key,int x,int y){

    if(key==GLUT_KEY_UP){
        xRotAngle-=5.0f;
    }
    else if(key==GLUT_KEY_DOWN){
        xRotAngle+=5.0f;
    }
    else if(key==GLUT_KEY_LEFT){
        yRotAngle-=5.0f;
    }
    else if(key==GLUT_KEY_RIGHT){
        yRotAngle+=5.0f;
    }
    //重新绘制
    glutPostRedisplay();
}

void timerFunc(int value)
{
	glutPostRedisplay();
	glutTimerFunc(10, timerFunc, 1);
}

int main(int argc, char* argv[])
{
	//菜单
	GLint iMainMenu;
	GLint iCoordinateaxisMenu;
	GLint iOrthoOrPerspectMenu;
	GLint iPrintmatrix;
	GLint iCylinderOrTorus;
	//初始化glut 
    glutInit(&argc,argv);
    //使用双缓冲区、深度缓冲区。
    glutInitDisplayMode(GLUT_DOUBLE|GLUT_RGBA|GLUT_DEPTH);
    //获取系统的宽像素
    SCREEN_WIDTH=glutGet(GLUT_SCREEN_WIDTH);
    //获取系统的高像素
    SCREEN_HEIGHT=glutGet(GLUT_SCREEN_HEIGHT);
	//创建窗口,窗口名字为OpenGL Transform Demo
    glutCreateWindow("OpenGL Transform Demo");
    //设置窗口大小
    glutReshapeWindow(windowWidth,windowHeight);
    //窗口居中显示
    glutPositionWindow((SCREEN_WIDTH-windowWidth)/2,(SCREEN_HEIGHT-windowHeight)/2);
    //窗口大小变化时的处理函数
    glutReshapeFunc(changSize);
    //设置显示回调函数 
    glutDisplayFunc(renderScreen);
    //设置按键输入处理回调函数
    glutSpecialFunc(specialKey);
	//菜单回调函数
	iCoordinateaxisMenu=glutCreateMenu(processMenu);
	//添加菜单
	glutAddMenuEntry("Display coordinate axis",1);
	glutAddMenuEntry("Don't dispaly coordinate axis",2);
	iOrthoOrPerspectMenu=glutCreateMenu(processMenu);
	glutAddMenuEntry("Ortho",3);
	glutAddMenuEntry("Perspect",4);
	iPrintmatrix=glutCreateMenu(processMenu);
	glutAddMenuEntry("Don't print Matrix",5);
	glutAddMenuEntry("Print Matrix",6);
	iCylinderOrTorus=glutCreateMenu(processMenu);
	glutAddMenuEntry("Cylinder",7);
	glutAddMenuEntry("Torus",8);
	iMainMenu=glutCreateMenu(processMenu);
	glutAddSubMenu("Whether Display coordinate axis",iCoordinateaxisMenu);
	glutAddSubMenu("Ortho Or Perspect",iOrthoOrPerspectMenu);
	glutAddSubMenu("Whether Print Matrix",iPrintmatrix);
	glutAddSubMenu("Cylinder or torus",iCylinderOrTorus);
	//将菜单榜定到鼠标右键上
	glutAttachMenu(GLUT_RIGHT_BUTTON);
	glutTimerFunc(10,timerFunc, 1);
    //设置全局渲染参数
    setupRederingState();
    glutMainLoop();
	
    return 0;
}


math3d.h

// Math3d.h
// Header file for the Math3d library. The C-Runtime has math.h, this file and the
// accompanying math.c are meant to suppliment math.h by adding geometry/math routines
// useful for graphics, simulation, and physics applications (3D stuff).
// Richard S. Wright Jr.
#ifndef _MATH3D_LIBRARY__
#define _MATH3D_LIBRARY__

#include <math.h>
#include <memory.h>

///////////////////////////////////////////////////////////////////////////////
// Data structures and containers
// Much thought went into how these are declared. Many libraries declare these
// as structures with x, y, z data members. However structure alignment issues
// could limit the portability of code based on such structures, or the binary
// compatibility of data files (more likely) that contain such structures across
// compilers/platforms. Arrays are always tightly packed, and are more efficient 
// for moving blocks of data around (usually).
typedef float	M3DVector3f[3];		// Vector of three floats (x, y, z)
typedef double	M3DVector3d[3];		// Vector of three doubles (x, y, z)

typedef float M3DVector4f[4];		// Lesser used... Do we really need these?
typedef double M3DVector4d[4];		// Yes, occasionaly

typedef float M3DVector2f[2];		// 3D points = 3D Vectors, but we need a 
typedef double M3DVector2d[2];		// 2D representations sometimes... (x,y) order



// 3x3 matrix - column major. X vector is 0, 1, 2, etc.
//		0	3	6	
//		1	4	7
//		2	5	8
typedef float M3DMatrix33f[9];		// A 3 x 3 matrix, column major (floats) - OpenGL Style
typedef double M3DMatrix33d[9];		// A 3 x 3 matrix, column major (doubles) - OpenGL Style


// 4x4 matrix - column major. X vector is 0, 1, 2, etc.
//	0	4	8	12
//	1	5	9	13
//	2	6	10	14
//	3	7	11	15
typedef float M3DMatrix44f[16];		// A 4 X 4 matrix, column major (floats) - OpenGL style
typedef double M3DMatrix44d[16];	// A 4 x 4 matrix, column major (doubles) - OpenGL style


///////////////////////////////////////////////////////////////////////////////
// Useful constants
#define M3D_PI (3.14159265358979323846)
#define M3D_2PI (2.0 * M3D_PI)
#define M3D_PI_DIV_180 (0.017453292519943296)
#define M3D_INV_PI_DIV_180 (57.2957795130823229)


///////////////////////////////////////////////////////////////////////////////
// Useful shortcuts and macros
// Radians are king... but we need a way to swap back and forth
#define m3dDegToRad(x)	((x)*M3D_PI_DIV_180)
#define m3dRadToDeg(x)	((x)*M3D_INV_PI_DIV_180)

// Hour angles
#define m3dHrToDeg(x)	((x) * (1.0 / 15.0))
#define m3dHrToRad(x)	m3dDegToRad(m3dHrToDeg(x))

#define m3dDegToHr(x)	((x) * 15.0))
#define m3dRadToHr(x)	m3dDegToHr(m3dRadToDeg(x))


// Returns the same number if it is a power of
// two. Returns a larger integer if it is not a 
// power of two. The larger integer is the next
// highest power of two.
inline unsigned int m3dIsPOW2(unsigned int iValue)
    {
    unsigned int nPow2 = 1;
    
    while(iValue > nPow2)
        nPow2 = (nPow2 << 1);
    
    return nPow2;
    }


///////////////////////////////////////////////////////////////////////////////
// Inline accessor functions for people who just can't count to 3 - Vectors
#define	m3dGetVectorX(v) (v[0])
#define m3dGetVectorY(v) (v[1])
#define m3dGetVectorZ(v) (v[2])
#define m3dGetVectorW(v) (v[3])

#define m3dSetVectorX(v, x)	((v)[0] = (x))
#define m3dSetVectorY(v, y)	((v)[1] = (y))
#define m3dSetVectorZ(v, z)	((v)[2] = (z))
#define m3dSetVectorW(v, w)	((v)[3] = (w))

///////////////////////////////////////////////////////////////////////////////
// Inline vector functions
// Load Vector with (x, y, z, w).
inline void m3dLoadVector2(M3DVector2f v, float x, float y)
    { v[0] = x; v[1] = y; }
inline void m3dLoadVector2(M3DVector2d v, float x, float y)
    { v[0] = x; v[1] = y; }
inline void m3dLoadVector3(M3DVector3f v, float x, float y, float z) 
	{ v[0] = x; v[1] = y; v[2] = z; }
inline void m3dLoadVector3(M3DVector3d v, double x, double y, double z)
	{ v[0] = x; v[1] = y; v[2] = z; }
inline void m3dLoadVector4(M3DVector4f v, float x, float y, float z, float w) 
	{ v[0] = x; v[1] = y; v[2] = z; v[3] = w;}
inline void m3dLoadVector4(M3DVector4d v, double x, double y, double z, double w)
	{ v[0] = x; v[1] = y; v[2] = z; v[3] = w;}


////////////////////////////////////////////////////////////////////////////////
// Copy vector src into vector dst
inline void	m3dCopyVector2(M3DVector2f dst, const M3DVector2f src) { memcpy(dst, src, sizeof(M3DVector2f)); }
inline void	m3dCopyVector2(M3DVector2d dst, const M3DVector2d src) { memcpy(dst, src, sizeof(M3DVector2d)); }

inline void	m3dCopyVector3(M3DVector3f dst, const M3DVector3f src) { memcpy(dst, src, sizeof(M3DVector3f)); }
inline void	m3dCopyVector3(M3DVector3d dst, const M3DVector3d src) { memcpy(dst, src, sizeof(M3DVector3d)); }

inline void	m3dCopyVector4(M3DVector4f dst, const M3DVector4f src) { memcpy(dst, src, sizeof(M3DVector4f)); }
inline void	m3dCopyVector4(M3DVector4d dst, const M3DVector4d src) { memcpy(dst, src, sizeof(M3DVector4d)); }


////////////////////////////////////////////////////////////////////////////////
// Add Vectors (r, a, b) r = a + b
inline void m3dAddVectors2(M3DVector2f r, const M3DVector2f a, const M3DVector2f b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1];  }
inline void m3dAddVectors2(M3DVector2d r, const M3DVector2d a, const M3DVector2d b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1];  }

inline void m3dAddVectors3(M3DVector3f r, const M3DVector3f a, const M3DVector3f b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1]; r[2] = a[2] + b[2]; }
inline void m3dAddVectors3(M3DVector3d r, const M3DVector3d a, const M3DVector3d b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1]; r[2] = a[2] + b[2]; }

inline void m3dAddVectors4(M3DVector4f r, const M3DVector4f a, const M3DVector4f b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1]; r[2] = a[2] + b[2]; r[3] = a[3] + b[3]; }
inline void m3dAddVectors4(M3DVector4d r, const M3DVector4d a, const M3DVector4d b)
	{ r[0] = a[0] + b[0];	r[1] = a[1] + b[1]; r[2] = a[2] + b[2]; r[3] = a[3] + b[3]; }

////////////////////////////////////////////////////////////////////////////////
// Subtract Vectors (r, a, b) r = a - b
inline void m3dSubtractVectors2(M3DVector2f r, const M3DVector2f a, const M3DVector2f b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1];  }
inline void m3dSubtractVectors2(M3DVector2d r, const M3DVector2d a, const M3DVector2d b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1]; }

inline void m3dSubtractVectors3(M3DVector3f r, const M3DVector3f a, const M3DVector3f b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1]; r[2] = a[2] - b[2]; }
inline void m3dSubtractVectors3(M3DVector3d r, const M3DVector3d a, const M3DVector3d b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1]; r[2] = a[2] - b[2]; }

inline void m3dSubtractVectors4(M3DVector4f r, const M3DVector4f a, const M3DVector4f b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1]; r[2] = a[2] - b[2]; r[3] = a[3] - b[3]; }
inline void m3dSubtractVectors4(M3DVector4d r, const M3DVector4d a, const M3DVector4d b)
	{ r[0] = a[0] - b[0]; r[1] = a[1] - b[1]; r[2] = a[2] - b[2]; r[3] = a[3] - b[3]; }



///////////////////////////////////////////////////////////////////////////////////////
// Scale Vectors (in place)
inline void m3dScaleVector2(M3DVector2f v, float scale) 
	{ v[0] *= scale; v[1] *= scale; }
inline void m3dScaleVector2(M3DVector2d v, double scale) 
	{ v[0] *= scale; v[1] *= scale; }

inline void m3dScaleVector3(M3DVector3f v, float scale) 
	{ v[0] *= scale; v[1] *= scale; v[2] *= scale; }
inline void m3dScaleVector3(M3DVector3d v, double scale) 
	{ v[0] *= scale; v[1] *= scale; v[2] *= scale; }

inline void m3dScaleVector4(M3DVector4f v, float scale) 
	{ v[0] *= scale; v[1] *= scale; v[2] *= scale; v[3] *= scale; }
inline void m3dScaleVector4(M3DVector4d v, double scale) 
	{ v[0] *= scale; v[1] *= scale; v[2] *= scale; v[3] *= scale; }


//////////////////////////////////////////////////////////////////////////////////////
// Cross Product
// u x v = result
// We only need one version for floats, and one version for doubles. A 3 component
// vector fits in a 4 component vector. If  M3DVector4d or M3DVector4f are passed
// we will be OK because 4th component is not used.
inline void m3dCrossProduct(M3DVector3f result, const M3DVector3f u, const M3DVector3f v)
	{
	result[0] = u[1]*v[2] - v[1]*u[2];
	result[1] = -u[0]*v[2] + v[0]*u[2];
	result[2] = u[0]*v[1] - v[0]*u[1];
	}

inline void m3dCrossProduct(M3DVector3d result, const M3DVector3d u, const M3DVector3d v)
	{
	result[0] = u[1]*v[2] - v[1]*u[2];
	result[1] = -u[0]*v[2] + v[0]*u[2];
	result[2] = u[0]*v[1] - v[0]*u[1];
	}

//////////////////////////////////////////////////////////////////////////////////////
// Dot Product, only for three component vectors
// return u dot v
inline float m3dDotProduct(const M3DVector3f u, const M3DVector3f v)
	{ return u[0]*v[0] + u[1]*v[1] + u[2]*v[2]; }

inline double m3dDotProduct(const M3DVector3d u, const M3DVector3d v)
	{ return u[0]*v[0] + u[1]*v[1] + u[2]*v[2]; }

//////////////////////////////////////////////////////////////////////////////////////
// Angle between vectors, only for three component vectors. Angle is in radians...
inline float m3dGetAngleBetweenVectors(const M3DVector3f u, const M3DVector3f v)
    {
    float dTemp = m3dDotProduct(u, v);
    return float(acos(double(dTemp)));
    }

inline double m3dGetAngleBetweenVectors(const M3DVector3d u, const M3DVector3d v)
    {
    double dTemp = m3dDotProduct(u, v);
    return acos(dTemp);
    }

//////////////////////////////////////////////////////////////////////////////////////
// Get Square of a vectors length
// Only for three component vectors
inline float m3dGetVectorLengthSquared(const M3DVector3f u)
	{ return (u[0] * u[0]) + (u[1] * u[1]) + (u[2] * u[2]); }

inline double m3dGetVectorLengthSquared(const M3DVector3d u)
	{ return (u[0] * u[0]) + (u[1] * u[1]) + (u[2] * u[2]); }

//////////////////////////////////////////////////////////////////////////////////////
// Get lenght of vector
// Only for three component vectors.
inline float m3dGetVectorLength(const M3DVector3f u)
	{ return float(sqrt(double(m3dGetVectorLengthSquared(u)))); }

inline double m3dGetVectorLength(const M3DVector3d u)
	{ return sqrt(m3dGetVectorLengthSquared(u)); }

//////////////////////////////////////////////////////////////////////////////////////
// Normalize a vector
// Scale a vector to unit length. Easy, just scale the vector by it's length
inline void m3dNormalizeVector(M3DVector3f u)
	{ m3dScaleVector3(u, 1.0f / m3dGetVectorLength(u)); }

inline void m3dNormalizeVector(M3DVector3d u)
	{ m3dScaleVector3(u, 1.0 / m3dGetVectorLength(u)); }


//////////////////////////////////////////////////////////////////////////////////////
// Get the distance between two points. The distance between two points is just
// the magnitude of the difference between two vectors
// Located in math.cpp
float m3dGetDistanceSquared(const M3DVector3f u, const M3DVector3f v);
double m3dGetDistanceSquared(const M3DVector3d u, const M3DVector3d v);

inline double m3dGetDistance(const M3DVector3d u, const M3DVector3d v)
{ return sqrt(m3dGetDistanceSquared(u, v)); }

inline float m3dGetDistance(const M3DVector3f u, const M3DVector3f v)
{ return float(sqrt(m3dGetDistanceSquared(u, v))); }

inline float m3dGetMagnitudeSquared(const M3DVector3f u) { return u[0]*u[0] + u[1]*u[1] + u[2]*u[2]; }
inline double m3dGetMagnitudeSquared(const M3DVector3d u) { return u[0]*u[0] + u[1]*u[1] + u[2]*u[2]; }

inline float m3dGetMagnitude(const M3DVector3f u) { return float(sqrt(m3dGetMagnitudeSquared(u))); }
inline double m3dGetMagnitude(const M3DVector3d u) { return sqrt(m3dGetMagnitudeSquared(u)); }

	

//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// Matrix functions
// Both floating point and double precision 3x3 and 4x4 matricies are supported.
// No support is included for arbitrarily dimensioned matricies on purpose, since
// the 3x3 and 4x4 matrix routines are the most common for the purposes of this
// library. Matrices are column major, like OpenGL matrices.
// Unlike the vector functions, some of these are going to have to not be inlined,
// although many will be.


// Copy Matrix
// Brain-dead memcpy
inline void m3dCopyMatrix33(M3DMatrix33f dst, const M3DMatrix33f src)
	{ memcpy(dst, src, sizeof(M3DMatrix33f)); }

inline void m3dCopyMatrix33(M3DMatrix33d dst, const M3DMatrix33d src)
	{ memcpy(dst, src, sizeof(M3DMatrix33d)); }

inline void m3dCopyMatrix44(M3DMatrix44f dst, const M3DMatrix44f src)
	{ memcpy(dst, src, sizeof(M3DMatrix44f)); }

inline void m3dCopyMatrix44(M3DMatrix44d dst, const M3DMatrix44d src)
	{ memcpy(dst, src, sizeof(M3DMatrix44d)); }

// LoadIdentity
// Implemented in Math3d.cpp
void m3dLoadIdentity33(M3DMatrix33f m);
void m3dLoadIdentity33(M3DMatrix33d m);
void m3dLoadIdentity44(M3DMatrix44f m);
void m3dLoadIdentity44(M3DMatrix44d m);

/////////////////////////////////////////////////////////////////////////////
// Get/Set Column.
inline void m3dGetMatrixColumn33(M3DVector3f dst, const M3DMatrix33f src, int column)
	{ memcpy(dst, src + (3 * column), sizeof(float) * 3); }

inline void m3dGetMatrixColumn33(M3DVector3d dst, const M3DMatrix33d src, int column)
	{ memcpy(dst, src + (3 * column), sizeof(double) * 3); }

inline void m3dSetMatrixColumn33(M3DMatrix33f dst, const M3DVector3f src, int column)
	{ memcpy(dst + (3 * column), src, sizeof(float) * 3); }

inline void m3dSetMatrixColumn33(M3DMatrix33d dst, const M3DVector3d src, int column)
	{ memcpy(dst + (3 * column), src, sizeof(double) * 3); }

inline void m3dGetMatrixColumn44(M3DVector4f dst, const M3DMatrix44f src, int column)
	{ memcpy(dst, src + (4 * column), sizeof(float) * 4); }

inline void m3dGetMatrixColumn44(M3DVector4d dst, const M3DMatrix44d src, int column)
	{ memcpy(dst, src + (4 * column), sizeof(double) * 4); }

inline void m3dSetMatrixColumn44(M3DMatrix44f dst, const M3DVector4f src, int column)
	{ memcpy(dst + (4 * column), src, sizeof(float) * 4); }

inline void m3dSetMatrixColumn44(M3DMatrix44d dst, const M3DVector4d src, int column)
	{ memcpy(dst + (4 * column), src, sizeof(double) * 4); }


// Get/Set row purposely omitted... use the functions below.
// I don't think row vectors are useful for column major ordering...
// If I'm wrong, add them later.


//////////////////////////////////////////////////////////////////////////////
// Get/Set RowCol - Remember column major ordering...
// Provides for element addressing
inline void m3dSetMatrixRowCol33(M3DMatrix33f m, int row, int col, float value)
	{ m[(col * 3) + row] = value; }

inline float m3dGetMatrixRowCol33(const M3DMatrix33f m, int row, int col)
	{ return m[(col * 3) + row]; }

inline void m3dSetMatrixRowCol33(M3DMatrix33d m, int row, int col, double value)
	{ m[(col * 3) + row] = value; }

inline double m3dGetMatrixRowCol33(const M3DMatrix33d m, int row, int col)
	{ return m[(col * 3) + row]; }

inline void m3dSetMatrixRowCol44(M3DMatrix44f m, int row, int col, float value)
	{ m[(col * 4) + row] = value; }

inline float m3dGetMatrixRowCol44(const M3DMatrix44f m, int row, int col)
	{ return m[(col * 4) + row]; }

inline void m3dSetMatrixRowCol44(M3DMatrix44d m, int row, int col, double value)
	{ m[(col * 4) + row] = value; }

inline double m3dGetMatrixRowCol44(const M3DMatrix44d m, int row, int col)
	{ return m[(col * 4) + row]; }


///////////////////////////////////////////////////////////////////////////////
// Extract a rotation matrix from a 4x4 matrix
// Extracts the rotation matrix (3x3) from a 4x4 matrix
inline void m3dExtractRotation(M3DMatrix33f dst, const M3DMatrix44f src)
	{	
	memcpy(dst, src, sizeof(float) * 3); // X column
	memcpy(dst + 3, src + 4, sizeof(float) * 3); // Y column
	memcpy(dst + 6, src + 8, sizeof(float) * 3); // Z column
	}

// Ditto above, but for doubles
inline void m3dExtractRotation(M3DMatrix33d dst, const M3DMatrix44d src)
	{
	memcpy(dst, src, sizeof(double) * 3); // X column
	memcpy(dst + 3, src + 4, sizeof(double) * 3); // Y column
	memcpy(dst + 6, src + 8, sizeof(double) * 3); // Z column
	}

// Inject Rotation (3x3) into a full 4x4 matrix...
inline void m3dInjectRotation(M3DMatrix44f dst, const M3DMatrix33f src)
	{
	memcpy(dst, src, sizeof(float) * 4);
	memcpy(dst + 4, src + 4, sizeof(float) * 4);
	memcpy(dst + 8, src + 8, sizeof(float) * 4);
	}

// Ditto above for doubles
inline void m3dInjectRotation(M3DMatrix44d dst, const M3DMatrix33d src)
	{
	memcpy(dst, src, sizeof(double) * 4);
	memcpy(dst + 4, src + 4, sizeof(double) * 4);
	memcpy(dst + 8, src + 8, sizeof(double) * 4);
	}


////////////////////////////////////////////////////////////////////////////////
// MultMatrix
// Implemented in Math.cpp
void m3dMatrixMultiply44(M3DMatrix44f product, const M3DMatrix44f a, const M3DMatrix44f b);
void m3dMatrixMultiply44(M3DMatrix44d product, const M3DMatrix44d a, const M3DMatrix44d b);
void m3dMatrixMultiply33(M3DMatrix33f product, const M3DMatrix33f a, const M3DMatrix33f b);
void m3dMatrixMultiply33(M3DMatrix33d product, const M3DMatrix33d a, const M3DMatrix33d b);


// Transform - Does rotation and translation via a 4x4 matrix. Transforms
// a point or vector.
// By-the-way __inline means I'm asking the compiler to do a cost/benefit analysis. If 
// these are used frequently, they may not be inlined to save memory. I'm experimenting
// with this....
__inline void m3dTransformVector3(M3DVector3f vOut, const M3DVector3f v, const M3DMatrix44f m)
    {
    vOut[0] = m[0] * v[0] + m[4] * v[1] + m[8] *  v[2] + m[12];// * v[3];	 
    vOut[1] = m[1] * v[0] + m[5] * v[1] + m[9] *  v[2] + m[13];// * v[3];	
    vOut[2] = m[2] * v[0] + m[6] * v[1] + m[10] * v[2] + m[14];// * v[3];	
	//vOut[3] = m[3] * v[0] + m[7] * v[1] + m[11] * v[2] + m[15] * v[3];
    }

// Ditto above, but for doubles
__inline void m3dTransformVector3(M3DVector3d vOut, const M3DVector3d v, const M3DMatrix44d m)
    {
    vOut[0] = m[0] * v[0] + m[4] * v[1] + m[8] *  v[2] + m[12];// * v[3];	 
    vOut[1] = m[1] * v[0] + m[5] * v[1] + m[9] *  v[2] + m[13];// * v[3];	
    vOut[2] = m[2] * v[0] + m[6] * v[1] + m[10] * v[2] + m[14];// * v[3];	
	//vOut[3] = m[3] * v[0] + m[7] * v[1] + m[11] * v[2] + m[15] * v[3];
    }

__inline void m3dTransformVector4(M3DVector4f vOut, const M3DVector4f v, const M3DMatrix44f m)
    {
    vOut[0] = m[0] * v[0] + m[4] * v[1] + m[8] *  v[2] + m[12] * v[3];	 
    vOut[1] = m[1] * v[0] + m[5] * v[1] + m[9] *  v[2] + m[13] * v[3];	
    vOut[2] = m[2] * v[0] + m[6] * v[1] + m[10] * v[2] + m[14] * v[3];	
	vOut[3] = m[3] * v[0] + m[7] * v[1] + m[11] * v[2] + m[15] * v[3];
    }

// Ditto above, but for doubles
__inline void m3dTransformVector4(M3DVector4d vOut, const M3DVector4d v, const M3DMatrix44d m)
    {
    vOut[0] = m[0] * v[0] + m[4] * v[1] + m[8] *  v[2] + m[12] * v[3];	 
    vOut[1] = m[1] * v[0] + m[5] * v[1] + m[9] *  v[2] + m[13] * v[3];	
    vOut[2] = m[2] * v[0] + m[6] * v[1] + m[10] * v[2] + m[14] * v[3];	
	vOut[3] = m[3] * v[0] + m[7] * v[1] + m[11] * v[2] + m[15] * v[3];
    }



// Just do the rotation, not the translation... this is usually done with a 3x3
// Matrix.
__inline void m3dRotateVector(M3DVector3f vOut, const M3DVector3f p, const M3DMatrix33f m)
	{
    vOut[0] = m[0] * p[0] + m[3] * p[1] + m[6] * p[2];	
    vOut[1] = m[1] * p[0] + m[4] * p[1] + m[7] * p[2];	
    vOut[2] = m[2] * p[0] + m[5] * p[1] + m[8] * p[2];	
	}

// Ditto above, but for doubles
__inline void m3dRotateVector(M3DVector3d vOut, const M3DVector3d p, const M3DMatrix33d m)
	{
    vOut[0] = m[0] * p[0] + m[3] * p[1] + m[6] * p[2];	
    vOut[1] = m[1] * p[0] + m[4] * p[1] + m[7] * p[2];	
    vOut[2] = m[2] * p[0] + m[5] * p[1] + m[8] * p[2];	
	}


// Scale a matrix (I don't beleive in Scaling matricies ;-)
// Yes, it's faster to loop backwards... These could be 
// unrolled... but eh... if you find this is a bottleneck,
// then you should unroll it yourself
inline void m3dScaleMatrix33(M3DMatrix33f m, float scale)
{ for(int i = 8; i >=0; i--) m[i] *= scale; }

inline void m3dScaleMatrix33(M3DMatrix33d m, double scale)
{ for(int i = 8; i >=0; i--) m[i] *= scale; }

inline void m3dScaleMatrix44(M3DMatrix44f m, float scale)
{ for(int i = 15; i >=0; i--) m[i] *= scale; }

inline void m3dScaleMatrix44(M3DMatrix44d m, double scale)
{ for(int i = 15; i >=0; i--) m[i] *= scale; }


// Create a Rotation matrix
// Implemented in math.cpp
void m3dRotationMatrix33(M3DMatrix33f m, float angle, float x, float y, float z);
void m3dRotationMatrix33(M3DMatrix33d m, double angle, double x, double y, double z);
void m3dRotationMatrix44(M3DMatrix44f m, float angle, float x, float y, float z);
void m3dRotationMatrix44(M3DMatrix44d m, double angle, double x, double y, double z);

// Create a Translation matrix. Only 4x4 matrices have translation components
inline void m3dTranslationMatrix44(M3DMatrix44f m, float x, float y, float z)
{ m3dLoadIdentity44(m); m[12] = x; m[13] = y; m[14] = z; }

inline void m3dTranslationMatrix44(M3DMatrix44d m, double x, double y, double z)
{ m3dLoadIdentity44(m); m[12] = x; m[13] = y; m[14] = z; }


// Translate matrix. Only 4x4 matrices supported
inline void m3dTranslateMatrix44(M3DMatrix44f m, float x, float y, float z)
{ m[12] += x; m[13] += y; m[14] += z; }

inline void m3dTranslateMatrix44(M3DMatrix44d m, double x, double y, double z)
{ m[12] += x; m[13] += y; m[14] += z; }


// Scale matrix. Only 4x4 matrices supported
inline void m3dScaleMatrix44(M3DMatrix44f m, float x, float y, float z)
{ m[0] *= x; m[5] *= y; m[10] *= z; }

inline void m3dScaleMatrix44(M3DMatrix44d m, double x, double y, double z)
{ m[0] *= x; m[5] *= y; m[10] *= z; }


// Transpose/Invert - Only 4x4 matricies supported
#define TRANSPOSE44(dst, src)            
{                                        
    for (int j = 0; j < 4; j++)          
    {                                    
        for (int i = 0; i < 4; i++)      
        {                                
            dst[(j*4)+i] = src[(i*4)+j]; 
        }                                
    }                                    
}
inline void m3dTransposeMatrix44(M3DMatrix44f dst, const M3DMatrix44f src)
{ TRANSPOSE44(dst, src); }
inline void m3dTransposeMatrix44(M3DMatrix44d dst, const M3DMatrix44d src)
{ TRANSPOSE44(dst, src); }
bool m3dInvertMatrix44(M3DMatrix44f dst, const M3DMatrix44f src);
bool m3dInvertMatrix44(M3DMatrix44d dst, const M3DMatrix44d src);

///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
// Other Miscellaneous functions

// Find a normal from three points
// Implemented in math3d.cpp
void m3dFindNormal(M3DVector3f result, const M3DVector3f point1, const M3DVector3f point2, 
							const M3DVector3f point3);
void m3dFindNormal(M3DVector3d result, const M3DVector3d point1, const M3DVector3d point2, 
							const M3DVector3d point3);



// Calculates the signed distance of a point to a plane
inline float m3dGetDistanceToPlane(const M3DVector3f point, const M3DVector4f plane)
           { return point[0]*plane[0] + point[1]*plane[1] + point[2]*plane[2] + plane[3]; }

inline double m3dGetDistanceToPlane(const M3DVector3d point, const M3DVector4d plane)
           { return point[0]*plane[0] + point[1]*plane[1] + point[2]*plane[2] + plane[3]; }


// Get plane equation from three points and a normal
void m3dGetPlaneEquation(M3DVector4f planeEq, const M3DVector3f p1, const M3DVector3f p2, const M3DVector3f p3);
void m3dGetPlaneEquation(M3DVector4d planeEq, const M3DVector3d p1, const M3DVector3d p2, const M3DVector3d p3);

// Determine if a ray intersects a sphere
double m3dRaySphereTest(const M3DVector3d point, const M3DVector3d ray, const M3DVector3d sphereCenter, double sphereRadius);
float m3dRaySphereTest(const M3DVector3f point, const M3DVector3f ray, const M3DVector3f sphereCenter, float sphereRadius);

// Etc. etc.

///////////////////////////////////////////////////////////////////////////////////////////////////////
// Faster (and more robust) replacements for gluProject
void m3dProjectXY( M3DVector2f vPointOut, const M3DMatrix44f mModelView, const M3DMatrix44f mProjection, const int iViewPort[4], const M3DVector3f vPointIn);    
void m3dProjectXYZ(M3DVector3f vPointOut, const M3DMatrix44f mModelView, const M3DMatrix44f mProjection, const int iViewPort[4], const M3DVector3f vPointIn);



//////////////////////////////////////////////////////////////////////////////////////////////////
// This function does a three dimensional Catmull-Rom "spline" interpolation between p1 and p2
void m3dCatmullRom(M3DVector3f vOut, M3DVector3f vP0, M3DVector3f vP1, M3DVector3f vP2, M3DVector3f vP3, float t);
void m3dCatmullRom(M3DVector3d vOut, M3DVector3d vP0, M3DVector3d vP1, M3DVector3d vP2, M3DVector3d vP3, double t);

//////////////////////////////////////////////////////////////////////////////////////////////////
// Compare floats and doubles... 
inline bool m3dCloseEnough(float fCandidate, float fCompare, float fEpsilon)
    {
    return (fabs(fCandidate - fCompare) < fEpsilon);
    }
    
inline bool m3dCloseEnough(double dCandidate, double dCompare, double dEpsilon)
    {
    return (fabs(dCandidate - dCompare) < dEpsilon);
    }    
 
////////////////////////////////////////////////////////////////////////////
// Used for normal mapping. Finds the tangent bases for a triangle...
// Only a floating point implementation is provided.
void m3dCalculateTangentBasis(const M3DVector3f pvTriangle[3], const M3DVector2f pvTexCoords[3], const M3DVector3f N, M3DVector3f vTangent);

////////////////////////////////////////////////////////////////////////////
// Smoothly step between 0 and 1 between edge1 and edge 2
double m3dSmoothStep(double edge1, double edge2, double x);
float m3dSmoothStep(float edge1, float edge2, float x);

/////////////////////////////////////////////////////////////////////////////
// Planar shadow Matrix
void m3dMakePlanarShadowMatrix(M3DMatrix44d proj, const M3DVector4d planeEq, const M3DVector3d vLightPos);
void m3dMakePlanarShadowMatrix(M3DMatrix44f proj, const M3DVector4f planeEq, const M3DVector3f vLightPos);

double m3dClosestPointOnRay(M3DVector3d vPointOnRay, const M3DVector3d vRayOrigin, const M3DVector3d vUnitRayDir, 
							const M3DVector3d vPointInSpace);

float m3dClosestPointOnRay(M3DVector3f vPointOnRay, const M3DVector3f vRayOrigin, const M3DVector3f vUnitRayDir, 
							const M3DVector3f vPointInSpace);

#endif


Math3d.cpp

// Math3d.c
// Implementation of non-inlined functions in the Math3D Library
// Richard S. Wright Jr.

// These are pretty portable
#include "stdafx.h"
#include <math.h>
#include "math3d.h"


////////////////////////////////////////////////////////////
// LoadIdentity
// For 3x3 and 4x4 float and double matricies.
// 3x3 float
void m3dLoadIdentity33(M3DMatrix33f m)
	{
	// Don't be fooled, this is still column major
	static M3DMatrix33f	identity = { 1.0f, 0.0f, 0.0f ,
									 0.0f, 1.0f, 0.0f,
									 0.0f, 0.0f, 1.0f };

	memcpy(m, identity, sizeof(M3DMatrix33f));
	}

// 3x3 double
void m3dLoadIdentity33(M3DMatrix33d m)
	{
	// Don't be fooled, this is still column major
	static M3DMatrix33d	identity = { 1.0, 0.0, 0.0 ,
									 0.0, 1.0, 0.0,
									 0.0, 0.0, 1.0 };

	memcpy(m, identity, sizeof(M3DMatrix33d));
	}

// 4x4 float
void m3dLoadIdentity44(M3DMatrix44f m)
	{
	// Don't be fooled, this is still column major
	static M3DMatrix44f	identity = { 1.0f, 0.0f, 0.0f, 0.0f,
									 0.0f, 1.0f, 0.0f, 0.0f,
									 0.0f, 0.0f, 1.0f, 0.0f,
									 0.0f, 0.0f, 0.0f, 1.0f };

	memcpy(m, identity, sizeof(M3DMatrix44f));
	}

// 4x4 double
void m3dLoadIdentity44(M3DMatrix44d m)
	{
	static M3DMatrix44d	identity = { 1.0, 0.0, 0.0, 0.0,
									 0.0, 1.0, 0.0, 0.0,
									 0.0, 0.0, 1.0, 0.0,
									 0.0, 0.0, 0.0, 1.0 };

	memcpy(m, identity, sizeof(M3DMatrix44d));
	}


////////////////////////////////////////////////////////////////////////
// Return the square of the distance between two points
// Should these be inlined...?
float m3dGetDistanceSquared(const M3DVector3f u, const M3DVector3f v)
	{
	float x = u[0] - v[0];
	x = x*x;
	
	float y = u[1] - v[1];
	y = y*y;

	float z = u[2] - v[2];
	z = z*z;

	return (x + y + z);
    }

// Ditto above, but for doubles
double m3dGetDistanceSquared(const M3DVector3d u, const M3DVector3d v)
	{
	double x = u[0] - v[0];
	x = x*x;
	
	double y = u[1] - v[1];
	y = y*y;

	double z = u[2] - v[2];
	z = z*z;

	return (x + y + z);
	}

#define A(row,col)  a[(col<<2)+row]
#define B(row,col)  b[(col<<2)+row]
#define P(row,col)  product[(col<<2)+row]

///////////////////////////////////////////////////////////////////////////////
// Multiply two 4x4 matricies
void m3dMatrixMultiply44(M3DMatrix44f product, const M3DMatrix44f a, const M3DMatrix44f b )
{
	for (int i = 0; i < 4; i++) {
		float ai0=A(i,0),  ai1=A(i,1),  ai2=A(i,2),  ai3=A(i,3);
		P(i,0) = ai0 * B(0,0) + ai1 * B(1,0) + ai2 * B(2,0) + ai3 * B(3,0);
		P(i,1) = ai0 * B(0,1) + ai1 * B(1,1) + ai2 * B(2,1) + ai3 * B(3,1);
		P(i,2) = ai0 * B(0,2) + ai1 * B(1,2) + ai2 * B(2,2) + ai3 * B(3,2);
		P(i,3) = ai0 * B(0,3) + ai1 * B(1,3) + ai2 * B(2,3) + ai3 * B(3,3);
	}
}

// Ditto above, but for doubles
void m3dMatrixMultiply(M3DMatrix44d product, const M3DMatrix44d a, const M3DMatrix44d b )
{
	for (int i = 0; i < 4; i++) {
		double ai0=A(i,0),  ai1=A(i,1),  ai2=A(i,2),  ai3=A(i,3);
		P(i,0) = ai0 * B(0,0) + ai1 * B(1,0) + ai2 * B(2,0) + ai3 * B(3,0);
		P(i,1) = ai0 * B(0,1) + ai1 * B(1,1) + ai2 * B(2,1) + ai3 * B(3,1);
		P(i,2) = ai0 * B(0,2) + ai1 * B(1,2) + ai2 * B(2,2) + ai3 * B(3,2);
		P(i,3) = ai0 * B(0,3) + ai1 * B(1,3) + ai2 * B(2,3) + ai3 * B(3,3);
	}
}
#undef A
#undef B
#undef P


#define A33(row,col)  a[(col*3)+row]
#define B33(row,col)  b[(col*3)+row]
#define P33(row,col)  product[(col*3)+row]

///////////////////////////////////////////////////////////////////////////////
// Multiply two 3x3 matricies
void m3dMatrixMultiply33(M3DMatrix33f product, const M3DMatrix33f a, const M3DMatrix33f b )
{
	for (int i = 0; i < 3; i++) {
		float ai0=A33(i,0), ai1=A33(i,1),  ai2=A33(i,2);
		P33(i,0) = ai0 * B33(0,0) + ai1 * B33(1,0) + ai2 * B33(2,0);
		P33(i,1) = ai0 * B33(0,1) + ai1 * B33(1,1) + ai2 * B33(2,1);
		P33(i,2) = ai0 * B33(0,2) + ai1 * B33(1,2) + ai2 * B33(2,2);
	}
}

// Ditto above, but for doubles
void m3dMatrixMultiply44(M3DMatrix33d product, const M3DMatrix33d a, const M3DMatrix33d b )
{
	for (int i = 0; i < 3; i++) {
		double ai0=A33(i,0),  ai1=A33(i,1),  ai2=A33(i,2);
		P33(i,0) = ai0 * B33(0,0) + ai1 * B33(1,0) + ai2 * B33(2,0);
		P33(i,1) = ai0 * B33(0,1) + ai1 * B33(1,1) + ai2 * B33(2,1);
		P33(i,2) = ai0 * B33(0,2) + ai1 * B33(1,2) + ai2 * B33(2,2);
	}
}

#undef A33
#undef B33
#undef P33

#define M33(row,col)  m[col*3+row]

///////////////////////////////////////////////////////////////////////////////
// Creates a 3x3 rotation matrix, takes radians NOT degrees
void m3dRotationMatrix33(M3DMatrix33f m, float angle, float x, float y, float z)
	{
	
	float mag, s, c;
	float xx, yy, zz, xy, yz, zx, xs, ys, zs, one_c;

	s = float(sin(angle));
	c = float(cos(angle));

	mag = float(sqrt( x*x + y*y + z*z ));

	// Identity matrix
	if (mag == 0.0f) {
		m3dLoadIdentity33(m);
		return;
	}

	// Rotation matrix is normalized
	x /= mag;
	y /= mag;
	z /= mag;



	xx = x * x;
	yy = y * y;
	zz = z * z;
	xy = x * y;
	yz = y * z;
	zx = z * x;
	xs = x * s;
	ys = y * s;
	zs = z * s;
	one_c = 1.0f - c;

	M33(0,0) = (one_c * xx) + c;
	M33(0,1) = (one_c * xy) - zs;
	M33(0,2) = (one_c * zx) + ys;

	M33(1,0) = (one_c * xy) + zs;
	M33(1,1) = (one_c * yy) + c;
	M33(1,2) = (one_c * yz) - xs;

	M33(2,0) = (one_c * zx) - ys;
	M33(2,1) = (one_c * yz) + xs;
	M33(2,2) = (one_c * zz) + c;
	}

#undef M33

///////////////////////////////////////////////////////////////////////////////
// Creates a 4x4 rotation matrix, takes radians NOT degrees
void m3dRotationMatrix44(M3DMatrix44f m, float angle, float x, float y, float z)
	{
	float mag, s, c;
	float xx, yy, zz, xy, yz, zx, xs, ys, zs, one_c;

	s = float(sin(angle));
	c = float(cos(angle));

	mag = float(sqrt( x*x + y*y + z*z ));

	// Identity matrix
	if (mag == 0.0f) {
		m3dLoadIdentity44(m);
		return;
	}

	// Rotation matrix is normalized
	x /= mag;
	y /= mag;
	z /= mag;

    #define M(row,col)  m[col*4+row]

	xx = x * x;
	yy = y * y;
	zz = z * z;
	xy = x * y;
	yz = y * z;
	zx = z * x;
	xs = x * s;
	ys = y * s;
	zs = z * s;
	one_c = 1.0f - c;

	M(0,0) = (one_c * xx) + c;
	M(0,1) = (one_c * xy) - zs;
	M(0,2) = (one_c * zx) + ys;
	M(0,3) = 0.0f;

	M(1,0) = (one_c * xy) + zs;
	M(1,1) = (one_c * yy) + c;
	M(1,2) = (one_c * yz) - xs;
	M(1,3) = 0.0f;

	M(2,0) = (one_c * zx) - ys;
	M(2,1) = (one_c * yz) + xs;
	M(2,2) = (one_c * zz) + c;
	M(2,3) = 0.0f;

	M(3,0) = 0.0f;
	M(3,1) = 0.0f;
	M(3,2) = 0.0f;
	M(3,3) = 1.0f;

    #undef M
	}



///////////////////////////////////////////////////////////////////////////////
// Ditto above, but for doubles
void m3dRotationMatrix33(M3DMatrix33d m, double angle, double x, double y, double z)
	{
	double mag, s, c;
	double xx, yy, zz, xy, yz, zx, xs, ys, zs, one_c;

	s = sin(angle);
	c = cos(angle);

	mag = sqrt( x*x + y*y + z*z );

	// Identity matrix
	if (mag == 0.0) {
		m3dLoadIdentity33(m);
		return;
	}

	// Rotation matrix is normalized
	x /= mag;
	y /= mag;
	z /= mag;

    #define M(row,col)  m[col*3+row]

	xx = x * x;
	yy = y * y;
	zz = z * z;
	xy = x * y;
	yz = y * z;
	zx = z * x;
	xs = x * s;
	ys = y * s;
	zs = z * s;
	one_c = 1.0 - c;

	M(0,0) = (one_c * xx) + c;
	M(0,1) = (one_c * xy) - zs;
	M(0,2) = (one_c * zx) + ys;

	M(1,0) = (one_c * xy) + zs;
	M(1,1) = (one_c * yy) + c;
	M(1,2) = (one_c * yz) - xs;

	M(2,0) = (one_c * zx) - ys;
	M(2,1) = (one_c * yz) + xs;
	M(2,2) = (one_c * zz) + c;

    #undef M
	}


///////////////////////////////////////////////////////////////////////////////
// Creates a 4x4 rotation matrix, takes radians NOT degrees
void m3dRotationMatrix44(M3DMatrix44d m, double angle, double x, double y, double z)
	{
	double mag, s, c;
	double xx, yy, zz, xy, yz, zx, xs, ys, zs, one_c;

	s = sin(angle);
	c = cos(angle);

	mag = sqrt( x*x + y*y + z*z );

	// Identity matrix
	if (mag == 0.0) {
		m3dLoadIdentity44(m);
		return;
	}

	// Rotation matrix is normalized
	x /= mag;
	y /= mag;
	z /= mag;

    #define M(row,col)  m[col*4+row]

	xx = x * x;
	yy = y * y;
	zz = z * z;
	xy = x * y;
	yz = y * z;
	zx = z * x;
	xs = x * s;
	ys = y * s;
	zs = z * s;
	one_c = 1.0f - c;

	M(0,0) = (one_c * xx) + c;
	M(0,1) = (one_c * xy) - zs;
	M(0,2) = (one_c * zx) + ys;
	M(0,3) = 0.0;

	M(1,0) = (one_c * xy) + zs;
	M(1,1) = (one_c * yy) + c;
	M(1,2) = (one_c * yz) - xs;
	M(1,3) = 0.0;

	M(2,0) = (one_c * zx) - ys;
	M(2,1) = (one_c * yz) + xs;
	M(2,2) = (one_c * zz) + c;
	M(2,3) = 0.0;

	M(3,0) = 0.0;
	M(3,1) = 0.0;
	M(3,2) = 0.0;
	M(3,3) = 1.0;

    #undef M
  }

// Lifted from Mesa
/*
 * Compute inverse of 4x4 transformation matrix.
 * Code contributed by Jacques Leroy jle@star.be
 * Return GL_TRUE for success, GL_FALSE for failure (singular matrix)
 */
bool m3dInvertMatrix44(M3DMatrix44f dst, const M3DMatrix44f src )
    {
    #define SWAP_ROWS(a, b) { float *_tmp = a; (a)=(b); (b)=_tmp; }
    #define MAT(m,r,c) (m)[(c)*4+(r)]

	float wtmp[4][8];
	float m0, m1, m2, m3, s;
	float *r0, *r1, *r2, *r3;

	r0 = wtmp[0], r1 = wtmp[1], r2 = wtmp[2], r3 = wtmp[3];

	r0[0] = MAT(src,0,0), r0[1] = MAT(src,0,1),
	r0[2] = MAT(src,0,2), r0[3] = MAT(src,0,3),
	r0[4] = 1.0, r0[5] = r0[6] = r0[7] = 0.0,

	r1[0] = MAT(src,1,0), r1[1] = MAT(src,1,1),
	r1[2] = MAT(src,1,2), r1[3] = MAT(src,1,3),
	r1[5] = 1.0, r1[4] = r1[6] = r1[7] = 0.0,

	r2[0] = MAT(src,2,0), r2[1] = MAT(src,2,1),
	r2[2] = MAT(src,2,2), r2[3] = MAT(src,2,3),
	r2[6] = 1.0, r2[4] = r2[5] = r2[7] = 0.0,

	r3[0] = MAT(src,3,0), r3[1] = MAT(src,3,1),
	r3[2] = MAT(src,3,2), r3[3] = MAT(src,3,3),
	r3[7] = 1.0, r3[4] = r3[5] = r3[6] = 0.0;

	/* choose pivot - or die */
	if (fabs(r3[0])>fabs(r2[0])) SWAP_ROWS(r3, r2);
	if (fabs(r2[0])>fabs(r1[0])) SWAP_ROWS(r2, r1);
	if (fabs(r1[0])>fabs(r0[0])) SWAP_ROWS(r1, r0);
	if (0.0 == r0[0])  return false;

	/* eliminate first variable     */
	m1 = r1[0]/r0[0]; m2 = r2[0]/r0[0]; m3 = r3[0]/r0[0];
	s = r0[1]; r1[1] -= m1 * s; r2[1] -= m2 * s; r3[1] -= m3 * s;
	s = r0[2]; r1[2] -= m1 * s; r2[2] -= m2 * s; r3[2] -= m3 * s;
	s = r0[3]; r1[3] -= m1 * s; r2[3] -= m2 * s; r3[3] -= m3 * s;
	s = r0[4];
	if (s != 0.0) { r1[4] -= m1 * s; r2[4] -= m2 * s; r3[4] -= m3 * s; }
	s = r0[5];
	if (s != 0.0) { r1[5] -= m1 * s; r2[5] -= m2 * s; r3[5] -= m3 * s; }
	s = r0[6];
	if (s != 0.0) { r1[6] -= m1 * s; r2[6] -= m2 * s; r3[6] -= m3 * s; }
	s = r0[7];
	if (s != 0.0) { r1[7] -= m1 * s; r2[7] -= m2 * s; r3[7] -= m3 * s; }

	/* choose pivot - or die */
	if (fabs(r3[1])>fabs(r2[1])) SWAP_ROWS(r3, r2);
	if (fabs(r2[1])>fabs(r1[1])) SWAP_ROWS(r2, r1);
	if (0.0 == r1[1])  return false;

	/* eliminate second variable */
	m2 = r2[1]/r1[1]; m3 = r3[1]/r1[1];
	r2[2] -= m2 * r1[2]; r3[2] -= m3 * r1[2];
	r2[3] -= m2 * r1[3]; r3[3] -= m3 * r1[3];
	s = r1[4]; if (0.0 != s) { r2[4] -= m2 * s; r3[4] -= m3 * s; }
	s = r1[5]; if (0.0 != s) { r2[5] -= m2 * s; r3[5] -= m3 * s; }
	s = r1[6]; if (0.0 != s) { r2[6] -= m2 * s; r3[6] -= m3 * s; }
	s = r1[7]; if (0.0 != s) { r2[7] -= m2 * s; r3[7] -= m3 * s; }

	/* choose pivot - or die */
	if (fabs(r3[2])>fabs(r2[2])) SWAP_ROWS(r3, r2);
	if (0.0 == r2[2])  return false;

	/* eliminate third variable */
	m3 = r3[2]/r2[2];
	r3[3] -= m3 * r2[3], r3[4] -= m3 * r2[4],
	r3[5] -= m3 * r2[5], r3[6] -= m3 * r2[6],
	r3[7] -= m3 * r2[7];

	/* last check */
	if (0.0 == r3[3]) return false;

	s = 1.0f/r3[3];              /* now back substitute row 3 */
	r3[4] *= s; r3[5] *= s; r3[6] *= s; r3[7] *= s;

	m2 = r2[3];                 /* now back substitute row 2 */
	s  = 1.0f/r2[2];
	r2[4] = s * (r2[4] - r3[4] * m2), r2[5] = s * (r2[5] - r3[5] * m2),
	r2[6] = s * (r2[6] - r3[6] * m2), r2[7] = s * (r2[7] - r3[7] * m2);
	m1 = r1[3];
	r1[4] -= r3[4] * m1, r1[5] -= r3[5] * m1,
	r1[6] -= r3[6] * m1, r1[7] -= r3[7] * m1;
	m0 = r0[3];
	r0[4] -= r3[4] * m0, r0[5] -= r3[5] * m0,
	r0[6] -= r3[6] * m0, r0[7] -= r3[7] * m0;

	m1 = r1[2];                 /* now back substitute row 1 */
	s  = 1.0f/r1[1];
	r1[4] = s * (r1[4] - r2[4] * m1), r1[5] = s * (r1[5] - r2[5] * m1),
	r1[6] = s * (r1[6] - r2[6] * m1), r1[7] = s * (r1[7] - r2[7] * m1);
	m0 = r0[2];
	r0[4] -= r2[4] * m0, r0[5] -= r2[5] * m0,
	r0[6] -= r2[6] * m0, r0[7] -= r2[7] * m0;

	m0 = r0[1];                 /* now back substitute row 0 */
	s  = 1.0f/r0[0];
	r0[4] = s * (r0[4] - r1[4] * m0), r0[5] = s * (r0[5] - r1[5] * m0),
	r0[6] = s * (r0[6] - r1[6] * m0), r0[7] = s * (r0[7] - r1[7] * m0);

	MAT(dst,0,0) = r0[4]; MAT(dst,0,1) = r0[5],
	MAT(dst,0,2) = r0[6]; MAT(dst,0,3) = r0[7],
	MAT(dst,1,0) = r1[4]; MAT(dst,1,1) = r1[5],
	MAT(dst,1,2) = r1[6]; MAT(dst,1,3) = r1[7],
	MAT(dst,2,0) = r2[4]; MAT(dst,2,1) = r2[5],
	MAT(dst,2,2) = r2[6]; MAT(dst,2,3) = r2[7],
	MAT(dst,3,0) = r3[4]; MAT(dst,3,1) = r3[5],
	MAT(dst,3,2) = r3[6]; MAT(dst,3,3) = r3[7]; 

	return true;

	#undef MAT
	#undef SWAP_ROWS
	}


// Ditto above, but for doubles
bool m3dInvertMatrix44(M3DMatrix44d dst, const M3DMatrix44d src)
	{
    #define SWAP_ROWS(a, b) { double *_tmp = a; (a)=(b); (b)=_tmp; }
    #define MAT(m,r,c) (m)[(c)*4+(r)]

	double wtmp[4][8];
	double m0, m1, m2, m3, s;
	double *r0, *r1, *r2, *r3;

	r0 = wtmp[0], r1 = wtmp[1], r2 = wtmp[2], r3 = wtmp[3];

	r0[0] = MAT(src,0,0), r0[1] = MAT(src,0,1),
	r0[2] = MAT(src,0,2), r0[3] = MAT(src,0,3),
	r0[4] = 1.0, r0[5] = r0[6] = r0[7] = 0.0,

	r1[0] = MAT(src,1,0), r1[1] = MAT(src,1,1),
	r1[2] = MAT(src,1,2), r1[3] = MAT(src,1,3),
	r1[5] = 1.0, r1[4] = r1[6] = r1[7] = 0.0,

	r2[0] = MAT(src,2,0), r2[1] = MAT(src,2,1),
	r2[2] = MAT(src,2,2), r2[3] = MAT(src,2,3),
	r2[6] = 1.0, r2[4] = r2[5] = r2[7] = 0.0,

	r3[0] = MAT(src,3,0), r3[1] = MAT(src,3,1),
	r3[2] = MAT(src,3,2), r3[3] = MAT(src,3,3),
	r3[7] = 1.0, r3[4] = r3[5] = r3[6] = 0.0;

	// choose pivot - or die 
	if (fabs(r3[0])>fabs(r2[0])) SWAP_ROWS(r3, r2);
	if (fabs(r2[0])>fabs(r1[0])) SWAP_ROWS(r2, r1);
	if (fabs(r1[0])>fabs(r0[0])) SWAP_ROWS(r1, r0);
	if (0.0 == r0[0])  return false;

	// eliminate first variable     
	m1 = r1[0]/r0[0]; m2 = r2[0]/r0[0]; m3 = r3[0]/r0[0];
	s = r0[1]; r1[1] -= m1 * s; r2[1] -= m2 * s; r3[1] -= m3 * s;
	s = r0[2]; r1[2] -= m1 * s; r2[2] -= m2 * s; r3[2] -= m3 * s;
	s = r0[3]; r1[3] -= m1 * s; r2[3] -= m2 * s; r3[3] -= m3 * s;
	s = r0[4];
	if (s != 0.0) { r1[4] -= m1 * s; r2[4] -= m2 * s; r3[4] -= m3 * s; }
	s = r0[5];
	if (s != 0.0) { r1[5] -= m1 * s; r2[5] -= m2 * s; r3[5] -= m3 * s; }
	s = r0[6];
	if (s != 0.0) { r1[6] -= m1 * s; r2[6] -= m2 * s; r3[6] -= m3 * s; }
	s = r0[7];
	if (s != 0.0) { r1[7] -= m1 * s; r2[7] -= m2 * s; r3[7] -= m3 * s; }

	// choose pivot - or die 
	if (fabs(r3[1])>fabs(r2[1])) SWAP_ROWS(r3, r2);
	if (fabs(r2[1])>fabs(r1[1])) SWAP_ROWS(r2, r1);
	if (0.0 == r1[1])  return false;

	// eliminate second variable 
	m2 = r2[1]/r1[1]; m3 = r3[1]/r1[1];
	r2[2] -= m2 * r1[2]; r3[2] -= m3 * r1[2];
	r2[3] -= m2 * r1[3]; r3[3] -= m3 * r1[3];
	s = r1[4]; if (0.0 != s) { r2[4] -= m2 * s; r3[4] -= m3 * s; }
	s = r1[5]; if (0.0 != s) { r2[5] -= m2 * s; r3[5] -= m3 * s; }
	s = r1[6]; if (0.0 != s) { r2[6] -= m2 * s; r3[6] -= m3 * s; }
	s = r1[7]; if (0.0 != s) { r2[7] -= m2 * s; r3[7] -= m3 * s; }

	// choose pivot - or die 
	if (fabs(r3[2])>fabs(r2[2])) SWAP_ROWS(r3, r2);
	if (0.0 == r2[2])  return false;

	// eliminate third variable 
	m3 = r3[2]/r2[2];
	r3[3] -= m3 * r2[3], r3[4] -= m3 * r2[4],
	r3[5] -= m3 * r2[5], r3[6] -= m3 * r2[6],
	r3[7] -= m3 * r2[7];

	// last check 
	if (0.0 == r3[3]) return false;

	s = 1.0f/r3[3];              // now back substitute row 3 
	r3[4] *= s; r3[5] *= s; r3[6] *= s; r3[7] *= s;

	m2 = r2[3];                 // now back substitute row 2 
	s  = 1.0f/r2[2];
	r2[4] = s * (r2[4] - r3[4] * m2), r2[5] = s * (r2[5] - r3[5] * m2),
	r2[6] = s * (r2[6] - r3[6] * m2), r2[7] = s * (r2[7] - r3[7] * m2);
	m1 = r1[3];
	r1[4] -= r3[4] * m1, r1[5] -= r3[5] * m1,
	r1[6] -= r3[6] * m1, r1[7] -= r3[7] * m1;
	m0 = r0[3];
	r0[4] -= r3[4] * m0, r0[5] -= r3[5] * m0,
	r0[6] -= r3[6] * m0, r0[7] -= r3[7] * m0;

	m1 = r1[2];                 // now back substitute row 1 
	s  = 1.0f/r1[1];
	r1[4] = s * (r1[4] - r2[4] * m1), r1[5] = s * (r1[5] - r2[5] * m1),
	r1[6] = s * (r1[6] - r2[6] * m1), r1[7] = s * (r1[7] - r2[7] * m1);
	m0 = r0[2];
	r0[4] -= r2[4] * m0, r0[5] -= r2[5] * m0,
	r0[6] -= r2[6] * m0, r0[7] -= r2[7] * m0;

	m0 = r0[1];                 // now back substitute row 0 
	s  = 1.0f/r0[0];
	r0[4] = s * (r0[4] - r1[4] * m0), r0[5] = s * (r0[5] - r1[5] * m0),
	r0[6] = s * (r0[6] - r1[6] * m0), r0[7] = s * (r0[7] - r1[7] * m0);

	MAT(dst,0,0) = r0[4]; MAT(dst,0,1) = r0[5],
	MAT(dst,0,2) = r0[6]; MAT(dst,0,3) = r0[7],
	MAT(dst,1,0) = r1[4]; MAT(dst,1,1) = r1[5],
	MAT(dst,1,2) = r1[6]; MAT(dst,1,3) = r1[7],
	MAT(dst,2,0) = r2[4]; MAT(dst,2,1) = r2[5],
	MAT(dst,2,2) = r2[6]; MAT(dst,2,3) = r2[7],
	MAT(dst,3,0) = r3[4]; MAT(dst,3,1) = r3[5],
	MAT(dst,3,2) = r3[6]; MAT(dst,3,3) = r3[7]; 

	return true;

	#undef MAT
	#undef SWAP_ROWS

	return true;
	}



///////////////////////////////////////////////////////////////////////////////////////
// Get Window coordinates, discard Z...
void m3dProjectXY(const M3DMatrix44f mModelView, const M3DMatrix44f mProjection, const int iViewPort[4], const M3DVector3f vPointIn, M3DVector2f vPointOut)
	{
    M3DVector4f vBack, vForth;

	memcpy(vBack, vPointIn, sizeof(float)*3);
	vBack[3] = 1.0f;
    
    m3dTransformVector4(vForth, vBack, mModelView);
    m3dTransformVector4(vBack, vForth, mProjection);
    
    if(!m3dCloseEnough(vBack[3], 0.0f, 0.000001f)) {
        float div = 1.0f / vBack[3];
        vBack[0] *= div;
        vBack[1] *= div;
        }


    vPointOut[0] = vBack[0] * 0.5f + 0.5f;
    vPointOut[1] = vBack[1] * 0.5f + 0.5f;

    /* Map x,y to viewport */
    vPointOut[0] = (vPointOut[0] * iViewPort[2]) + iViewPort[0];
    vPointOut[1] = (vPointOut[1] * iViewPort[3]) + iViewPort[1];    
	}
    
///////////////////////////////////////////////////////////////////////////////////////
// Get window coordinates, we also want Z....
void m3dProjectXYZ(const M3DMatrix44f mModelView, const M3DMatrix44f mProjection, const int iViewPort[4], const M3DVector3f vPointIn, M3DVector3f vPointOut)
	{
    M3DVector4f vBack, vForth;

	memcpy(vBack, vPointIn, sizeof(float)*3);
	vBack[3] = 1.0f;
    
    m3dTransformVector4(vForth, vBack, mModelView);
    m3dTransformVector4(vBack, vForth, mProjection);
    
    if(!m3dCloseEnough(vBack[3], 0.0f, 0.000001f)) {
        float div = 1.0f / vBack[3];
        vBack[0] *= div;
        vBack[1] *= div;
        vBack[2] *= div; 
        }

    vPointOut[0] = vBack[0] * 0.5f + 0.5f;
    vPointOut[1] = vBack[1] * 0.5f + 0.5f;
    vPointOut[2] = vBack[2] * 0.5f + 0.5f;

    /* Map x,y to viewport */
    vPointOut[0] = (vPointOut[0] * iViewPort[2]) + iViewPort[0];
    vPointOut[1] = (vPointOut[1] * iViewPort[3]) + iViewPort[1];
	}



///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
// Misc. Utilities
///////////////////////////////////////////////////////////////////////////////
// Calculates the normal of a triangle specified by the three points
// p1, p2, and p3. Each pointer points to an array of three floats. The
// triangle is assumed to be wound counter clockwise. 
void m3dFindNormal(M3DVector3f result, const M3DVector3f point1, const M3DVector3f point2, 
							const M3DVector3f point3)
	{
	M3DVector3f v1,v2;		// Temporary vectors

	// Calculate two vectors from the three points. Assumes counter clockwise
	// winding!
	v1[0] = point1[0] - point2[0];
	v1[1] = point1[1] - point2[1];
	v1[2] = point1[2] - point2[2];

	v2[0] = point2[0] - point3[0];
	v2[1] = point2[1] - point3[1];
	v2[2] = point2[2] - point3[2];

	// Take the cross product of the two vectors to get
	// the normal vector.
	m3dCrossProduct(result, v1, v2);
	}



// Ditto above, but for doubles
void m3dFindNormal(M3DVector3d result, const M3DVector3d point1, const M3DVector3d point2, 
							const M3DVector3d point3)
	{
	M3DVector3d v1,v2;		// Temporary vectors

	// Calculate two vectors from the three points. Assumes counter clockwise
	// winding!
	v1[0] = point1[0] - point2[0];
	v1[1] = point1[1] - point2[1];
	v1[2] = point1[2] - point2[2];

	v2[0] = point2[0] - point3[0];
	v2[1] = point2[1] - point3[1];
	v2[2] = point2[2] - point3[2];

	// Take the cross product of the two vectors to get
	// the normal vector.
	m3dCrossProduct(result, v1, v2);
	}



/////////////////////////////////////////////////////////////////////////////////////////
// Calculate the plane equation of the plane that the three specified points lay in. The
// points are given in clockwise winding order, with normal pointing out of clockwise face
// planeEq contains the A,B,C, and D of the plane equation coefficients
void m3dGetPlaneEquation(M3DVector4f planeEq, const M3DVector3f p1, const M3DVector3f p2, const M3DVector3f p3)
{
    // Get two vectors... do the cross product
    M3DVector3f v1, v2;

    // V1 = p3 - p1
    v1[0] = p3[0] - p1[0];
    v1[1] = p3[1] - p1[1];
    v1[2] = p3[2] - p1[2];

    // V2 = P2 - p1
    v2[0] = p2[0] - p1[0];
    v2[1] = p2[1] - p1[1];
    v2[2] = p2[2] - p1[2];

    // Unit normal to plane - Not sure which is the best way here
    m3dCrossProduct(planeEq, v1, v2);
    m3dNormalizeVector(planeEq);
    // Back substitute to get D
    planeEq[3] = -(planeEq[0] * p3[0] + planeEq[1] * p3[1] + planeEq[2] * p3[2]);
}


// Ditto above, but for doubles
void m3dGetPlaneEquation(M3DVector4d planeEq, const M3DVector3d p1, const M3DVector3d p2, const M3DVector3d p3)
{
    // Get two vectors... do the cross product
    M3DVector3d v1, v2;

    // V1 = p3 - p1
    v1[0] = p3[0] - p1[0];
    v1[1] = p3[1] - p1[1];
    v1[2] = p3[2] - p1[2];

    // V2 = P2 - p1
    v2[0] = p2[0] - p1[0];
    v2[1] = p2[1] - p1[1];
    v2[2] = p2[2] - p1[2];

    // Unit normal to plane - Not sure which is the best way here
    m3dCrossProduct(planeEq, v1, v2);
    m3dNormalizeVector(planeEq);
    // Back substitute to get D
    planeEq[3] = -(planeEq[0] * p3[0] + planeEq[1] * p3[1] + planeEq[2] * p3[2]);
}


//////////////////////////////////////////////////////////////////////////////////////////////////
// This function does a three dimensional Catmull-Rom curve interpolation. Pass four points, and a
// floating point number between 0.0 and 1.0. The curve is interpolated between the middle two points.
// Coded by RSW
// http://www.mvps.org/directx/articles/catmull/
void m3dCatmullRom3(M3DVector3f vOut, M3DVector3f vP0, M3DVector3f vP1, M3DVector3f vP2, M3DVector3f vP3, float t)
    {
    // Unrolled loop to speed things up a little bit...
    float t2 = t * t;
    float t3 = t2 * t;

    // X    
    vOut[0] = 0.5f * ( ( 2.0f * vP1[0]) +
                       (-vP0[0] + vP2[0]) * t +
                       (2.0f * vP0[0] - 5.0f *vP1[0] + 4.0f * vP2[0] - vP3[0]) * t2 +
                       (-vP0[0] + 3.0f*vP1[0] - 3.0f *vP2[0] + vP3[0]) * t3);
    // Y
    vOut[1] = 0.5f * ( ( 2.0f * vP1[1]) +
                       (-vP0[1] + vP2[1]) * t +
                       (2.0f * vP0[1] - 5.0f *vP1[1] + 4.0f * vP2[1] - vP3[1]) * t2 +
                       (-vP0[1] + 3.0f*vP1[1] - 3.0f *vP2[1] + vP3[1]) * t3);

    // Z
    vOut[2] = 0.5f * ( ( 2.0f * vP1[2]) +
                       (-vP0[2] + vP2[2]) * t +
                       (2.0f * vP0[2] - 5.0f *vP1[2] + 4.0f * vP2[2] - vP3[2]) * t2 +
                       (-vP0[2] + 3.0f*vP1[2] - 3.0f *vP2[2] + vP3[2]) * t3);
    }


//////////////////////////////////////////////////////////////////////////////////////////////////
// This function does a three dimensional Catmull-Rom curve interpolation. Pass four points, and a
// floating point number between 0.0 and 1.0. The curve is interpolated between the middle two points.
// Coded by RSW
// http://www.mvps.org/directx/articles/catmull/
void m3dCatmullRom3(M3DVector3d vOut, M3DVector3d vP0, M3DVector3d vP1, M3DVector3d vP2, M3DVector3d vP3, double t)
    {
    // Unrolled loop to speed things up a little bit...
    double t2 = t * t;
    double t3 = t2 * t;

    // X    
    vOut[0] = 0.5 * ( ( 2.0 * vP1[0]) +
                       (-vP0[0] + vP2[0]) * t +
                       (2.0 * vP0[0] - 5.0 *vP1[0] + 4.0 * vP2[0] - vP3[0]) * t2 +
                       (-vP0[0] + 3.0*vP1[0] - 3.0 *vP2[0] + vP3[0]) * t3);
    // Y
    vOut[1] = 0.5 * ( ( 2.0 * vP1[1]) +
                       (-vP0[1] + vP2[1]) * t +
                       (2.0 * vP0[1] - 5.0 *vP1[1] + 4.0 * vP2[1] - vP3[1]) * t2 +
                       (-vP0[1] + 3*vP1[1] - 3.0 *vP2[1] + vP3[1]) * t3);

    // Z
    vOut[2] = 0.5 * ( ( 2.0 * vP1[2]) +
                       (-vP0[2] + vP2[2]) * t +
                       (2.0 * vP0[2] - 5.0 *vP1[2] + 4.0 * vP2[2] - vP3[2]) * t2 +
                       (-vP0[2] + 3.0*vP1[2] - 3.0 *vP2[2] + vP3[2]) * t3);
    }


///////////////////////////////////////////////////////////////////////////////
// Determine if the ray (starting at point) intersects the sphere centered at
// sphereCenter with radius sphereRadius
// Return value is < 0 if the ray does not intersect
// Return value is 0.0 if ray is tangent
// Positive value is distance to the intersection point
// Algorithm from "3D Math Primer for Graphics and Game Development"
double m3dRaySphereTest(const M3DVector3d point, const M3DVector3d ray, const M3DVector3d sphereCenter, double sphereRadius)
	{
	//m3dNormalizeVector(ray);	// Make sure ray is unit length

	M3DVector3d rayToCenter;	// Ray to center of sphere
	rayToCenter[0] =  sphereCenter[0] - point[0];	
	rayToCenter[1] =  sphereCenter[1] - point[1];
	rayToCenter[2] =  sphereCenter[2] - point[2];
	
	// Project rayToCenter on ray to test
	double a = m3dDotProduct(rayToCenter, ray);
	
	// Distance to center of sphere
	double distance2 = m3dDotProduct(rayToCenter, rayToCenter);	// Or length

	
	double dRet = (sphereRadius * sphereRadius) - distance2 + (a*a);
	
	if(dRet > 0.0)			// Return distance to intersection
		dRet = a - sqrt(dRet);
	
	return dRet;
	}

///////////////////////////////////////////////////////////////////////////////
// Determine if the ray (starting at point) intersects the sphere centered at
// ditto above, but uses floating point math
float m3dRaySphereTest(const M3DVector3f point, const M3DVector3f ray, const M3DVector3f sphereCenter, float sphereRadius)
	{
	//m3dNormalizeVectorf(ray);	// Make sure ray is unit length

	M3DVector3f rayToCenter;	// Ray to center of sphere
	rayToCenter[0] =  sphereCenter[0] - point[0];	
	rayToCenter[1] =  sphereCenter[1] - point[1];
	rayToCenter[2] =  sphereCenter[2] - point[2];
	
	// Project rayToCenter on ray to test
	float a = m3dDotProduct(rayToCenter, ray);
	
	// Distance to center of sphere
	float distance2 = m3dDotProduct(rayToCenter, rayToCenter);	// Or length
	
	float dRet = (sphereRadius * sphereRadius) - distance2 + (a*a);
	
	if(dRet > 0.0)			// Return distance to intersection
		dRet = a - sqrtf(dRet);
	
	return dRet;
	}


///////////////////////////////////////////////////////////////////////////////////////////////////
// Calculate the tangent basis for a triangle on the surface of a model
// This vector is needed for most normal mapping shaders 
void m3dCalculateTangentBasis(const M3DVector3f vTriangle[3], const M3DVector2f vTexCoords[3], const M3DVector3f N, M3DVector3f vTangent)
{
    M3DVector3f dv2v1, dv3v1;
    float dc2c1t, dc2c1b, dc3c1t, dc3c1b;
    float M;
    
    m3dSubtractVectors3(dv2v1, vTriangle[1], vTriangle[0]);
    m3dSubtractVectors3(dv3v1, vTriangle[2], vTriangle[0]);
    
    dc2c1t = vTexCoords[1][0] - vTexCoords[0][0];
    dc2c1b = vTexCoords[1][1] - vTexCoords[0][1];
    dc3c1t = vTexCoords[2][0] - vTexCoords[0][0];
    dc3c1b = vTexCoords[2][1] - vTexCoords[0][1];
    
    M = (dc2c1t * dc3c1b) - (dc3c1t * dc2c1b);
    M = 1.0f / M;
    
    m3dScaleVector3(dv2v1, dc3c1b);
    m3dScaleVector3(dv3v1, dc2c1b);
    
    m3dSubtractVectors3(vTangent, dv2v1, dv3v1);
    m3dScaleVector3(vTangent, M);  // This potentially changes the direction of the vector
    m3dNormalizeVector(vTangent);

    M3DVector3f B;
    m3dCrossProduct(B, N, vTangent);
    m3dCrossProduct(vTangent, B, N);
    m3dNormalizeVector(vTangent);
    }
	
	
////////////////////////////////////////////////////////////////////////////
// Smoothly step between 0 and 1 between edge1 and edge 2
double m3dSmoothStep(double edge1, double edge2, double x)
    {
    double t;
    t = (x - edge1) / (edge2 - edge1);
    if(t > 1.0)
        t = 1.0;
        
    if(t < 0.0)
        t = 0.0f;
        
    return t * t * ( 3.0 - 2.0 * t);
    }

////////////////////////////////////////////////////////////////////////////
// Smoothly step between 0 and 1 between edge1 and edge 2
float m3dSmoothStep(float edge1, float edge2, float x)
    {
    float t;
    t = (x - edge1) / (edge2 - edge1);
    if(t > 1.0f)
        t = 1.0f;
        
    if(t < 0.0)
        t = 0.0f;
        
    return t * t * ( 3.0f - 2.0f * t);
    }
	
	

///////////////////////////////////////////////////////////////////////////
// Creae a projection to "squish" an object into the plane.
// Use m3dGetPlaneEquationf(planeEq, point1, point2, point3);
// to get a plane equation.
void m3dMakePlanarShadowMatrix(M3DMatrix44f proj, const M3DVector4f planeEq, const M3DVector3f vLightPos)
	{
	// These just make the code below easier to read. They will be 
	// removed by the optimizer.	
	float a = planeEq[0];
	float b = planeEq[1];
	float c = planeEq[2];
	float d = planeEq[3];

	float dx = -vLightPos[0];
	float dy = -vLightPos[1];
	float dz = -vLightPos[2];

	// Now build the projection matrix
	proj[0] = b * dy + c * dz;
	proj[1] = -a * dy;
	proj[2] = -a * dz;
	proj[3] = 0.0;

	proj[4] = -b * dx;
	proj[5] = a * dx + c * dz;
	proj[6] = -b * dz;
	proj[7] = 0.0;

	proj[8] = -c * dx;
	proj[9] = -c * dy;
	proj[10] = a * dx + b * dy;
	proj[11] = 0.0;

	proj[12] = -d * dx;
	proj[13] = -d * dy;
	proj[14] = -d * dz;
	proj[15] = a * dx + b * dy + c * dz;
	// Shadow matrix ready
	}
	
	
///////////////////////////////////////////////////////////////////////////
// Creae a projection to "squish" an object into the plane.
// Use m3dGetPlaneEquationd(planeEq, point1, point2, point3);
// to get a plane equation.
void m3dMakePlanarShadowMatrix(M3DMatrix44d proj, const M3DVector4d planeEq, const M3DVector3f vLightPos)
	{
	// These just make the code below easier to read. They will be 
	// removed by the optimizer.	
	double a = planeEq[0];
	double b = planeEq[1];
	double c = planeEq[2];
	double d = planeEq[3];

	double dx = -vLightPos[0];
	double dy = -vLightPos[1];
	double dz = -vLightPos[2];

	// Now build the projection matrix
	proj[0] = b * dy + c * dz;
	proj[1] = -a * dy;
	proj[2] = -a * dz;
	proj[3] = 0.0;

	proj[4] = -b * dx;
	proj[5] = a * dx + c * dz;
	proj[6] = -b * dz;
	proj[7] = 0.0;

	proj[8] = -c * dx;
	proj[9] = -c * dy;
	proj[10] = a * dx + b * dy;
	proj[11] = 0.0;

	proj[12] = -d * dx;
	proj[13] = -d * dy;
	proj[14] = -d * dz;
	proj[15] = a * dx + b * dy + c * dz;
	// Shadow matrix ready
	}


/////////////////////////////////////////////////////////////////////////////
// I want to know the point on a ray, closest to another given point in space.
// As a bonus, return the distance squared of the two points.
// In: vRayOrigin is the origin of the ray.
// In: vUnitRayDir is the unit vector of the ray
// In: vPointInSpace is the point in space
// Out: vPointOnRay is the poing on the ray closest to vPointInSpace
// Return: The square of the distance to the ray
double m3dClosestPointOnRay(M3DVector3d vPointOnRay, const M3DVector3d vRayOrigin, const M3DVector3d vUnitRayDir, 
											const M3DVector3d vPointInSpace)
	{
	M3DVector3d v;
	m3dSubtractVectors3(v, vPointInSpace, vRayOrigin);
	
	double t = m3dDotProduct(vUnitRayDir, v);
	
	// This is the point on the ray
	vPointOnRay[0] = vRayOrigin[0] + (t * vUnitRayDir[0]);
	vPointOnRay[1] = vRayOrigin[1] + (t * vUnitRayDir[1]);
	vPointOnRay[2] = vRayOrigin[2] + (t * vUnitRayDir[2]);
	
	return m3dGetDistanceSquared(vPointOnRay, vPointInSpace);
	}

// ditto above... but with floats
float m3dClosestPointOnRay(M3DVector3f vPointOnRay, const M3DVector3f vRayOrigin, const M3DVector3f vUnitRayDir, 
							 const M3DVector3f vPointInSpace)
	{
	M3DVector3f v;
	m3dSubtractVectors3(v, vPointInSpace, vRayOrigin);
	
	float t = m3dDotProduct(vUnitRayDir, v);
	
	// This is the point on the ray
	vPointOnRay[0] = vRayOrigin[0] + (t * vUnitRayDir[0]);
	vPointOnRay[1] = vRayOrigin[1] + (t * vUnitRayDir[1]);
	vPointOnRay[2] = vRayOrigin[2] + (t * vUnitRayDir[2]);
	
	return m3dGetDistanceSquared(vPointOnRay, vPointInSpace);
	}


原文地址:https://www.cnblogs.com/jiangu66/p/3167818.html