// SONIC ROBO BLAST 2 //----------------------------------------------------------------------------- // Copyright (C) 1993-1996 by id Software, Inc. // Copyright (C) 1998-2000 by DooM Legacy Team. // Copyright (C) 1999-2024 by Sonic Team Junior. // Copyright (C) 2009 by Stephen McGranahan. // // This program is free software distributed under the // terms of the GNU General Public License, version 2. // See the 'LICENSE' file for more details. //----------------------------------------------------------------------------- /// \file m_fixed.c /// \brief Fixed point implementation #if 0 //#ifndef NO_M #include #define HAVE_SQRT #if 0 //#ifndef _WIN32 // MSVCRT does not have *f() functions #define HAVE_SQRTF #endif #endif #include "doomdef.h" #include "m_fixed.h" fixed_t FixedSqrt(fixed_t x) { #ifdef HAVE_SQRT const float fx = FIXED_TO_FLOAT(x); float fr; #ifdef HAVE_SQRTF fr = sqrtf(fx); #else fr = (float)sqrt(fx); #endif return FLOAT_TO_FIXED(fr); #else // The neglected art of Fixed Point arithmetic // Jetro Lauha // Seminar Presentation // Assembly 2006, 3rd- 6th August 2006 // (Revised: September 13, 2006) // URL: http://jet.ro/files/The_neglected_art_of_Fixed_Point_arithmetic_20060913.pdf register UINT32 root, remHi, remLo, testDiv, count; root = 0; /* Clear root */ remHi = 0; /* Clear high part of partial remainder */ remLo = x; /* Get argument into low part of partial remainder */ count = (15 + (FRACBITS >> 1)); /* Load loop counter */ do { remHi = (remHi << 2) | (remLo >> 30); remLo <<= 2; /* get 2 bits of arg */ root <<= 1; /* Get ready for the next bit in the root */ testDiv = (root << 1) + 1; /* Test radical */ if (remHi >= testDiv) { remHi -= testDiv; root += 1; } } while (count-- != 0); return root; #endif } fixed_t FixedHypot(fixed_t x, fixed_t y) { fixed_t ax, yx, yx2, yx1; if (abs(y) > abs(x)) // |y|>|x| { ax = abs(y); // |y| => ax yx = FixedDiv(x, y); // (x/y) } else // |x|>|y| { ax = abs(x); // |x| => ax yx = FixedDiv(y, x); // (x/y) } yx2 = FixedMul(yx, yx); // (x/y)^2 yx1 = FixedSqrt(1 * FRACUNIT + yx2); // (1 + (x/y)^2)^1/2 return FixedMul(ax, yx1); // |x|*((1 + (x/y)^2)^1/2) } vector2_t *FV2_Load(vector2_t *vec, fixed_t x, fixed_t y) { vec->x = x; vec->y = y; return vec; } vector2_t *FV2_UnLoad(vector2_t *vec, fixed_t *x, fixed_t *y) { *x = vec->x; *y = vec->y; return vec; } vector2_t *FV2_Copy(vector2_t *a_o, const vector2_t *a_i) { return M_Memcpy(a_o, a_i, sizeof(vector2_t)); } vector2_t *FV2_AddEx(const vector2_t *a_i, const vector2_t *a_c, vector2_t *a_o) { a_o->x = a_i->x + a_c->x; a_o->y = a_i->y + a_c->y; return a_o; } vector2_t *FV2_Add(vector2_t *a_i, const vector2_t *a_c) { return FV2_AddEx(a_i, a_c, a_i); } vector2_t *FV2_SubEx(const vector2_t *a_i, const vector2_t *a_c, vector2_t *a_o) { a_o->x = a_i->x - a_c->x; a_o->y = a_i->y - a_c->y; return a_o; } vector2_t *FV2_Sub(vector2_t *a_i, const vector2_t *a_c) { return FV2_SubEx(a_i, a_c, a_i); } vector2_t *FV2_MulEx(const vector2_t *a_i, fixed_t a_c, vector2_t *a_o) { a_o->x = FixedMul(a_i->x, a_c); a_o->y = FixedMul(a_i->y, a_c); return a_o; } vector2_t *FV2_Mul(vector2_t *a_i, fixed_t a_c) { return FV2_MulEx(a_i, a_c, a_i); } vector2_t *FV2_DivideEx(const vector2_t *a_i, fixed_t a_c, vector2_t *a_o) { a_o->x = FixedDiv(a_i->x, a_c); a_o->y = FixedDiv(a_i->y, a_c); return a_o; } vector2_t *FV2_Divide(vector2_t *a_i, fixed_t a_c) { return FV2_DivideEx(a_i, a_c, a_i); } // Vector Complex Math vector2_t *FV2_Midpoint(const vector2_t *a_1, const vector2_t *a_2, vector2_t *a_o) { a_o->x = FixedDiv(a_2->x - a_1->x, 2 * FRACUNIT); a_o->y = FixedDiv(a_2->y - a_1->y, 2 * FRACUNIT); a_o->x = a_1->x + a_o->x; a_o->y = a_1->y + a_o->y; return a_o; } fixed_t FV2_Distance(const vector2_t *p1, const vector2_t *p2) { fixed_t xs = FixedMul(p2->x - p1->x, p2->x - p1->x); fixed_t ys = FixedMul(p2->y - p1->y, p2->y - p1->y); return FixedSqrt(xs + ys); } fixed_t FV2_Magnitude(const vector2_t *a_normal) { fixed_t xs = FixedMul(a_normal->x, a_normal->x); fixed_t ys = FixedMul(a_normal->y, a_normal->y); return FixedSqrt(xs + ys); } // Also returns the magnitude fixed_t FV2_NormalizeEx(const vector2_t *a_normal, vector2_t *a_o) { fixed_t magnitude = FV2_Magnitude(a_normal); a_o->x = FixedDiv(a_normal->x, magnitude); a_o->y = FixedDiv(a_normal->y, magnitude); return magnitude; } fixed_t FV2_Normalize(vector2_t *a_normal) { return FV2_NormalizeEx(a_normal, a_normal); } vector2_t *FV2_NegateEx(const vector2_t *a_1, vector2_t *a_o) { a_o->x = -a_1->x; a_o->y = -a_1->y; return a_o; } vector2_t *FV2_Negate(vector2_t *a_1) { return FV2_NegateEx(a_1, a_1); } boolean FV2_Equal(const vector2_t *a_1, const vector2_t *a_2) { fixed_t Epsilon = FRACUNIT / FRACUNIT; if ((abs(a_2->x - a_1->x) > Epsilon) || (abs(a_2->y - a_1->y) > Epsilon)) { return true; } return false; } fixed_t FV2_Dot(const vector2_t *a_1, const vector2_t *a_2) { return (FixedMul(a_1->x, a_2->x) + FixedMul(a_1->y, a_2->y)); } // // Point2Vec // // Given two points, create a vector between them. // vector2_t *FV2_Point2Vec(const vector2_t *point1, const vector2_t *point2, vector2_t *a_o) { a_o->x = point1->x - point2->x; a_o->y = point1->y - point2->y; return a_o; } vector3_t *FV3_Load(vector3_t *vec, fixed_t x, fixed_t y, fixed_t z) { vec->x = x; vec->y = y; vec->z = z; return vec; } vector4_t *FV4_Load(vector4_t *vec, fixed_t x, fixed_t y, fixed_t z, fixed_t a) { vec->x = x; vec->y = y; vec->z = z; vec->a = a; return vec; } vector3_t *FV3_UnLoad(vector3_t *vec, fixed_t *x, fixed_t *y, fixed_t *z) { *x = vec->x; *y = vec->y; *z = vec->z; return vec; } vector4_t *FV4_UnLoad(vector4_t *vec, fixed_t *x, fixed_t *y, fixed_t *z, fixed_t *a) { *x = vec->x; *y = vec->y; *z = vec->z; *a = vec->a; return vec; } vector3_t *FV3_Copy(vector3_t *a_o, const vector3_t *a_i) { return M_Memcpy(a_o, a_i, sizeof(vector3_t)); } vector4_t *FV4_Copy(vector4_t *a_o, const vector4_t *a_i) { return M_Memcpy(a_o, a_i, sizeof(vector4_t)); } vector3_t *FV3_AddEx(const vector3_t *a_i, const vector3_t *a_c, vector3_t *a_o) { a_o->x = a_i->x + a_c->x; a_o->y = a_i->y + a_c->y; a_o->z = a_i->z + a_c->z; return a_o; } vector4_t *FV4_AddEx(const vector4_t *a_i, const vector4_t *a_c, vector4_t *a_o) { a_o->x = a_i->x + a_c->x; a_o->y = a_i->y + a_c->y; a_o->z = a_i->z + a_c->z; a_o->a = a_i->a + a_c->a; return a_o; } vector3_t *FV3_Add(vector3_t *a_i, const vector3_t *a_c) { return FV3_AddEx(a_i, a_c, a_i); } vector4_t *FV4_Add(vector4_t *a_i, const vector4_t *a_c) { return FV4_AddEx(a_i, a_c, a_i); } vector3_t *FV3_SubEx(const vector3_t *a_i, const vector3_t *a_c, vector3_t *a_o) { a_o->x = a_i->x - a_c->x; a_o->y = a_i->y - a_c->y; a_o->z = a_i->z - a_c->z; return a_o; } vector4_t *FV4_SubEx(const vector4_t *a_i, const vector4_t *a_c, vector4_t *a_o) { a_o->x = a_i->x - a_c->x; a_o->y = a_i->y - a_c->y; a_o->z = a_i->z - a_c->z; a_o->a = a_i->a - a_c->a; return a_o; } vector3_t *FV3_Sub(vector3_t *a_i, const vector3_t *a_c) { return FV3_SubEx(a_i, a_c, a_i); } vector4_t *FV4_Sub(vector4_t *a_i, const vector4_t *a_c) { return FV4_SubEx(a_i, a_c, a_i); } vector3_t *FV3_MulEx(const vector3_t *a_i, fixed_t a_c, vector3_t *a_o) { a_o->x = FixedMul(a_i->x, a_c); a_o->y = FixedMul(a_i->y, a_c); a_o->z = FixedMul(a_i->z, a_c); return a_o; } vector4_t *FV4_MulEx(const vector4_t *a_i, fixed_t a_c, vector4_t *a_o) { a_o->x = FixedMul(a_i->x, a_c); a_o->y = FixedMul(a_i->y, a_c); a_o->z = FixedMul(a_i->z, a_c); a_o->a = FixedMul(a_i->a, a_c); return a_o; } vector3_t *FV3_Mul(vector3_t *a_i, fixed_t a_c) { return FV3_MulEx(a_i, a_c, a_i); } vector4_t *FV4_Mul(vector4_t *a_i, fixed_t a_c) { return FV4_MulEx(a_i, a_c, a_i); } vector3_t *FV3_DivideEx(const vector3_t *a_i, fixed_t a_c, vector3_t *a_o) { a_o->x = FixedDiv(a_i->x, a_c); a_o->y = FixedDiv(a_i->y, a_c); a_o->z = FixedDiv(a_i->z, a_c); return a_o; } vector4_t *FV4_DivideEx(const vector4_t *a_i, fixed_t a_c, vector4_t *a_o) { a_o->x = FixedDiv(a_i->x, a_c); a_o->y = FixedDiv(a_i->y, a_c); a_o->z = FixedDiv(a_i->z, a_c); a_o->a = FixedDiv(a_i->a, a_c); return a_o; } vector3_t *FV3_Divide(vector3_t *a_i, fixed_t a_c) { return FV3_DivideEx(a_i, a_c, a_i); } vector4_t *FV4_Divide(vector4_t *a_i, fixed_t a_c) { return FV4_DivideEx(a_i, a_c, a_i); } // Vector Complex Math vector3_t *FV3_Midpoint(const vector3_t *a_1, const vector3_t *a_2, vector3_t *a_o) { a_o->x = FixedDiv(a_2->x - a_1->x, 2 * FRACUNIT); a_o->y = FixedDiv(a_2->y - a_1->y, 2 * FRACUNIT); a_o->z = FixedDiv(a_2->z - a_1->z, 2 * FRACUNIT); a_o->x = a_1->x + a_o->x; a_o->y = a_1->y + a_o->y; a_o->z = a_1->z + a_o->z; return a_o; } vector4_t *FV4_Midpoint(const vector4_t *a_1, const vector4_t *a_2, vector4_t *a_o) { a_o->x = FixedDiv(a_2->x - a_1->x, 2 * FRACUNIT); a_o->y = FixedDiv(a_2->y - a_1->y, 2 * FRACUNIT); a_o->z = FixedDiv(a_2->z - a_1->z, 2 * FRACUNIT); a_o->a = FixedDiv(a_2->a - a_1->a, 2 * FRACUNIT); a_o->x = a_1->x + a_o->x; a_o->y = a_1->y + a_o->y; a_o->z = a_1->z + a_o->z; a_o->a = a_1->z + a_o->a; return a_o; } fixed_t FV3_Distance(const vector3_t *p1, const vector3_t *p2) { fixed_t xs = FixedMul(p2->x - p1->x, p2->x - p1->x); fixed_t ys = FixedMul(p2->y - p1->y, p2->y - p1->y); fixed_t zs = FixedMul(p2->z - p1->z, p2->z - p1->z); return FixedSqrt(xs + ys + zs); } fixed_t FV4_Distance(const vector4_t *p1, const vector4_t *p2) { fixed_t xs = FixedMul(p2->x - p1->x, p2->x - p1->x); fixed_t ys = FixedMul(p2->y - p1->y, p2->y - p1->y); fixed_t zs = FixedMul(p2->z - p1->z, p2->z - p1->z); fixed_t za = FixedMul(p2->a - p1->a, p2->a - p1->a); return FixedSqrt(xs + ys + zs + za); } fixed_t FV3_Magnitude(const vector3_t *a_normal) { fixed_t xs = FixedMul(a_normal->x, a_normal->x); fixed_t ys = FixedMul(a_normal->y, a_normal->y); fixed_t zs = FixedMul(a_normal->z, a_normal->z); return FixedSqrt(xs + ys + zs); } fixed_t FV4_Magnitude(const vector4_t *a_normal) { fixed_t xs = FixedMul(a_normal->x, a_normal->x); fixed_t ys = FixedMul(a_normal->y, a_normal->y); fixed_t zs = FixedMul(a_normal->z, a_normal->z); fixed_t as = FixedMul(a_normal->a, a_normal->a); return FixedSqrt(xs + ys + zs + as); } // Also returns the magnitude fixed_t FV3_NormalizeEx(const vector3_t *a_normal, vector3_t *a_o) { fixed_t magnitude = FV3_Magnitude(a_normal); a_o->x = FixedDiv(a_normal->x, magnitude); a_o->y = FixedDiv(a_normal->y, magnitude); a_o->z = FixedDiv(a_normal->z, magnitude); return magnitude; } fixed_t FV4_NormalizeEx(const vector4_t *a_normal, vector4_t *a_o) { fixed_t magnitude = FV4_Magnitude(a_normal); a_o->x = FixedDiv(a_normal->x, magnitude); a_o->y = FixedDiv(a_normal->y, magnitude); a_o->z = FixedDiv(a_normal->z, magnitude); a_o->a = FixedDiv(a_normal->a, magnitude); return magnitude; } fixed_t FV3_Normalize(vector3_t *a_normal) { return FV3_NormalizeEx(a_normal, a_normal); } fixed_t FV4_Normalize(vector4_t *a_normal) { return FV4_NormalizeEx(a_normal, a_normal); } vector3_t *FV3_NegateEx(const vector3_t *a_1, vector3_t *a_o) { a_o->x = -a_1->x; a_o->y = -a_1->y; a_o->z = -a_1->z; return a_o; } vector4_t *FV4_NegateEx(const vector4_t *a_1, vector4_t *a_o) { a_o->x = -a_1->x; a_o->y = -a_1->y; a_o->z = -a_1->z; a_o->a = -a_1->a; return a_o; } vector3_t *FV3_Negate(vector3_t *a_1) { return FV3_NegateEx(a_1, a_1); } vector4_t *FV4_Negate(vector4_t *a_1) { return FV4_NegateEx(a_1, a_1); } boolean FV3_Equal(const vector3_t *a_1, const vector3_t *a_2) { fixed_t Epsilon = FRACUNIT / FRACUNIT; if ((abs(a_2->x - a_1->x) > Epsilon) || (abs(a_2->y - a_1->y) > Epsilon) || (abs(a_2->z - a_1->z) > Epsilon)) { return true; } return false; } boolean FV4_Equal(const vector4_t *a_1, const vector4_t *a_2) { fixed_t Epsilon = FRACUNIT / FRACUNIT; if ((abs(a_2->x - a_1->x) > Epsilon) || (abs(a_2->y - a_1->y) > Epsilon) || (abs(a_2->z - a_1->z) > Epsilon) || (abs(a_2->a - a_1->a) > Epsilon)) { return true; } return false; } fixed_t FV3_Dot(const vector3_t *a_1, const vector3_t *a_2) { return (FixedMul(a_1->x, a_2->x) + FixedMul(a_1->y, a_2->y) + FixedMul(a_1->z, a_2->z)); } fixed_t FV4_Dot(const vector4_t *a_1, const vector4_t *a_2) { return (FixedMul(a_1->x, a_2->x) + FixedMul(a_1->y, a_2->y) + FixedMul(a_1->z, a_2->z) + FixedMul(a_1->a, a_2->a)); } vector3_t *FV3_Cross(const vector3_t *a_1, const vector3_t *a_2, vector3_t *a_o) { a_o->x = FixedMul(a_1->y, a_2->z) - FixedMul(a_1->z, a_2->y); a_o->y = FixedMul(a_1->z, a_2->x) - FixedMul(a_1->x, a_2->z); a_o->z = FixedMul(a_1->x, a_2->y) - FixedMul(a_1->y, a_2->x); return a_o; } // // ClosestPointOnLine // // Finds the point on a line closest // to the specified point. // vector3_t *FV3_ClosestPointOnLine(const vector3_t *Line, const vector3_t *p, vector3_t *out) { // Determine t (the length of the vector from "Line[0]" to "p") vector3_t c, V; fixed_t t, d = 0; FV3_SubEx(p, &Line[0], &c); FV3_SubEx(&Line[1], &Line[0], &V); FV3_NormalizeEx(&V, &V); d = FV3_Distance(&Line[0], &Line[1]); t = FV3_Dot(&V, &c); // Check to see if "t" is beyond the extents of the line segment if (t < 0) { return FV3_Copy(out, &Line[0]); } if (t > d) { return FV3_Copy(out, &Line[1]); } // Return the point between "Line[0]" and "Line[1]" FV3_Mul(&V, t); return FV3_AddEx(&Line[0], &V, out); } // // ClosestPointOnVector // // Similar to ClosestPointOnLine, but uses a vector instead of two points. // void FV3_ClosestPointOnVector(const vector3_t *dir, const vector3_t *p, vector3_t *out) { fixed_t t = FV3_Dot(dir, p); // Return the point on the line closest FV3_MulEx(dir, t, out); return; } // // ClosestPointOnTriangle // // Given a triangle and a point, // the closest point on the edge of // the triangle is returned. // void FV3_ClosestPointOnTriangle(const vector3_t *tri, const vector3_t *point, vector3_t *result) { UINT8 i; fixed_t dist, closestdist; vector3_t EdgePoints[3]; vector3_t Line[2]; FV3_Copy(&Line[0], &tri[0]); FV3_Copy(&Line[1], &tri[1]); FV3_ClosestPointOnLine(Line, point, &EdgePoints[0]); FV3_Copy(&Line[0], &tri[1]); FV3_Copy(&Line[1], &tri[2]); FV3_ClosestPointOnLine(Line, point, &EdgePoints[1]); FV3_Copy(&Line[0], &tri[2]); FV3_Copy(&Line[1], &tri[0]); FV3_ClosestPointOnLine(Line, point, &EdgePoints[2]); // Find the closest one of the three FV3_Copy(result, &EdgePoints[0]); closestdist = FV3_Distance(point, &EdgePoints[0]); for (i = 1; i < 3; i++) { dist = FV3_Distance(point, &EdgePoints[i]); if (dist < closestdist) { closestdist = dist; FV3_Copy(result, &EdgePoints[i]); } } // We now have the closest point! Whee! } // // Point2Vec // // Given two points, create a vector between them. // vector3_t *FV3_Point2Vec(const vector3_t *point1, const vector3_t *point2, vector3_t *a_o) { a_o->x = point1->x - point2->x; a_o->y = point1->y - point2->y; a_o->z = point1->z - point2->z; return a_o; } // // Normal // // Calculates the normal of a polygon. // fixed_t FV3_Normal(const vector3_t *a_triangle, vector3_t *a_normal) { vector3_t a_1; vector3_t a_2; FV3_Point2Vec(&a_triangle[2], &a_triangle[0], &a_1); FV3_Point2Vec(&a_triangle[1], &a_triangle[0], &a_2); FV3_Cross(&a_1, &a_2, a_normal); return FV3_NormalizeEx(a_normal, a_normal); } // // Strength // // Measures the 'strength' of a vector in a particular direction. // fixed_t FV3_Strength(const vector3_t *a_1, const vector3_t *dir) { vector3_t normal; fixed_t dist = FV3_NormalizeEx(a_1, &normal); fixed_t dot = FV3_Dot(&normal, dir); FV3_ClosestPointOnVector(dir, a_1, &normal); dist = FV3_Magnitude(&normal); if (dot < 0) // Not facing same direction, so negate result. dist = -dist; return dist; } // // PlaneDistance // // Calculates distance between a plane and the origin. // fixed_t FV3_PlaneDistance(const vector3_t *a_normal, const vector3_t *a_point) { return -(FixedMul(a_normal->x, a_point->x) + FixedMul(a_normal->y, a_point->y) + FixedMul(a_normal->z, a_point->z)); } boolean FV3_IntersectedPlane(const vector3_t *a_triangle, const vector3_t *a_line, vector3_t *a_normal, fixed_t *originDistance) { fixed_t distance1 = 0, distance2 = 0; FV3_Normal(a_triangle, a_normal); *originDistance = FV3_PlaneDistance(a_normal, &a_triangle[0]); distance1 = (FixedMul(a_normal->x, a_line[0].x) + FixedMul(a_normal->y, a_line[0].y) + FixedMul(a_normal->z, a_line[0].z)) + *originDistance; distance2 = (FixedMul(a_normal->x, a_line[1].x) + FixedMul(a_normal->y, a_line[1].y) + FixedMul(a_normal->z, a_line[1].z)) + *originDistance; // Positive or zero number means no intersection if (FixedMul(distance1, distance2) >= 0) return false; return true; } // // PlaneIntersection // // Returns the distance from // rOrigin to where the ray // intersects the plane. Assumes // you already know it intersects // the plane. // fixed_t FV3_PlaneIntersection(const vector3_t *pOrigin, const vector3_t *pNormal, const vector3_t *rOrigin, const vector3_t *rVector) { fixed_t d = -(FV3_Dot(pNormal, pOrigin)); fixed_t number = FV3_Dot(pNormal, rOrigin) + d; fixed_t denom = FV3_Dot(pNormal, rVector); return -FixedDiv(number, denom); } // // IntersectRaySphere // Input : rO - origin of ray in world space // rV - vector describing direction of ray in world space // sO - Origin of sphere // sR - radius of sphere // Notes : Normalized directional vectors expected // Return: distance to sphere in world units, -1 if no intersection. // fixed_t FV3_IntersectRaySphere(const vector3_t *rO, const vector3_t *rV, const vector3_t *sO, fixed_t sR) { vector3_t Q; fixed_t c, v, d; FV3_SubEx(sO, rO, &Q); c = FV3_Magnitude(&Q); v = FV3_Dot(&Q, rV); d = FixedMul(sR, sR) - (FixedMul(c, c) - FixedMul(v, v)); // If there was no intersection, return -1 if (d < 0 * FRACUNIT) return (-1 * FRACUNIT); // Return the distance to the [first] intersecting point return (v - FixedSqrt(d)); } // // IntersectionPoint // // This returns the intersection point of the line that intersects the plane // vector3_t *FV3_IntersectionPoint(const vector3_t *vNormal, const vector3_t *vLine, fixed_t distance, vector3_t *ReturnVec) { vector3_t vLineDir; // Variables to hold the point and the line's direction fixed_t Numerator = 0, Denominator = 0, dist = 0; // Here comes the confusing part. We need to find the 3D point that is actually // on the plane. Here are some steps to do that: // 1) First we need to get the vector of our line, Then normalize it so it's a length of 1 FV3_Point2Vec(&vLine[1], &vLine[0], &vLineDir); // Get the Vector of the line FV3_NormalizeEx(&vLineDir, &vLineDir); // Normalize the lines vector // 2) Use the plane equation (distance = Ax + By + Cz + D) to find the distance from one of our points to the plane. // Here I just chose a arbitrary point as the point to find that distance. You notice we negate that // distance. We negate the distance because we want to eventually go BACKWARDS from our point to the plane. // By doing this is will basically bring us back to the plane to find our intersection point. Numerator = -(FixedMul(vNormal->x, vLine[0].x) + // Use the plane equation with the normal and the line FixedMul(vNormal->y, vLine[0].y) + FixedMul(vNormal->z, vLine[0].z) + distance); // 3) If we take the dot product between our line vector and the normal of the polygon, // this will give us the cosine of the angle between the 2 (since they are both normalized - length 1). // We will then divide our Numerator by this value to find the offset towards the plane from our arbitrary point. Denominator = FV3_Dot(vNormal, &vLineDir); // Get the dot product of the line's vector and the normal of the plane // Since we are using division, we need to make sure we don't get a divide by zero error // If we do get a 0, that means that there are INFINITE points because the the line is // on the plane (the normal is perpendicular to the line - (Normal.Vector = 0)). // In this case, we should just return any point on the line. if (Denominator == 0 * FRACUNIT) // Check so we don't divide by zero { ReturnVec->x = vLine[0].x; ReturnVec->y = vLine[0].y; ReturnVec->z = vLine[0].z; return ReturnVec; // Return an arbitrary point on the line } // We divide the (distance from the point to the plane) by (the dot product) // to get the distance (dist) that we need to move from our arbitrary point. We need // to then times this distance (dist) by our line's vector (direction). When you times // a scalar (single number) by a vector you move along that vector. That is what we are // doing. We are moving from our arbitrary point we chose from the line BACK to the plane // along the lines vector. It seems logical to just get the numerator, which is the distance // from the point to the line, and then just move back that much along the line's vector. // Well, the distance from the plane means the SHORTEST distance. What about in the case that // the line is almost parallel with the polygon, but doesn't actually intersect it until half // way down the line's length. The distance from the plane is short, but the distance from // the actual intersection point is pretty long. If we divide the distance by the dot product // of our line vector and the normal of the plane, we get the correct length. Cool huh? dist = FixedDiv(Numerator, Denominator); // Divide to get the multiplying (percentage) factor // Now, like we said above, we times the dist by the vector, then add our arbitrary point. // This essentially moves the point along the vector to a certain distance. This now gives // us the intersection point. Yay! // Return the intersection point ReturnVec->x = vLine[0].x + FixedMul(vLineDir.x, dist); ReturnVec->y = vLine[0].y + FixedMul(vLineDir.y, dist); ReturnVec->z = vLine[0].z + FixedMul(vLineDir.z, dist); return ReturnVec; } // // PointOnLineSide // // If on the front side of the line, returns 1. // If on the back side of the line, returns 0. // 2D only. // UINT8 FV3_PointOnLineSide(const vector3_t *point, const vector3_t *line) { fixed_t s1 = FixedMul((point->y - line[0].y), (line[1].x - line[0].x)); fixed_t s2 = FixedMul((point->x - line[0].x), (line[1].y - line[0].y)); return (UINT8)(s1 - s2 < 0); } // // PointInsideBox // // Given four points of a box, // determines if the supplied point is // inside the box or not. // boolean FV3_PointInsideBox(const vector3_t *point, const vector3_t *box) { vector3_t lastLine[2]; FV3_Load(&lastLine[0], box[3].x, box[3].y, box[3].z); FV3_Load(&lastLine[1], box[0].x, box[0].y, box[0].z); if (FV3_PointOnLineSide(point, &box[0]) || FV3_PointOnLineSide(point, &box[1]) || FV3_PointOnLineSide(point, &box[2]) || FV3_PointOnLineSide(point, lastLine)) return false; return true; } // // LoadIdentity // // Loads the identity matrix into a matrix // void FM_LoadIdentity(matrix_t* matrix) { #define M(row,col) matrix->m[col * 4 + row] memset(matrix, 0x00, sizeof(matrix_t)); M(0, 0) = FRACUNIT; M(1, 1) = FRACUNIT; M(2, 2) = FRACUNIT; M(3, 3) = FRACUNIT; #undef M } // // CreateObjectMatrix // // Creates a matrix that can be used for // adjusting the position of an object // void FM_CreateObjectMatrix(matrix_t *matrix, fixed_t x, fixed_t y, fixed_t z, fixed_t anglex, fixed_t angley, fixed_t anglez, fixed_t upx, fixed_t upy, fixed_t upz, fixed_t radius) { vector3_t upcross; vector3_t upvec; vector3_t basevec; FV3_Load(&upvec, upx, upy, upz); FV3_Load(&basevec, anglex, angley, anglez); FV3_Cross(&upvec, &basevec, &upcross); FV3_Normalize(&upcross); FM_LoadIdentity(matrix); matrix->m[0] = upcross.x; matrix->m[1] = upcross.y; matrix->m[2] = upcross.z; matrix->m[3] = 0 * FRACUNIT; matrix->m[4] = upx; matrix->m[5] = upy; matrix->m[6] = upz; matrix->m[7] = 0; matrix->m[8] = anglex; matrix->m[9] = angley; matrix->m[10] = anglez; matrix->m[11] = 0; matrix->m[12] = x - FixedMul(upx, radius); matrix->m[13] = y - FixedMul(upy, radius); matrix->m[14] = z - FixedMul(upz, radius); matrix->m[15] = FRACUNIT; } // // MultMatrixVec // // Multiplies a vector by the specified matrix // const vector3_t *FM_MultMatrixVec3(const matrix_t *matrix, const vector3_t *vec, vector3_t *out) { #define M(row,col) matrix->m[col * 4 + row] out->x = FixedMul(vec->x, M(0, 0)) + FixedMul(vec->y, M(0, 1)) + FixedMul(vec->z, M(0, 2)) + M(0, 3); out->y = FixedMul(vec->x, M(1, 0)) + FixedMul(vec->y, M(1, 1)) + FixedMul(vec->z, M(1, 2)) + M(1, 3); out->z = FixedMul(vec->x, M(2, 0)) + FixedMul(vec->y, M(2, 1)) + FixedMul(vec->z, M(2, 2)) + M(2, 3); #undef M return out; } const vector4_t *FM_MultMatrixVec4(const matrix_t *matrix, const vector4_t *vec, vector4_t *out) { #define M(row,col) matrix->m[col * 4 + row] out->x = FixedMul(vec->x, M(0, 0)) + FixedMul(vec->y, M(0, 1)) + FixedMul(vec->z, M(0, 2)) + FixedMul(vec->a, M(0, 3)); out->y = FixedMul(vec->x, M(1, 0)) + FixedMul(vec->y, M(1, 1)) + FixedMul(vec->z, M(1, 2)) + FixedMul(vec->a, M(1, 3)); out->z = FixedMul(vec->x, M(2, 0)) + FixedMul(vec->y, M(2, 1)) + FixedMul(vec->z, M(2, 2)) + FixedMul(vec->a, M(2, 3)); out->a = FixedMul(vec->x, M(3, 0)) + FixedMul(vec->y, M(3, 1)) + FixedMul(vec->z, M(3, 2)) + FixedMul(vec->a, M(3, 3)); #undef M return out; } // // MultMatrix // // Multiples one matrix into another // void FM_MultMatrix(matrix_t *dest, const matrix_t *multme) { matrix_t result; UINT8 i, j; #define M(row,col) multme->m[col * 4 + row] #define D(row,col) dest->m[col * 4 + row] #define R(row,col) result.m[col * 4 + row] for (i = 0; i < 4; i++) { for (j = 0; j < 4; j++) R(i, j) = FixedMul(D(i, 0), M(0, j)) + FixedMul(D(i, 1), M(1, j)) + FixedMul(D(i, 2), M(2, j)) + FixedMul(D(i, 3), M(3, j)); } M_Memcpy(dest, &result, sizeof(matrix_t)); #undef R #undef D #undef M } // // Translate // // Translates a matrix // void FM_Translate(matrix_t *dest, fixed_t x, fixed_t y, fixed_t z) { matrix_t trans; #define M(row,col) trans.m[col * 4 + row] memset(&trans, 0x00, sizeof(matrix_t)); M(0, 0) = M(1, 1) = M(2, 2) = M(3, 3) = FRACUNIT; M(0, 3) = x; M(1, 3) = y; M(2, 3) = z; FM_MultMatrix(dest, &trans); #undef M } // // Scale // // Scales a matrix // void FM_Scale(matrix_t *dest, fixed_t x, fixed_t y, fixed_t z) { matrix_t scale; #define M(row,col) scale.m[col * 4 + row] memset(&scale, 0x00, sizeof(matrix_t)); M(3, 3) = FRACUNIT; M(0, 0) = x; M(1, 1) = y; M(2, 2) = z; FM_MultMatrix(dest, &scale); #undef M } #ifdef M_TESTCASE //#define MULDIV_TEST #define SQRT_TEST static inline void M_print(INT64 a) { const fixed_t w = (a >> FRACBITS); fixed_t f = a % FRACUNIT; fixed_t d = FRACUNIT; if (f == 0) { printf("%d", (fixed_t)w); return; } else while (f != 1 && f / 2 == f >> 1) { d /= 2; f /= 2; } if (w == 0) printf("%d/%d", (fixed_t)f, d); else printf("%d+(%d/%d)", (fixed_t)w, (fixed_t)f, d); } FUNCMATH FUNCINLINE static inline fixed_t FixedMulC(fixed_t a, fixed_t b) { return (fixed_t)((((INT64)a * b)) / FRACUNIT); } FUNCMATH FUNCINLINE static inline fixed_t FixedDivC2(fixed_t a, fixed_t b) { INT64 ret; if (b == 0) I_Error("FixedDiv: divide by zero"); ret = (((INT64)a * FRACUNIT)) / b; if ((ret > INT32_MAX) || (ret < INT32_MIN)) I_Error("FixedDiv: divide by zero"); return (fixed_t)ret; } FUNCMATH FUNCINLINE static inline fixed_t FixedDivC(fixed_t a, fixed_t b) { if ((abs(a) >> (FRACBITS - 2)) >= abs(b)) return (a^b) < 0 ? INT32_MIN : INT32_MAX; return FixedDivC2(a, b); } FUNCMATH FUNCINLINE static inline fixed_t FixedSqrtC(fixed_t x) { const float fx = FIXED_TO_FLOAT(x); float fr; #ifdef HAVE_SQRTF fr = sqrtf(fx); #else fr = (float)sqrt(fx); #endif return FLOAT_TO_FIXED(fr); } int main(int argc, char** argv) { int n = 10; INT64 a, b; fixed_t c, d; (void)argc; (void)argv; #ifdef MULDIV_TEST for (a = 1; a <= INT32_MAX; a += FRACUNIT) for (b = 0; b <= INT32_MAX; b += FRACUNIT) { c = FixedMul(a, b); d = FixedMulC(a, b); if (c != d) { printf("("); M_print(a); printf(") * ("); M_print(b); printf(") = ("); M_print(c); printf(") != ("); M_print(d); printf(") \n"); n--; printf("%d != %d\n", c, d); } c = FixedDiv(a, b); d = FixedDivC(a, b); if (c != d) { printf("("); M_print(a); printf(") / ("); M_print(b); printf(") = ("); M_print(c); printf(") != ("); M_print(d); printf(")\n"); n--; printf("%d != %d\n", c, d); } if (n <= 0) exit(-1); } #endif #ifdef SQRT_TEST for (a = 0; a <= INT32_MAX; a += 1) { c = FixedSqrt(a); d = FixedSqrtC(a); b = abs(c - d); if (b > 1) { printf("sqrt("); M_print(a); printf(") = {("); M_print(c); printf(") != ("); M_print(d); printf(")} \n"); //n--; printf("%d != %d {", c, d); M_print(b); printf("}\n"); } if (n <= 0) exit(-1); } #endif exit(0); } static void *cpu_cpy(void *dest, const void *src, size_t n) { return memcpy(dest, src, n); } void *(*M_Memcpy)(void* dest, const void* src, size_t n) = cpu_cpy; void I_Error(const char *error, ...) { (void)error; exit(-1); } #endif