dolphin/Source/Plugins/Plugin_VideoSoftware/Src/TransformUnit.cpp
Soren Jorvang 1619e176ff Use brute force to link the software plugin.
Not yet functional.


git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@7050 8ced0084-cf51-0410-be5f-012b33b47a6e
2011-02-03 19:55:30 +00:00

511 lines
16 KiB
C++

// Copyright (C) 2003-2009 Dolphin Project.
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, version 2.0.
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License 2.0 for more details.
// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/
// Official SVN repository and contact information can be found at
// http://code.google.com/p/dolphin-emu/
#include "Common.h"
#include <math.h>
#include "TransformUnit.h"
#include "XFMemLoader.h"
#include "CPMemLoader.h"
#include "BPMemLoader.h"
#include "NativeVertexFormat.h"
#include "Vec3.h"
namespace TransformUnit
{
void MultiplyVec2Mat24(const Vec3 &vec, const float *mat, Vec3 &result)
{
result.x = mat[0] * vec.x + mat[1] * vec.y + mat[2] + mat[3];
result.y = mat[4] * vec.x + mat[5] * vec.y + mat[6] + mat[7];
result.z = 1.0f;
}
void MultiplyVec2Mat34(const Vec3 &vec, const float *mat, Vec3 &result)
{
result.x = mat[0] * vec.x + mat[1] * vec.y + mat[2] + mat[3];
result.y = mat[4] * vec.x + mat[5] * vec.y + mat[6] + mat[7];
result.z = mat[8] * vec.x + mat[9] * vec.y + mat[10] + mat[11];
}
void MultiplyVec3Mat33(const Vec3 &vec, const float *mat, Vec3 &result)
{
result.x = mat[0] * vec.x + mat[1] * vec.y + mat[2] * vec.z;
result.y = mat[3] * vec.x + mat[4] * vec.y + mat[5] * vec.z;
result.z = mat[6] * vec.x + mat[7] * vec.y + mat[8] * vec.z;
}
void MultiplyVec3Mat24(const Vec3 &vec, const float *mat, Vec3 &result)
{
result.x = mat[0] * vec.x + mat[1] * vec.y + mat[2] * vec.z + mat[3];
result.y = mat[4] * vec.x + mat[5] * vec.y + mat[6] * vec.z + mat[7];
result.z = 1.0f;
}
void MultiplyVec3Mat34(const Vec3 &vec, const float *mat, Vec3 &result)
{
result.x = mat[0] * vec.x + mat[1] * vec.y + mat[2] * vec.z + mat[3];
result.y = mat[4] * vec.x + mat[5] * vec.y + mat[6] * vec.z + mat[7];
result.z = mat[8] * vec.x + mat[9] * vec.y + mat[10] * vec.z + mat[11];
}
void MultipleVec3Perspective(const Vec3 &vec, const float *proj, Vec4 &result)
{
result.x = proj[0] * vec.x + proj[1] * vec.z;
result.y = proj[2] * vec.y + proj[3] * vec.z;
//result.z = (proj[4] * vec.z + proj[5]);
result.z = (proj[4] * vec.z + proj[5]) * (1.0f - (float)1e-7);
result.w = -vec.z;
}
void MultipleVec3Ortho(const Vec3 &vec, const float *proj, Vec4 &result)
{
result.x = proj[0] * vec.x + proj[1];
result.y = proj[2] * vec.y + proj[3];
result.z = proj[4] * vec.z + proj[5];
result.w = 1;
}
void TransformPosition(const InputVertexData *src, OutputVertexData *dst)
{
const float* mat = (const float*)&swxfregs.posMatrices[src->posMtx * 4];
MultiplyVec3Mat34(src->position, mat, dst->mvPosition);
if (swxfregs.projection[6] == 0)
{
MultipleVec3Perspective(dst->mvPosition, swxfregs.projection, dst->projectedPosition);
}
else
{
MultipleVec3Ortho(dst->mvPosition, swxfregs.projection, dst->projectedPosition);
}
}
void TransformNormal(const InputVertexData *src, bool nbt, OutputVertexData *dst)
{
const float* mat = (const float*)&swxfregs.normalMatrices[(src->posMtx & 31) * 3];
if (nbt)
{
MultiplyVec3Mat33(src->normal[0], mat, dst->normal[0]);
MultiplyVec3Mat33(src->normal[1], mat, dst->normal[1]);
MultiplyVec3Mat33(src->normal[2], mat, dst->normal[2]);
dst->normal[0].normalize();
}
else
{
MultiplyVec3Mat33(src->normal[0], mat, dst->normal[0]);
dst->normal[0].normalize();
}
}
inline void TransformTexCoordRegular(const TexMtxInfo &texinfo, int coordNum, bool specialCase, const InputVertexData *srcVertex, OutputVertexData *dstVertex)
{
const Vec3 *src;
switch (texinfo.sourcerow)
{
case XF_SRCGEOM_INROW:
src = &srcVertex->position;
break;
case XF_SRCNORMAL_INROW:
src = &srcVertex->normal[0];
break;
case XF_SRCBINORMAL_T_INROW:
src = &srcVertex->normal[1];
break;
case XF_SRCBINORMAL_B_INROW:
src = &srcVertex->normal[2];
break;
default:
_assert_(texinfo.sourcerow >= XF_SRCTEX0_INROW && texinfo.sourcerow <= XF_SRCTEX7_INROW);
src = (Vec3*)srcVertex->texCoords[texinfo.sourcerow - XF_SRCTEX0_INROW];
break;
}
const float *mat = (const float*)&swxfregs.posMatrices[srcVertex->texMtx[coordNum] * 4];
Vec3 *dst = &dstVertex->texCoords[coordNum];
if (texinfo.projection == XF_TEXPROJ_ST)
{
if (texinfo.inputform == XF_TEXINPUT_AB11 || specialCase)
MultiplyVec2Mat24(*src, mat, *dst);
else
MultiplyVec3Mat24(*src, mat, *dst);
}
else // texinfo.projection == XF_TEXPROJ_STQ
{
_assert_(!specialCase);
if (texinfo.inputform == XF_TEXINPUT_AB11)
MultiplyVec2Mat34(*src, mat, *dst);
else
MultiplyVec3Mat34(*src, mat, *dst);
}
if (swxfregs.dualTexTrans)
{
Vec3 tempCoord;
// normalize
const PostMtxInfo &postInfo = swxfregs.postMtxInfo[coordNum];
const float *postMat = (const float*)&swxfregs.postMatrices[postInfo.index * 4];
if (specialCase)
{
// no normalization
// q of input is 1
// q of output is unknown
tempCoord.x = dst->x;
tempCoord.y = dst->y;
dst->x = postMat[0] * tempCoord.x + postMat[1] * tempCoord.y + postMat[2] + postMat[3];
dst->y = postMat[4] * tempCoord.x + postMat[5] * tempCoord.y + postMat[6] + postMat[7];
dst->z = 1.0f;
}
else
{
if (postInfo.normalize)
tempCoord = dst->normalized();
else
tempCoord = *dst;
MultiplyVec3Mat34(tempCoord, postMat, *dst);
}
}
}
struct LightPointer
{
u32 reserved[3];
u8 color[4];
Vec3 cosatt;
Vec3 distatt;
Vec3 pos;
Vec3 dir;
};
inline void AddIntegerColor(const u8 *src, Vec3 &dst)
{
dst.x += src[1];
dst.y += src[2];
dst.z += src[3];
}
inline void AddScaledIntegerColor(const u8 *src, float scale, Vec3 &dst)
{
dst.x += src[1] * scale;
dst.y += src[2] * scale;
dst.z += src[3] * scale;
}
inline float Clamp(float val, float a, float b)
{
return val<a?a:val>b?b:val;
}
inline float SafeDivide(float n, float d)
{
return (d==0)?(n>0?1:0):n/d;
}
void LightColor(const Vec3 &pos, const Vec3 &normal, u8 lightNum, const LitChannel &chan, Vec3 &lightCol)
{
const LightPointer *light = (const LightPointer*)&swxfregs.lights[0x10*lightNum];
if (!(chan.attnfunc & 1)) {
// atten disabled
switch (chan.diffusefunc) {
case LIGHTDIF_NONE:
AddIntegerColor(light->color, lightCol);
break;
case LIGHTDIF_SIGN:
{
Vec3 ldir = (light->pos - pos).normalized();
float diffuse = ldir * normal;
AddScaledIntegerColor(light->color, diffuse, lightCol);
}
break;
case LIGHTDIF_CLAMP:
{
Vec3 ldir = (light->pos - pos).normalized();
float diffuse = max(0.0f, ldir * normal);
AddScaledIntegerColor(light->color, diffuse, lightCol);
}
break;
default: _assert_(0);
}
}
else { // spec and spot
// not sure about divide by zero checks
Vec3 ldir = light->pos - pos;
float attn;
if (chan.attnfunc == 3) { // spot
float dist2 = ldir.length2();
float dist = sqrtf(dist2);
ldir = ldir / dist;
attn = max(0.0f, ldir * light->dir);
float cosAtt = light->cosatt.x + (light->cosatt.y * attn) + (light->cosatt.z * attn * attn);
float distAtt = light->distatt.x + (light->distatt.y * dist) + (light->distatt.z * dist2);
attn = SafeDivide(max(0.0f, cosAtt), distAtt);
}
else if (chan.attnfunc == 1) { // specular
// donko - what is going on here? 655.36 is a guess but seems about right.
attn = (light->pos * normal) > -655.36 ? max(0.0f, (light->dir * normal)) : 0;
ldir.set(1.0f, attn, attn * attn);
float cosAtt = max(0.0f, light->cosatt * ldir);
float distAtt = light->distatt * ldir;
attn = SafeDivide(max(0.0f, cosAtt), distAtt);
} else {
PanicAlert("LightColor");
return;
}
switch (chan.diffusefunc) {
case LIGHTDIF_NONE:
AddScaledIntegerColor(light->color, attn, lightCol);
break;
case LIGHTDIF_SIGN:
{
float difAttn = ldir * normal;
AddScaledIntegerColor(light->color, attn * difAttn, lightCol);
}
break;
case LIGHTDIF_CLAMP:
{
float difAttn = max(0.0f, ldir * normal);
AddScaledIntegerColor(light->color, attn * difAttn, lightCol);
}
break;
default: _assert_(0);
}
}
}
void LightAlpha(const Vec3 &pos, const Vec3 &normal, u8 lightNum, const LitChannel &chan, float &lightCol)
{
const LightPointer *light = (const LightPointer*)&swxfregs.lights[0x10*lightNum];
if (!(chan.attnfunc & 1)) {
// atten disabled
switch (chan.diffusefunc) {
case LIGHTDIF_NONE:
lightCol += light->color[0];
break;
case LIGHTDIF_SIGN:
{
Vec3 ldir = (light->pos - pos).normalized();
float diffuse = ldir * normal;
lightCol += light->color[0] * diffuse;
}
break;
case LIGHTDIF_CLAMP:
{
Vec3 ldir = (light->pos - pos).normalized();
float diffuse = max(0.0f, ldir * normal);
lightCol += light->color[0] * diffuse;
}
break;
default: _assert_(0);
}
}
else { // spec and spot
Vec3 ldir = light->pos - pos;
float attn;
if (chan.attnfunc == 3) { // spot
float dist2 = ldir.length2();
float dist = sqrtf(dist2);
ldir = ldir / dist;
attn = max(0.0f, ldir * light->dir);
float cosAtt = light->cosatt.x + (light->cosatt.y * attn) + (light->cosatt.z * attn * attn);
float distAtt = light->distatt.x + (light->distatt.y * dist) + (light->distatt.z * dist2);
attn = SafeDivide(max(0.0f, cosAtt), distAtt);
}
else /* if (chan.attnfunc == 1) */ { // specular
// donko - what is going on here? 655.36 is a guess but seems about right.
attn = (light->pos * normal) > -655.36 ? max(0.0f, (light->dir * normal)) : 0;
ldir.set(1.0f, attn, attn * attn);
float cosAtt = light->cosatt * ldir;
float distAtt = light->distatt * ldir;
attn = SafeDivide(max(0.0f, cosAtt), distAtt);
}
switch (chan.diffusefunc) {
case LIGHTDIF_NONE:
lightCol += light->color[0] * attn;
break;
case LIGHTDIF_SIGN:
{
float difAttn = ldir * normal;
lightCol += light->color[0] * attn * difAttn;
}
break;
case LIGHTDIF_CLAMP:
{
float difAttn = max(0.0f, ldir * normal);
lightCol += light->color[0] * attn * difAttn;
}
break;
default: _assert_(0);
}
}
}
void TransformColor(const InputVertexData *src, OutputVertexData *dst)
{
for (u32 chan = 0; chan < swxfregs.nNumChans; chan++)
{
// abgr
u8 matcolor[4];
u8 chancolor[4];
// color
LitChannel &colorchan = swxfregs.color[chan];
if (colorchan.matsource)
*(u32*)matcolor = *(u32*)src->color[chan]; // vertex
else
*(u32*)matcolor = swxfregs.matColor[chan];
if (colorchan.enablelighting)
{
Vec3 lightCol;
if (colorchan.ambsource)
{
// vertex
lightCol.x = src->color[chan][1];
lightCol.y = src->color[chan][2];
lightCol.z = src->color[chan][3];
}
else
{
u8 *ambColor = (u8*)&swxfregs.ambColor[chan];
lightCol.x = ambColor[1];
lightCol.y = ambColor[2];
lightCol.z = ambColor[3];
}
u8 mask = colorchan.GetFullLightMask();
for (int i = 0; i < 8; ++i) {
if (mask&(1<<i))
LightColor(dst->mvPosition, dst->normal[0], i, colorchan, lightCol);
}
float inv = 1.0f / 255.0f;
chancolor[1] = (u8)(matcolor[1] * Clamp(lightCol.x * inv, 0.0f, 1.0f));
chancolor[2] = (u8)(matcolor[2] * Clamp(lightCol.y * inv, 0.0f, 1.0f));
chancolor[3] = (u8)(matcolor[3] * Clamp(lightCol.z * inv, 0.0f, 1.0f));
}
else
{
*(u32*)chancolor = *(u32*)matcolor;
}
// alpha
LitChannel &alphachan = swxfregs.alpha[chan];
if (alphachan.matsource)
matcolor[0] = src->color[chan][0]; // vertex
else
matcolor[0] = swxfregs.matColor[chan] & 0xff;
if (swxfregs.alpha[chan].enablelighting)
{
float lightCol;
if (alphachan.ambsource)
lightCol = src->color[chan][0]; // vertex
else
lightCol = (float)(swxfregs.ambColor[chan] & 0xff);
u8 mask = alphachan.GetFullLightMask();
for (int i = 0; i < 8; ++i) {
if (mask&(1<<i))
LightAlpha(dst->mvPosition, dst->normal[0], i, alphachan, lightCol);
}
chancolor[0] = (u8)(matcolor[0] * Clamp(lightCol / 255.0f, 0.0f, 1.0f));
}
else
{
chancolor[0] = matcolor[0];
}
// abgr -> rgba
*(u32*)dst->color[chan] = Common::swap32(*(u32*)chancolor);
}
}
void TransformTexCoord(const InputVertexData *src, OutputVertexData *dst, bool specialCase)
{
for (u32 coordNum = 0; coordNum < swxfregs.numTexGens; coordNum++)
{
const TexMtxInfo &texinfo = swxfregs.texMtxInfo[coordNum];
switch (texinfo.texgentype)
{
case XF_TEXGEN_REGULAR:
TransformTexCoordRegular(texinfo, coordNum, specialCase, src, dst);
break;
case XF_TEXGEN_EMBOSS_MAP:
{
const LightPointer *light = (const LightPointer*)&swxfregs.lights[0x10*texinfo.embosslightshift];
Vec3 ldir = (light->pos - dst->mvPosition).normalized();
float d1 = ldir * dst->normal[1];
float d2 = ldir * dst->normal[2];
dst->texCoords[coordNum].x = dst->texCoords[texinfo.embosssourceshift].x + d1;
dst->texCoords[coordNum].y = dst->texCoords[texinfo.embosssourceshift].y + d2;
dst->texCoords[coordNum].z = dst->texCoords[texinfo.embosssourceshift].z;
}
break;
case XF_TEXGEN_COLOR_STRGBC0:
_assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
_assert_(texinfo.inputform == XF_TEXINPUT_AB11);
dst->texCoords[coordNum].x = (float)dst->color[0][0] / 255.0f;
dst->texCoords[coordNum].y = (float)dst->color[0][1] / 255.0f;
dst->texCoords[coordNum].z = 1.0f;
break;
case XF_TEXGEN_COLOR_STRGBC1:
_assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
_assert_(texinfo.inputform == XF_TEXINPUT_AB11);
dst->texCoords[coordNum].x = (float)dst->color[1][0] / 255.0f;
dst->texCoords[coordNum].y = (float)dst->color[1][1] / 255.0f;
dst->texCoords[coordNum].z = 1.0f;
break;
default:
ERROR_LOG(VIDEO, "Bad tex gen type %i", texinfo.texgentype);
}
}
for (u32 coordNum = 0; coordNum < swxfregs.numTexGens; coordNum++)
{
dst->texCoords[coordNum][0] *= (bpmem.texcoords[coordNum].s.scale_minus_1 + 1);
dst->texCoords[coordNum][1] *= (bpmem.texcoords[coordNum].t.scale_minus_1 + 1);
}
}
}