merge 16951:17122
[blender.git] / source / blender / render / intern / source / rayshade.c
1 /**
2  * $Id$
3  *
4  * ***** BEGIN GPL LICENSE BLOCK *****
5  *
6  * This program is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU General Public License
8  * as published by the Free Software Foundation; either version 2
9  * of the License, or (at your option) any later version. 
10  *
11  * This program is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14  * GNU General Public License for more details.
15  *
16  * You should have received a copy of the GNU General Public License
17  * along with this program; if not, write to the Free Software Foundation,
18  * Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
19  *
20  * The Original Code is Copyright (C) 1990-1998 NeoGeo BV.
21  * All rights reserved.
22  *
23  * Contributors: 2004/2005 Blender Foundation, full recode
24  *
25  * ***** END GPL LICENSE BLOCK *****
26  */
27
28 #include <math.h>
29 #include <string.h>
30 #include <stdlib.h>
31 #include <float.h>
32
33 #include "MEM_guardedalloc.h"
34
35 #include "DNA_material_types.h"
36 #include "DNA_lamp_types.h"
37
38 #include "BKE_global.h"
39 #include "BKE_node.h"
40 #include "BKE_utildefines.h"
41
42 #include "BLI_arithb.h"
43 #include "BLI_blenlib.h"
44 #include "BLI_jitter.h"
45 #include "BLI_rand.h"
46
47 #include "PIL_time.h"
48
49 #include "render_types.h"
50 #include "renderpipeline.h"
51 #include "rendercore.h"
52 #include "renderdatabase.h"
53 #include "pixelblending.h"
54 #include "pixelshading.h"
55 #include "shading.h"
56 #include "texture.h"
57
58 #include "RE_raytrace.h"
59
60 #define RAY_TRA         1
61 #define RAY_TRAFLIP     2
62
63 #define DEPTH_SHADOW_TRA  10
64
65 /* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ */
66 /* defined in pipeline.c, is hardcopy of active dynamic allocated Render */
67 /* only to be used here in this file, it's for speed */
68 extern struct Render R;
69 /* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ */
70
71 static void vlr_face_coords(RayFace *face, float **v1, float **v2, float **v3, float **v4)
72 {
73         VlakRen *vlr= (VlakRen*)face;
74
75         *v1 = (vlr->v1)? vlr->v1->co: NULL;
76         *v2 = (vlr->v2)? vlr->v2->co: NULL;
77         *v3 = (vlr->v3)? vlr->v3->co: NULL;
78         *v4 = (vlr->v4)? vlr->v4->co: NULL;
79 }
80
81 static int vlr_check_intersect(Isect *is, int ob, RayFace *face)
82 {
83         ObjectInstanceRen *obi= RAY_OBJECT_GET((Render*)is->userdata, ob);
84         VlakRen *vlr = (VlakRen*)face;
85
86         /* for baking selected to active non-traceable materials might still
87          * be in the raytree */
88         if(!(vlr->mat->mode & MA_TRACEBLE))
89                 return 0;
90
91         /* I know... cpu cycle waste, might do smarter once */
92         if(is->mode==RE_RAY_MIRROR)
93                 return !(vlr->mat->mode & MA_ONLYCAST);
94         else
95                 return (is->lay & obi->lay);
96 }
97
98 static float *vlr_get_transform(void *userdata, int i)
99 {
100         ObjectInstanceRen *obi= RAY_OBJECT_GET((Render*)userdata, i);
101
102         return (obi->flag & R_TRANSFORMED)? (float*)obi->mat: NULL;
103 }
104
105 void freeraytree(Render *re)
106 {
107         if(re->raytree) {
108                 RE_ray_tree_free(re->raytree);
109                 re->raytree= NULL;
110         }
111 }
112
113 void makeraytree(Render *re)
114 {
115         ObjectInstanceRen *obi;
116         ObjectRen *obr;
117         VlakRen *vlr= NULL;
118         float min[3], max[3], co1[3], co2[3], co3[3], co4[3];
119         double lasttime= PIL_check_seconds_timer();
120         int v, totv = 0, totface = 0;
121
122         INIT_MINMAX(min, max);
123
124         /* first min max raytree space */
125         for(obi=re->instancetable.first; obi; obi=obi->next) {
126                 obr= obi->obr;
127
128                 if(re->excludeob && obr->ob == re->excludeob)
129                         continue;
130
131                 for(v=0;v<obr->totvlak;v++) {
132                         if((v & 255)==0) vlr= obr->vlaknodes[v>>8].vlak;
133                         else vlr++;
134                         /* baking selected to active needs non-traceable too */
135                         if((re->flag & R_BAKE_TRACE) || (vlr->mat->mode & MA_TRACEBLE)) {       
136                                 if((vlr->mat->mode & MA_WIRE)==0) {     
137                                         VECCOPY(co1, vlr->v1->co);
138                                         VECCOPY(co2, vlr->v2->co);
139                                         VECCOPY(co3, vlr->v3->co);
140
141                                         if(obi->flag & R_TRANSFORMED) {
142                                                 Mat4MulVecfl(obi->mat, co1);
143                                                 Mat4MulVecfl(obi->mat, co2);
144                                                 Mat4MulVecfl(obi->mat, co3);
145                                         }
146
147                                         DO_MINMAX(co1, min, max);
148                                         DO_MINMAX(co2, min, max);
149                                         DO_MINMAX(co3, min, max);
150
151                                         if(vlr->v4) {
152                                                 VECCOPY(co4, vlr->v4->co);
153                                                 if(obi->flag & R_TRANSFORMED)
154                                                         Mat4MulVecfl(obi->mat, co4);
155                                                 DO_MINMAX(co4, min, max);
156                                         }
157
158                                         totface++;
159                                 }
160                         }
161                 }
162         }
163
164         re->raytree= RE_ray_tree_create(re->r.ocres, totface, min, max,
165                 vlr_face_coords, vlr_check_intersect, vlr_get_transform, re);
166
167         if(min[0] > max[0]) { /* empty raytree */
168                 RE_ray_tree_done(re->raytree);
169                 return; 
170         }
171
172         for(obi=re->instancetable.first; obi; obi=obi->next) {
173                 obr= obi->obr;
174
175                 if(re->excludeob && obr->ob == re->excludeob)
176                         continue;
177
178                 for(v=0; v<obr->totvlak; v++, totv++) {
179                         if((v & 255)==0) {
180                                 double time= PIL_check_seconds_timer();
181
182                                 vlr= obr->vlaknodes[v>>8].vlak;
183                                 if(re->test_break())
184                                         break;
185                                 if(time-lasttime>1.0f) {
186                                         char str[32];
187                                         sprintf(str, "Filling Octree: %d", totv);
188                                         re->i.infostr= str;
189                                         re->stats_draw(&re->i);
190                                         re->i.infostr= NULL;
191                                         lasttime= time;
192                                 }
193                         }
194                         else vlr++;
195                         
196                         if((re->flag & R_BAKE_TRACE) || (vlr->mat->mode & MA_TRACEBLE))
197                                 if((vlr->mat->mode & MA_WIRE)==0)
198                                         RE_ray_tree_add_face(re->raytree, RAY_OBJECT_SET(re, obi), vlr);
199                 }
200         }
201
202         RE_ray_tree_done(re->raytree);
203         
204         re->i.infostr= NULL;
205         re->stats_draw(&re->i);
206 }
207
208 static void shade_ray(Isect *is, ShadeInput *shi, ShadeResult *shr)
209 {
210         VlakRen *vlr= (VlakRen*)is->face;
211         ObjectInstanceRen *obi= RAY_OBJECT_GET(&R, is->ob);
212         int osatex= 0;
213         
214         /* set up view vector */
215         VECCOPY(shi->view, is->vec);
216
217         /* render co */
218         shi->co[0]= is->start[0]+is->labda*(shi->view[0]);
219         shi->co[1]= is->start[1]+is->labda*(shi->view[1]);
220         shi->co[2]= is->start[2]+is->labda*(shi->view[2]);
221         
222         Normalize(shi->view);
223
224         shi->obi= obi;
225         shi->obr= obi->obr;
226         shi->vlr= vlr;
227         shi->mat= vlr->mat;
228         memcpy(&shi->r, &shi->mat->r, 23*sizeof(float));        // note, keep this synced with render_types.h
229         shi->har= shi->mat->har;
230         
231         // Osa structs we leave unchanged now
232         SWAP(int, osatex, shi->osatex);
233         
234         shi->dxco[0]= shi->dxco[1]= shi->dxco[2]= 0.0f;
235         shi->dyco[0]= shi->dyco[1]= shi->dyco[2]= 0.0f;
236         
237         // but, set Osa stuff to zero where it can confuse texture code
238         if(shi->mat->texco & (TEXCO_NORM|TEXCO_REFL) ) {
239                 shi->dxno[0]= shi->dxno[1]= shi->dxno[2]= 0.0f;
240                 shi->dyno[0]= shi->dyno[1]= shi->dyno[2]= 0.0f;
241         }
242
243         if(vlr->v4) {
244                 if(is->isect==2) 
245                         shade_input_set_triangle_i(shi, obi, vlr, 2, 1, 3);
246                 else
247                         shade_input_set_triangle_i(shi, obi, vlr, 0, 1, 3);
248         }
249         else {
250                 shade_input_set_triangle_i(shi, obi, vlr, 0, 1, 2);
251         }
252
253         shi->u= is->u;
254         shi->v= is->v;
255         shi->dx_u= shi->dx_v= shi->dy_u= shi->dy_v=  0.0f;
256
257         shade_input_set_normals(shi);
258
259         /* point normals to viewing direction */
260         if(INPR(shi->facenor, shi->view) < 0.0f)
261                 shade_input_flip_normals(shi);
262
263         shade_input_set_shade_texco(shi);
264         
265         if(is->mode==RE_RAY_SHADOW_TRA) 
266                 if(shi->mat->nodetree && shi->mat->use_nodes) {
267                         ntreeShaderExecTree(shi->mat->nodetree, shi, shr);
268                         shi->mat= vlr->mat;             /* shi->mat is being set in nodetree */
269                 }
270                 else
271                         shade_color(shi, shr);
272         else {
273                 if(shi->mat->nodetree && shi->mat->use_nodes) {
274                         ntreeShaderExecTree(shi->mat->nodetree, shi, shr);
275                         shi->mat= vlr->mat;             /* shi->mat is being set in nodetree */
276                 }
277                 else
278                         shade_material_loop(shi, shr);
279                 
280                 /* raytrace likes to separate the spec color */
281                 VECSUB(shr->diff, shr->combined, shr->spec);
282         }       
283         
284         SWAP(int, osatex, shi->osatex);  // XXXXX!!!!
285
286 }
287
288 static int refraction(float *refract, float *n, float *view, float index)
289 {
290         float dot, fac;
291
292         VECCOPY(refract, view);
293         
294         dot= view[0]*n[0] + view[1]*n[1] + view[2]*n[2];
295
296         if(dot>0.0f) {
297                 index = 1.0f/index;
298                 fac= 1.0f - (1.0f - dot*dot)*index*index;
299                 if(fac<= 0.0f) return 0;
300                 fac= -dot*index + sqrt(fac);
301         }
302         else {
303                 fac= 1.0f - (1.0f - dot*dot)*index*index;
304                 if(fac<= 0.0f) return 0;
305                 fac= -dot*index - sqrt(fac);
306         }
307
308         refract[0]= index*view[0] + fac*n[0];
309         refract[1]= index*view[1] + fac*n[1];
310         refract[2]= index*view[2] + fac*n[2];
311
312         return 1;
313 }
314
315 /* orn = original face normal */
316 static void reflection(float *ref, float *n, float *view, float *orn)
317 {
318         float f1;
319         
320         f1= -2.0f*(n[0]*view[0]+ n[1]*view[1]+ n[2]*view[2]);
321         
322         ref[0]= (view[0]+f1*n[0]);
323         ref[1]= (view[1]+f1*n[1]);
324         ref[2]= (view[2]+f1*n[2]);
325
326         if(orn) {
327                 /* test phong normals, then we should prevent vector going to the back */
328                 f1= ref[0]*orn[0]+ ref[1]*orn[1]+ ref[2]*orn[2];
329                 if(f1>0.0f) {
330                         f1+= .01f;
331                         ref[0]-= f1*orn[0];
332                         ref[1]-= f1*orn[1];
333                         ref[2]-= f1*orn[2];
334                 }
335         }
336 }
337
338 #if 0
339 static void color_combine(float *result, float fac1, float fac2, float *col1, float *col2)
340 {
341         float col1t[3], col2t[3];
342         
343         col1t[0]= sqrt(col1[0]);
344         col1t[1]= sqrt(col1[1]);
345         col1t[2]= sqrt(col1[2]);
346         col2t[0]= sqrt(col2[0]);
347         col2t[1]= sqrt(col2[1]);
348         col2t[2]= sqrt(col2[2]);
349
350         result[0]= (fac1*col1t[0] + fac2*col2t[0]);
351         result[0]*= result[0];
352         result[1]= (fac1*col1t[1] + fac2*col2t[1]);
353         result[1]*= result[1];
354         result[2]= (fac1*col1t[2] + fac2*col2t[2]);
355         result[2]*= result[2];
356 }
357 #endif
358
359 static float shade_by_transmission(Isect *is, ShadeInput *shi, ShadeResult *shr)
360 {
361         float dx, dy, dz, d, p;
362
363         if (0 == (shi->mat->mode & (MA_RAYTRANSP|MA_ZTRA)))
364                 return -1;
365            
366         if (shi->mat->tx_limit <= 0.0f) {
367                 d= 1.0f;
368         } 
369         else {
370                 /* shi.co[] calculated by shade_ray() */
371                 dx= shi->co[0] - is->start[0];
372                 dy= shi->co[1] - is->start[1];
373                 dz= shi->co[2] - is->start[2];
374                 d= sqrt(dx*dx+dy*dy+dz*dz);
375                 if (d > shi->mat->tx_limit)
376                         d= shi->mat->tx_limit;
377
378                 p = shi->mat->tx_falloff;
379                 if(p < 0.0f) p= 0.0f;
380                 else if (p > 10.0f) p= 10.0f;
381
382                 shr->alpha *= pow(d, p);
383                 if (shr->alpha > 1.0f)
384                         shr->alpha= 1.0f;
385         }
386
387         return d;
388 }
389
390 static void ray_fadeout_endcolor(float *col, ShadeInput *origshi, ShadeInput *shi, ShadeResult *shr, Isect *isec, float *vec)
391 {
392         /* un-intersected rays get either rendered material color or sky color */
393         if (origshi->mat->fadeto_mir == MA_RAYMIR_FADETOMAT) {
394                 VECCOPY(col, shr->combined);
395         } else if (origshi->mat->fadeto_mir == MA_RAYMIR_FADETOSKY) {
396                 VECCOPY(shi->view, vec);
397                 Normalize(shi->view);
398                 
399                 shadeSkyView(col, isec->start, shi->view, NULL);
400                 shadeSunView(col, shi->view);
401         }
402 }
403
404 static void ray_fadeout(Isect *is, ShadeInput *shi, float *col, float *blendcol, float dist_mir)
405 {
406         /* if fading out, linear blend against fade color */
407         float blendfac;
408
409         blendfac = 1.0 - VecLenf(shi->co, is->start)/dist_mir;
410         
411         col[0] = col[0]*blendfac + (1.0 - blendfac)*blendcol[0];
412         col[1] = col[1]*blendfac + (1.0 - blendfac)*blendcol[1];
413         col[2] = col[2]*blendfac + (1.0 - blendfac)*blendcol[2];
414 }
415
416 /* the main recursive tracer itself */
417 static void traceray(ShadeInput *origshi, ShadeResult *origshr, short depth, float *start, float *vec, float *col, ObjectInstanceRen *obi, VlakRen *vlr, int traflag)
418 {
419         ShadeInput shi;
420         ShadeResult shr;
421         Isect isec;
422         float f, f1, fr, fg, fb;
423         float ref[3], maxsize=RE_ray_tree_max_size(R.raytree);
424         float dist_mir = origshi->mat->dist_mir;
425
426         /* Warning, This is not that nice, and possibly a bit slow for every ray,
427         however some variables were not initialized properly in, unless using shade_input_initialize(...), we need to do a memset */
428         memset(&shi, 0, sizeof(ShadeInput)); 
429         /* end warning! - Campbell */
430         
431         VECCOPY(isec.start, start);
432         if (dist_mir > 0.0) {
433                 isec.end[0]= start[0]+dist_mir*vec[0];
434                 isec.end[1]= start[1]+dist_mir*vec[1];
435                 isec.end[2]= start[2]+dist_mir*vec[2];
436         } else {
437                 isec.end[0]= start[0]+maxsize*vec[0];
438                 isec.end[1]= start[1]+maxsize*vec[1];
439                 isec.end[2]= start[2]+maxsize*vec[2];
440         }
441         isec.mode= RE_RAY_MIRROR;
442         isec.faceorig= (RayFace*)vlr;
443         isec.oborig= RAY_OBJECT_SET(&R, obi);
444
445         if(RE_ray_tree_intersect(R.raytree, &isec)) {
446                 float d= 1.0f;
447                 
448                 shi.mask= origshi->mask;
449                 shi.osatex= origshi->osatex;
450                 shi.depth= 1;                                   /* only used to indicate tracing */
451                 shi.thread= origshi->thread;
452                 //shi.sample= 0; // memset above, so dont need this
453                 shi.xs= origshi->xs;
454                 shi.ys= origshi->ys;
455                 shi.lay= origshi->lay;
456                 shi.passflag= SCE_PASS_COMBINED; /* result of tracing needs no pass info */
457                 shi.combinedflag= 0xFFFFFF;              /* ray trace does all options */
458                 //shi.do_preview= 0; // memset above, so dont need this
459                 shi.light_override= origshi->light_override;
460                 shi.mat_override= origshi->mat_override;
461                 
462                 memset(&shr, 0, sizeof(ShadeResult));
463                 
464                 shade_ray(&isec, &shi, &shr);
465                 if (traflag & RAY_TRA)
466                         d= shade_by_transmission(&isec, &shi, &shr);
467                 
468                 if(depth>0) {
469
470                         if(shi.mat->mode_l & (MA_RAYTRANSP|MA_ZTRA) && shr.alpha < 1.0f) {
471                                 float nf, f, f1, refract[3], tracol[4];
472                                 
473                                 tracol[0]= shi.r;
474                                 tracol[1]= shi.g;
475                                 tracol[2]= shi.b;
476                                 tracol[3]= col[3];      // we pass on and accumulate alpha
477                                 
478                                 if(shi.mat->mode & MA_RAYTRANSP) {
479                                         /* odd depths: use normal facing viewer, otherwise flip */
480                                         if(traflag & RAY_TRAFLIP) {
481                                                 float norm[3];
482                                                 norm[0]= - shi.vn[0];
483                                                 norm[1]= - shi.vn[1];
484                                                 norm[2]= - shi.vn[2];
485                                                 if (!refraction(refract, norm, shi.view, shi.ang))
486                                                         reflection(refract, norm, shi.view, shi.vn);
487                                         }
488                                         else {
489                                                 if (!refraction(refract, shi.vn, shi.view, shi.ang))
490                                                         reflection(refract, shi.vn, shi.view, shi.vn);
491                                         }
492                                         traflag |= RAY_TRA;
493                                         traceray(origshi, origshr, depth-1, shi.co, refract, tracol, shi.obi, shi.vlr, traflag ^ RAY_TRAFLIP);
494                                 }
495                                 else
496                                         traceray(origshi, origshr, depth-1, shi.co, shi.view, tracol, shi.obi, shi.vlr, 0);
497                                 
498                                 f= shr.alpha; f1= 1.0f-f;
499                                 nf= d * shi.mat->filter;
500                                 fr= 1.0f+ nf*(shi.r-1.0f);
501                                 fg= 1.0f+ nf*(shi.g-1.0f);
502                                 fb= 1.0f+ nf*(shi.b-1.0f);
503                                 shr.diff[0]= f*shr.diff[0] + f1*fr*tracol[0];
504                                 shr.diff[1]= f*shr.diff[1] + f1*fg*tracol[1];
505                                 shr.diff[2]= f*shr.diff[2] + f1*fb*tracol[2];
506                                 
507                                 shr.spec[0] *=f;
508                                 shr.spec[1] *=f;
509                                 shr.spec[2] *=f;
510
511                                 col[3]= f1*tracol[3] + f;
512                         }
513                         else 
514                                 col[3]= 1.0f;
515
516                         if(shi.mat->mode_l & MA_RAYMIRROR) {
517                                 f= shi.ray_mirror;
518                                 if(f!=0.0f) f*= fresnel_fac(shi.view, shi.vn, shi.mat->fresnel_mir_i, shi.mat->fresnel_mir);
519                         }
520                         else f= 0.0f;
521                         
522                         if(f!=0.0f) {
523                                 float mircol[4];
524                                 
525                                 reflection(ref, shi.vn, shi.view, NULL);                        
526                                 traceray(origshi, origshr, depth-1, shi.co, ref, mircol, shi.obi, shi.vlr, 0);
527                         
528                                 f1= 1.0f-f;
529
530                                 /* combine */
531                                 //color_combine(col, f*fr*(1.0f-shr.spec[0]), f1, col, shr.diff);
532                                 //col[0]+= shr.spec[0];
533                                 //col[1]+= shr.spec[1];
534                                 //col[2]+= shr.spec[2];
535                                 
536                                 fr= shi.mirr;
537                                 fg= shi.mirg;
538                                 fb= shi.mirb;
539                 
540                                 col[0]= f*fr*(1.0f-shr.spec[0])*mircol[0] + f1*shr.diff[0] + shr.spec[0];
541                                 col[1]= f*fg*(1.0f-shr.spec[1])*mircol[1] + f1*shr.diff[1] + shr.spec[1];
542                                 col[2]= f*fb*(1.0f-shr.spec[2])*mircol[2] + f1*shr.diff[2] + shr.spec[2];
543                         }
544                         else {
545                                 col[0]= shr.diff[0] + shr.spec[0];
546                                 col[1]= shr.diff[1] + shr.spec[1];
547                                 col[2]= shr.diff[2] + shr.spec[2];
548                         }
549                         
550                         if (dist_mir > 0.0) {
551                                 float blendcol[3];
552                                 
553                                 /* max ray distance set, but found an intersection, so fade this color
554                                  * out towards the sky/material color for a smooth transition */
555                                 ray_fadeout_endcolor(blendcol, origshi, &shi, origshr, &isec, vec);
556                                 ray_fadeout(&isec, &shi, col, blendcol, dist_mir);
557                         }
558                 }
559                 else {
560                         col[0]= shr.diff[0] + shr.spec[0];
561                         col[1]= shr.diff[1] + shr.spec[1];
562                         col[2]= shr.diff[2] + shr.spec[2];
563                 }
564                 
565         }
566         else {
567                 ray_fadeout_endcolor(col, origshi, &shi, origshr, &isec, vec);
568         }
569 }
570
571 /* **************** jitter blocks ********** */
572
573 /* calc distributed planar energy */
574
575 static void DP_energy(float *table, float *vec, int tot, float xsize, float ysize)
576 {
577         int x, y, a;
578         float *fp, force[3], result[3];
579         float dx, dy, dist, min;
580         
581         min= MIN2(xsize, ysize);
582         min*= min;
583         result[0]= result[1]= 0.0f;
584         
585         for(y= -1; y<2; y++) {
586                 dy= ysize*y;
587                 for(x= -1; x<2; x++) {
588                         dx= xsize*x;
589                         fp= table;
590                         for(a=0; a<tot; a++, fp+= 2) {
591                                 force[0]= vec[0] - fp[0]-dx;
592                                 force[1]= vec[1] - fp[1]-dy;
593                                 dist= force[0]*force[0] + force[1]*force[1];
594                                 if(dist < min && dist>0.0f) {
595                                         result[0]+= force[0]/dist;
596                                         result[1]+= force[1]/dist;
597                                 }
598                         }
599                 }
600         }
601         vec[0] += 0.1*min*result[0]/(float)tot;
602         vec[1] += 0.1*min*result[1]/(float)tot;
603         // cyclic clamping
604         vec[0]= vec[0] - xsize*floor(vec[0]/xsize + 0.5);
605         vec[1]= vec[1] - ysize*floor(vec[1]/ysize + 0.5);
606 }
607
608 // random offset of 1 in 2
609 static void jitter_plane_offset(float *jitter1, float *jitter2, int tot, float sizex, float sizey, float ofsx, float ofsy)
610 {
611         float dsizex= sizex*ofsx;
612         float dsizey= sizey*ofsy;
613         float hsizex= 0.5*sizex, hsizey= 0.5*sizey;
614         int x;
615         
616         for(x=tot; x>0; x--, jitter1+=2, jitter2+=2) {
617                 jitter2[0]= jitter1[0] + dsizex;
618                 jitter2[1]= jitter1[1] + dsizey;
619                 if(jitter2[0] > hsizex) jitter2[0]-= sizex;
620                 if(jitter2[1] > hsizey) jitter2[1]-= sizey;
621         }
622 }
623
624 /* called from convertBlenderScene.c */
625 /* we do this in advance to get consistant random, not alter the render seed, and be threadsafe */
626 void init_jitter_plane(LampRen *lar)
627 {
628         float *fp;
629         int x, iter=12, tot= lar->ray_totsamp;
630         
631         /* test if already initialized */
632         if(lar->jitter) return;
633         
634         /* at least 4, or max threads+1 tables */
635         if(BLENDER_MAX_THREADS < 4) x= 4;
636         else x= BLENDER_MAX_THREADS+1;
637         fp= lar->jitter= MEM_callocN(x*tot*2*sizeof(float), "lamp jitter tab");
638         
639         /* if 1 sample, we leave table to be zero's */
640         if(tot>1) {
641                 
642                 /* set per-lamp fixed seed */
643                 BLI_srandom(tot);
644                 
645                 /* fill table with random locations, area_size large */
646                 for(x=0; x<tot; x++, fp+=2) {
647                         fp[0]= (BLI_frand()-0.5)*lar->area_size;
648                         fp[1]= (BLI_frand()-0.5)*lar->area_sizey;
649                 }
650                 
651                 while(iter--) {
652                         fp= lar->jitter;
653                         for(x=tot; x>0; x--, fp+=2) {
654                                 DP_energy(lar->jitter, fp, tot, lar->area_size, lar->area_sizey);
655                         }
656                 }
657         }       
658         /* create the dithered tables (could just check lamp type!) */
659         jitter_plane_offset(lar->jitter, lar->jitter+2*tot, tot, lar->area_size, lar->area_sizey, 0.5f, 0.0f);
660         jitter_plane_offset(lar->jitter, lar->jitter+4*tot, tot, lar->area_size, lar->area_sizey, 0.5f, 0.5f);
661         jitter_plane_offset(lar->jitter, lar->jitter+6*tot, tot, lar->area_size, lar->area_sizey, 0.0f, 0.5f);
662 }
663
664 /* table around origin, -0.5*size to 0.5*size */
665 static float *give_jitter_plane(LampRen *lar, int thread, int xs, int ys)
666 {
667         int tot;
668         
669         tot= lar->ray_totsamp;
670                         
671         if(lar->ray_samp_type & LA_SAMP_JITTER) {
672                 /* made it threadsafe */
673                 
674                 if(lar->xold[thread]!=xs || lar->yold[thread]!=ys) {
675                         jitter_plane_offset(lar->jitter, lar->jitter+2*(thread+1)*tot, tot, lar->area_size, lar->area_sizey, BLI_thread_frand(thread), BLI_thread_frand(thread));
676                         lar->xold[thread]= xs; 
677                         lar->yold[thread]= ys;
678                 }
679                 return lar->jitter+2*(thread+1)*tot;
680         }
681         if(lar->ray_samp_type & LA_SAMP_DITHER) {
682                 return lar->jitter + 2*tot*((xs & 1)+2*(ys & 1));
683         }
684         
685         return lar->jitter;
686 }
687
688
689 /* **************** QMC sampling *************** */
690
691 static void halton_sample(double *ht_invprimes, double *ht_nums, double *v)
692 {
693         // incremental halton sequence generator, from:
694         // "Instant Radiosity", Keller A.
695         unsigned int i;
696         
697         for (i = 0; i < 2; i++)
698         {
699                 double r = fabs((1.0 - ht_nums[i]) - 1e-10);
700                 
701                 if (ht_invprimes[i] >= r)
702                 {
703                         double lasth;
704                         double h = ht_invprimes[i];
705                         
706                         do {
707                                 lasth = h;
708                                 h *= ht_invprimes[i];
709                         } while (h >= r);
710                         
711                         ht_nums[i] += ((lasth + h) - 1.0);
712                 }
713                 else
714                         ht_nums[i] += ht_invprimes[i];
715                 
716                 v[i] = (float)ht_nums[i];
717         }
718 }
719
720 /* Generate Hammersley points in [0,1)^2
721  * From Lucille renderer */
722 static void hammersley_create(double *out, int n)
723 {
724         double p, t;
725         int k, kk;
726
727         for (k = 0; k < n; k++) {
728                 t = 0;
729                 for (p = 0.5, kk = k; kk; p *= 0.5, kk >>= 1) {
730                         if (kk & 1) {           /* kk mod 2 = 1         */
731                                 t += p;
732                         }
733                 }
734         
735                 out[2 * k + 0] = (double)k / (double)n;
736                 out[2 * k + 1] = t;
737         }
738 }
739
740 static struct QMCSampler *QMC_initSampler(int type, int tot)
741 {       
742         QMCSampler *qsa = MEM_callocN(sizeof(QMCSampler), "qmc sampler");
743         qsa->samp2d = MEM_callocN(2*sizeof(double)*tot, "qmc sample table");
744
745         qsa->tot = tot;
746         qsa->type = type;
747         
748         if (qsa->type==SAMP_TYPE_HAMMERSLEY) 
749                 hammersley_create(qsa->samp2d, qsa->tot);
750                 
751         return qsa;
752 }
753
754 static void QMC_initPixel(QMCSampler *qsa, int thread)
755 {
756         if (qsa->type==SAMP_TYPE_HAMMERSLEY)
757         {
758                 /* hammersley sequence is fixed, already created in QMCSampler init.
759                  * per pixel, gets a random offset. We create separate offsets per thread, for write-safety */
760                 qsa->offs[thread][0] = 0.5 * BLI_thread_frand(thread);
761                 qsa->offs[thread][1] = 0.5 * BLI_thread_frand(thread);
762         }
763         else {  /* SAMP_TYPE_HALTON */
764                 
765                 /* generate a new randomised halton sequence per pixel
766                  * to alleviate qmc artifacts and make it reproducable 
767                  * between threads/frames */
768                 double ht_invprimes[2], ht_nums[2];
769                 double r[2];
770                 int i;
771         
772                 ht_nums[0] = BLI_thread_frand(thread);
773                 ht_nums[1] = BLI_thread_frand(thread);
774                 ht_invprimes[0] = 0.5;
775                 ht_invprimes[1] = 1.0/3.0;
776                 
777                 for (i=0; i< qsa->tot; i++) {
778                         halton_sample(ht_invprimes, ht_nums, r);
779                         qsa->samp2d[2*i+0] = r[0];
780                         qsa->samp2d[2*i+1] = r[1];
781                 }
782         }
783 }
784
785 static void QMC_freeSampler(QMCSampler *qsa)
786 {
787         MEM_freeN(qsa->samp2d);
788         MEM_freeN(qsa);
789 }
790
791 static void QMC_getSample(double *s, QMCSampler *qsa, int thread, int num)
792 {
793         if (qsa->type == SAMP_TYPE_HAMMERSLEY) {
794                 s[0] = fmod(qsa->samp2d[2*num+0] + qsa->offs[thread][0], 1.0f);
795                 s[1] = fmod(qsa->samp2d[2*num+1] + qsa->offs[thread][1], 1.0f);
796         }
797         else { /* SAMP_TYPE_HALTON */
798                 s[0] = qsa->samp2d[2*num+0];
799                 s[1] = qsa->samp2d[2*num+1];
800         }
801 }
802
803 /* phong weighted disc using 'blur' for exponent, centred on 0,0 */
804 static void QMC_samplePhong(float *vec, QMCSampler *qsa, int thread, int num, float blur)
805 {
806         double s[2];
807         float phi, pz, sqr;
808         
809         QMC_getSample(s, qsa, thread, num);
810
811         phi = s[0]*2*M_PI;
812         pz = pow(s[1], blur);
813         sqr = sqrt(1.0f-pz*pz);
814
815         vec[0] = cos(phi)*sqr;
816         vec[1] = sin(phi)*sqr;
817         vec[2] = 0.0f;
818 }
819
820 /* rect of edge lengths sizex, sizey, centred on 0.0,0.0 i.e. ranging from -sizex/2 to +sizey/2 */
821 static void QMC_sampleRect(float *vec, QMCSampler *qsa, int thread, int num, float sizex, float sizey)
822 {
823         double s[2];
824
825         QMC_getSample(s, qsa, thread, num);
826                 
827         vec[0] = (s[0] - 0.5) * sizex;
828         vec[1] = (s[1] - 0.5) * sizey;
829         vec[2] = 0.0f;
830 }
831
832 /* disc of radius 'radius', centred on 0,0 */
833 static void QMC_sampleDisc(float *vec, QMCSampler *qsa, int thread, int num, float radius)
834 {
835         double s[2];
836         float phi, sqr;
837         
838         QMC_getSample(s, qsa, thread, num);
839         
840         phi = s[0]*2*M_PI;
841         sqr = sqrt(s[1]);
842
843         vec[0] = cos(phi)*sqr* radius/2.0;
844         vec[1] = sin(phi)*sqr* radius/2.0;
845         vec[2] = 0.0f;
846 }
847
848 /* uniform hemisphere sampling */
849 static void QMC_sampleHemi(float *vec, QMCSampler *qsa, int thread, int num)
850 {
851         double s[2];
852         float phi, sqr;
853         
854         QMC_getSample(s, qsa, thread, num);
855         
856         phi = s[0]*2.f*M_PI;    
857         sqr = sqrt(s[1]);
858
859         vec[0] = cos(phi)*sqr;
860         vec[1] = sin(phi)*sqr;
861         vec[2] = 1.f - s[1]*s[1];
862 }
863
864 #if 0 /* currently not used */
865 /* cosine weighted hemisphere sampling */
866 static void QMC_sampleHemiCosine(float *vec, QMCSampler *qsa, int thread, int num)
867 {
868         double s[2];
869         float phi, sqr;
870         
871         QMC_getSample(s, qsa, thread, num);
872         
873         phi = s[0]*2.f*M_PI;    
874         sqr = s[1]*sqrt(2-s[1]*s[1]);
875
876         vec[0] = cos(phi)*sqr;
877         vec[1] = sin(phi)*sqr;
878         vec[2] = 1.f - s[1]*s[1];
879
880 }
881 #endif
882
883 /* called from convertBlenderScene.c */
884 void init_render_qmcsampler(Render *re)
885 {
886         re->qmcsamplers= MEM_callocN(sizeof(ListBase)*BLENDER_MAX_THREADS, "QMCListBase");
887 }
888
889 static QMCSampler *get_thread_qmcsampler(Render *re, int thread, int type, int tot)
890 {
891         QMCSampler *qsa;
892
893         /* create qmc samplers as needed, since recursion makes it hard to
894          * predict how many are needed */
895
896         for(qsa=re->qmcsamplers[thread].first; qsa; qsa=qsa->next) {
897                 if(qsa->type == type && qsa->tot == tot && !qsa->used) {
898                         qsa->used= 1;
899                         return qsa;
900                 }
901         }
902
903         qsa= QMC_initSampler(type, tot);
904         qsa->used= 1;
905         BLI_addtail(&re->qmcsamplers[thread], qsa);
906
907         return qsa;
908 }
909
910 static void release_thread_qmcsampler(Render *re, int thread, QMCSampler *qsa)
911 {
912         qsa->used= 0;
913 }
914
915 void free_render_qmcsampler(Render *re)
916 {
917         QMCSampler *qsa, *next;
918         int a;
919
920         if(re->qmcsamplers) {
921                 for(a=0; a<BLENDER_MAX_THREADS; a++) {
922                         for(qsa=re->qmcsamplers[a].first; qsa; qsa=next) {
923                                 next= qsa->next;
924                                 QMC_freeSampler(qsa);
925                         }
926
927                         re->qmcsamplers[a].first= re->qmcsamplers[a].last= NULL;
928                 }
929
930                 MEM_freeN(re->qmcsamplers);
931                 re->qmcsamplers= NULL;
932         }
933 }
934
935 static int adaptive_sample_variance(int samples, float *col, float *colsq, float thresh)
936 {
937         float var[3], mean[3];
938
939         /* scale threshold just to give a bit more precision in input rather than dealing with 
940          * tiny tiny numbers in the UI */
941         thresh /= 2;
942         
943         mean[0] = col[0] / (float)samples;
944         mean[1] = col[1] / (float)samples;
945         mean[2] = col[2] / (float)samples;
946
947         var[0] = (colsq[0] / (float)samples) - (mean[0]*mean[0]);
948         var[1] = (colsq[1] / (float)samples) - (mean[1]*mean[1]);
949         var[2] = (colsq[2] / (float)samples) - (mean[2]*mean[2]);
950         
951         if ((var[0] * 0.4 < thresh) && (var[1] * 0.3 < thresh) && (var[2] * 0.6 < thresh))
952                 return 1;
953         else
954                 return 0;
955 }
956
957 static int adaptive_sample_contrast_val(int samples, float prev, float val, float thresh)
958 {
959         /* if the last sample's contribution to the total value was below a small threshold
960          * (i.e. the samples taken are very similar), then taking more samples that are probably 
961          * going to be the same is wasting effort */
962         if (fabs( prev/(float)(samples-1) - val/(float)samples ) < thresh) {
963                 return 1;
964         } else
965                 return 0;
966 }
967
968 static float get_avg_speed(ShadeInput *shi)
969 {
970         float pre_x, pre_y, post_x, post_y, speedavg;
971         
972         pre_x = (shi->winspeed[0] == PASS_VECTOR_MAX)?0.0:shi->winspeed[0];
973         pre_y = (shi->winspeed[1] == PASS_VECTOR_MAX)?0.0:shi->winspeed[1];
974         post_x = (shi->winspeed[2] == PASS_VECTOR_MAX)?0.0:shi->winspeed[2];
975         post_y = (shi->winspeed[3] == PASS_VECTOR_MAX)?0.0:shi->winspeed[3];
976         
977         speedavg = (sqrt(pre_x*pre_x + pre_y*pre_y) + sqrt(post_x*post_x + post_y*post_y)) / 2.0;
978         
979         return speedavg;
980 }
981
982 /* ***************** main calls ************** */
983
984
985 static void trace_refract(float *col, ShadeInput *shi, ShadeResult *shr)
986 {
987         QMCSampler *qsa=NULL;
988         int samp_type;
989         
990         float samp3d[3], orthx[3], orthy[3];
991         float v_refract[3], v_refract_new[3];
992         float sampcol[4], colsq[4];
993         
994         float blur = pow(1.0 - shi->mat->gloss_tra, 3);
995         short max_samples = shi->mat->samp_gloss_tra;
996         float adapt_thresh = shi->mat->adapt_thresh_tra;
997         
998         int samples=0;
999         
1000         colsq[0] = colsq[1] = colsq[2] = 0.0;
1001         col[0] = col[1] = col[2] = 0.0;
1002         col[3]= shr->alpha;
1003         
1004         if (blur > 0.0) {
1005                 if (adapt_thresh != 0.0) samp_type = SAMP_TYPE_HALTON;
1006                 else samp_type = SAMP_TYPE_HAMMERSLEY;
1007                         
1008                 /* all samples are generated per pixel */
1009                 qsa = get_thread_qmcsampler(&R, shi->thread, samp_type, max_samples);
1010                 QMC_initPixel(qsa, shi->thread);
1011         } else 
1012                 max_samples = 1;
1013         
1014
1015         while (samples < max_samples) {         
1016                 refraction(v_refract, shi->vn, shi->view, shi->ang);
1017                 
1018                 if (max_samples > 1) {
1019                         /* get a quasi-random vector from a phong-weighted disc */
1020                         QMC_samplePhong(samp3d, qsa, shi->thread, samples, blur);
1021                                                 
1022                         VecOrthoBasisf(v_refract, orthx, orthy);
1023                         VecMulf(orthx, samp3d[0]);
1024                         VecMulf(orthy, samp3d[1]);
1025                                 
1026                         /* and perturb the refraction vector in it */
1027                         VecAddf(v_refract_new, v_refract, orthx);
1028                         VecAddf(v_refract_new, v_refract_new, orthy);
1029                         
1030                         Normalize(v_refract_new);
1031                 } else {
1032                         /* no blurriness, use the original normal */
1033                         VECCOPY(v_refract_new, v_refract);
1034                 }
1035         
1036                 traceray(shi, shr, shi->mat->ray_depth_tra, shi->co, v_refract_new, sampcol, shi->obi, shi->vlr, RAY_TRA|RAY_TRAFLIP);
1037         
1038                 col[0] += sampcol[0];
1039                 col[1] += sampcol[1];
1040                 col[2] += sampcol[2];
1041                 col[3] += sampcol[3];
1042                 
1043                 /* for variance calc */
1044                 colsq[0] += sampcol[0]*sampcol[0];
1045                 colsq[1] += sampcol[1]*sampcol[1];
1046                 colsq[2] += sampcol[2]*sampcol[2];
1047                 
1048                 samples++;
1049                 
1050                 /* adaptive sampling */
1051                 if (adapt_thresh < 1.0 && samples > max_samples/2) 
1052                 {
1053                         if (adaptive_sample_variance(samples, col, colsq, adapt_thresh))
1054                                 break;
1055                         
1056                         /* if the pixel so far is very dark, we can get away with less samples */
1057                         if ( (col[0] + col[1] + col[2])/3.0/(float)samples < 0.01 )
1058                                 max_samples--;
1059                 }
1060         }
1061         
1062         col[0] /= (float)samples;
1063         col[1] /= (float)samples;
1064         col[2] /= (float)samples;
1065         col[3] /= (float)samples;
1066         
1067         if (qsa)
1068                 release_thread_qmcsampler(&R, shi->thread, qsa);
1069 }
1070
1071 static void trace_reflect(float *col, ShadeInput *shi, ShadeResult *shr, float fresnelfac)
1072 {
1073         QMCSampler *qsa=NULL;
1074         int samp_type;
1075         
1076         float samp3d[3], orthx[3], orthy[3];
1077         float v_nor_new[3], v_reflect[3];
1078         float sampcol[4], colsq[4];
1079                 
1080         float blur = pow(1.0 - shi->mat->gloss_mir, 3);
1081         short max_samples = shi->mat->samp_gloss_mir;
1082         float adapt_thresh = shi->mat->adapt_thresh_mir;
1083         float aniso = 1.0 - shi->mat->aniso_gloss_mir;
1084         
1085         int samples=0;
1086         
1087         col[0] = col[1] = col[2] = 0.0;
1088         colsq[0] = colsq[1] = colsq[2] = 0.0;
1089         
1090         if (blur > 0.0) {
1091                 if (adapt_thresh != 0.0) samp_type = SAMP_TYPE_HALTON;
1092                 else samp_type = SAMP_TYPE_HAMMERSLEY;
1093                         
1094                 /* all samples are generated per pixel */
1095                 qsa = get_thread_qmcsampler(&R, shi->thread, samp_type, max_samples);
1096                 QMC_initPixel(qsa, shi->thread);
1097         } else 
1098                 max_samples = 1;
1099         
1100         while (samples < max_samples) {
1101                                 
1102                 if (max_samples > 1) {
1103                         /* get a quasi-random vector from a phong-weighted disc */
1104                         QMC_samplePhong(samp3d, qsa, shi->thread, samples, blur);
1105
1106                         /* find the normal's perpendicular plane, blurring along tangents
1107                          * if tangent shading enabled */
1108                         if (shi->mat->mode & (MA_TANGENT_V)) {
1109                                 Crossf(orthx, shi->vn, shi->tang);      // bitangent
1110                                 VECCOPY(orthy, shi->tang);
1111                                 VecMulf(orthx, samp3d[0]);
1112                                 VecMulf(orthy, samp3d[1]*aniso);
1113                         } else {
1114                                 VecOrthoBasisf(shi->vn, orthx, orthy);
1115                                 VecMulf(orthx, samp3d[0]);
1116                                 VecMulf(orthy, samp3d[1]);
1117                         }
1118
1119                         /* and perturb the normal in it */
1120                         VecAddf(v_nor_new, shi->vn, orthx);
1121                         VecAddf(v_nor_new, v_nor_new, orthy);
1122                         Normalize(v_nor_new);
1123                 } else {
1124                         /* no blurriness, use the original normal */
1125                         VECCOPY(v_nor_new, shi->vn);
1126                 }
1127                 
1128                 if((shi->vlr->flag & R_SMOOTH)) 
1129                         reflection(v_reflect, v_nor_new, shi->view, shi->facenor);
1130                 else
1131                         reflection(v_reflect, v_nor_new, shi->view, NULL);
1132                 
1133                 traceray(shi, shr, shi->mat->ray_depth, shi->co, v_reflect, sampcol, shi->obi, shi->vlr, 0);
1134
1135                 
1136                 col[0] += sampcol[0];
1137                 col[1] += sampcol[1];
1138                 col[2] += sampcol[2];
1139         
1140                 /* for variance calc */
1141                 colsq[0] += sampcol[0]*sampcol[0];
1142                 colsq[1] += sampcol[1]*sampcol[1];
1143                 colsq[2] += sampcol[2]*sampcol[2];
1144                 
1145                 samples++;
1146
1147                 /* adaptive sampling */
1148                 if (adapt_thresh > 0.0 && samples > max_samples/3) 
1149                 {
1150                         if (adaptive_sample_variance(samples, col, colsq, adapt_thresh))
1151                                 break;
1152                         
1153                         /* if the pixel so far is very dark, we can get away with less samples */
1154                         if ( (col[0] + col[1] + col[2])/3.0/(float)samples < 0.01 )
1155                                 max_samples--;
1156                 
1157                         /* reduce samples when reflection is dim due to low ray mirror blend value or fresnel factor
1158                          * and when reflection is blurry */
1159                         if (fresnelfac < 0.1 * (blur+1)) {
1160                                 max_samples--;
1161                                 
1162                                 /* even more for very dim */
1163                                 if (fresnelfac < 0.05 * (blur+1)) 
1164                                         max_samples--;
1165                         }
1166                 }
1167         }
1168         
1169         col[0] /= (float)samples;
1170         col[1] /= (float)samples;
1171         col[2] /= (float)samples;
1172         
1173         if (qsa)
1174                 release_thread_qmcsampler(&R, shi->thread, qsa);
1175 }
1176
1177 /* extern call from render loop */
1178 void ray_trace(ShadeInput *shi, ShadeResult *shr)
1179 {
1180         float i, f, f1, fr, fg, fb;
1181         float mircol[4], tracol[4];
1182         float diff[3];
1183         int do_tra, do_mir;
1184         
1185         do_tra= ((shi->mat->mode & (MA_RAYTRANSP)) && shr->alpha!=1.0f);
1186         do_mir= ((shi->mat->mode & MA_RAYMIRROR) && shi->ray_mirror!=0.0f);
1187
1188         
1189         /* raytrace mirror amd refract like to separate the spec color */
1190         if(shi->combinedflag & SCE_PASS_SPEC)
1191                 VECSUB(diff, shr->combined, shr->spec) /* no ; */
1192         else
1193                 VECCOPY(diff, shr->combined);
1194         
1195         if(do_tra) {
1196                 float olddiff[3];
1197                 
1198                 trace_refract(tracol, shi, shr);
1199                 
1200                 f= shr->alpha; f1= 1.0f-f;
1201                 fr= 1.0f+ shi->mat->filter*(shi->r-1.0f);
1202                 fg= 1.0f+ shi->mat->filter*(shi->g-1.0f);
1203                 fb= 1.0f+ shi->mat->filter*(shi->b-1.0f);
1204                 
1205                 /* for refract pass */
1206                 VECCOPY(olddiff, diff);
1207                 
1208                 diff[0]= f*diff[0] + f1*fr*tracol[0];
1209                 diff[1]= f*diff[1] + f1*fg*tracol[1];
1210                 diff[2]= f*diff[2] + f1*fb*tracol[2];
1211                 
1212                 if(shi->passflag & SCE_PASS_REFRACT)
1213                         VECSUB(shr->refr, diff, olddiff);
1214                 
1215                 if(!(shi->combinedflag & SCE_PASS_REFRACT))
1216                         VECSUB(diff, diff, shr->refr);
1217                 
1218                 shr->alpha= tracol[3];
1219         }
1220         
1221         if(do_mir) {
1222         
1223                 i= shi->ray_mirror*fresnel_fac(shi->view, shi->vn, shi->mat->fresnel_mir_i, shi->mat->fresnel_mir);
1224                 if(i!=0.0f) {
1225                 
1226                         trace_reflect(mircol, shi, shr, i);
1227                         
1228                         fr= i*shi->mirr;
1229                         fg= i*shi->mirg;
1230                         fb= i*shi->mirb;
1231
1232                         if(shi->passflag & SCE_PASS_REFLECT) {
1233                                 /* mirror pass is not blocked out with spec */
1234                                 shr->refl[0]= fr*mircol[0] - fr*diff[0];
1235                                 shr->refl[1]= fg*mircol[1] - fg*diff[1];
1236                                 shr->refl[2]= fb*mircol[2] - fb*diff[2];
1237                         }
1238                         
1239                         if(shi->combinedflag & SCE_PASS_REFLECT) {
1240                                 
1241                                 f= fr*(1.0f-shr->spec[0]);      f1= 1.0f-i;
1242                                 diff[0]= f*mircol[0] + f1*diff[0];
1243                                 
1244                                 f= fg*(1.0f-shr->spec[1]);      f1= 1.0f-i;
1245                                 diff[1]= f*mircol[1] + f1*diff[1];
1246                                 
1247                                 f= fb*(1.0f-shr->spec[2]);      f1= 1.0f-i;
1248                                 diff[2]= f*mircol[2] + f1*diff[2];
1249                         }
1250                 }
1251         }
1252         /* put back together */
1253         if(shi->combinedflag & SCE_PASS_SPEC)
1254                 VECADD(shr->combined, diff, shr->spec) /* no ; */
1255         else
1256                 VECCOPY(shr->combined, diff);
1257 }
1258
1259 /* color 'shadfac' passes through 'col' with alpha and filter */
1260 /* filter is only applied on alpha defined transparent part */
1261 static void addAlphaLight(float *shadfac, float *col, float alpha, float filter)
1262 {
1263         float fr, fg, fb;
1264         
1265         fr= 1.0f+ filter*(col[0]-1.0f);
1266         fg= 1.0f+ filter*(col[1]-1.0f);
1267         fb= 1.0f+ filter*(col[2]-1.0f);
1268         
1269         shadfac[0]= alpha*col[0] + fr*(1.0f-alpha)*shadfac[0];
1270         shadfac[1]= alpha*col[1] + fg*(1.0f-alpha)*shadfac[1];
1271         shadfac[2]= alpha*col[2] + fb*(1.0f-alpha)*shadfac[2];
1272         
1273         shadfac[3]= (1.0f-alpha)*shadfac[3];
1274 }
1275
1276 static void ray_trace_shadow_tra(Isect *is, int depth, int traflag)
1277 {
1278         /* ray to lamp, find first face that intersects, check alpha properties,
1279            if it has col[3]>0.0f  continue. so exit when alpha is full */
1280         ShadeInput shi;
1281         ShadeResult shr;
1282
1283         if(RE_ray_tree_intersect(R.raytree, is)) {
1284                 float d= 1.0f;
1285                 /* we got a face */
1286                 
1287                 /* Warning, This is not that nice, and possibly a bit slow for every ray,
1288                 however some variables were not initialized properly in, unless using shade_input_initialize(...), we need to do a memset */
1289                 memset(&shi, 0, sizeof(ShadeInput)); 
1290                 /* end warning! - Campbell */
1291                 
1292                 shi.depth= 1;                                   /* only used to indicate tracing */
1293                 shi.mask= 1;
1294                 
1295                 /*shi.osatex= 0;
1296                 shi.thread= shi.sample= 0;
1297                 shi.lay= 0;
1298                 shi.passflag= 0;
1299                 shi.combinedflag= 0;
1300                 shi.do_preview= 0;
1301                 shi.light_override= NULL;
1302                 shi.mat_override= NULL;*/
1303                 
1304                 shade_ray(is, &shi, &shr);
1305                 if (traflag & RAY_TRA)
1306                         d= shade_by_transmission(is, &shi, &shr);
1307                 
1308                 /* mix colors based on shadfac (rgb + amount of light factor) */
1309                 addAlphaLight(is->col, shr.diff, shr.alpha, d*shi.mat->filter);
1310                 
1311                 if(depth>0 && is->col[3]>0.0f) {
1312                         
1313                         /* adapt isect struct */
1314                         VECCOPY(is->start, shi.co);
1315                         is->oborig= RAY_OBJECT_SET(&R, shi.obi);
1316                         is->faceorig= (RayFace*)shi.vlr;
1317
1318                         ray_trace_shadow_tra(is, depth-1, traflag | RAY_TRA);
1319                 }
1320         }
1321 }
1322
1323 /* not used, test function for ambient occlusion (yaf: pathlight) */
1324 /* main problem; has to be called within shading loop, giving unwanted recursion */
1325 int ray_trace_shadow_rad(ShadeInput *ship, ShadeResult *shr)
1326 {
1327         static int counter=0, only_one= 0;
1328         extern float hashvectf[];
1329         Isect isec;
1330         ShadeInput shi;
1331         ShadeResult shr_t;
1332         float vec[3], accum[3], div= 0.0f, maxsize= RE_ray_tree_max_size(R.raytree);
1333         int a;
1334         
1335         if(only_one) {
1336                 return 0;
1337         }
1338         only_one= 1;
1339         
1340         accum[0]= accum[1]= accum[2]= 0.0f;
1341         isec.mode= RE_RAY_MIRROR;
1342         isec.faceorig= (RayFace*)ship->vlr;
1343         isec.oborig= RAY_OBJECT_SET(&R, ship->obi);
1344         
1345         for(a=0; a<8*8; a++) {
1346                 
1347                 counter+=3;
1348                 counter %= 768;
1349                 VECCOPY(vec, hashvectf+counter);
1350                 if(ship->vn[0]*vec[0]+ship->vn[1]*vec[1]+ship->vn[2]*vec[2]>0.0f) {
1351                         vec[0]-= vec[0];
1352                         vec[1]-= vec[1];
1353                         vec[2]-= vec[2];
1354                 }
1355                 VECCOPY(isec.start, ship->co);
1356                 isec.end[0]= isec.start[0] + maxsize*vec[0];
1357                 isec.end[1]= isec.start[1] + maxsize*vec[1];
1358                 isec.end[2]= isec.start[2] + maxsize*vec[2];
1359                 
1360                 if(RE_ray_tree_intersect(R.raytree, &isec)) {
1361                         float fac;
1362                         
1363                         /* Warning, This is not that nice, and possibly a bit slow for every ray,
1364                         however some variables were not initialized properly in, unless using shade_input_initialize(...), we need to do a memset */
1365                         memset(&shi, 0, sizeof(ShadeInput)); 
1366                         /* end warning! - Campbell */
1367                         
1368                         shade_ray(&isec, &shi, &shr_t);
1369                         fac= isec.labda*isec.labda;
1370                         fac= 1.0f;
1371                         accum[0]+= fac*(shr_t.diff[0]+shr_t.spec[0]);
1372                         accum[1]+= fac*(shr_t.diff[1]+shr_t.spec[1]);
1373                         accum[2]+= fac*(shr_t.diff[2]+shr_t.spec[2]);
1374                         div+= fac;
1375                 }
1376                 else div+= 1.0f;
1377         }
1378         
1379         if(div!=0.0f) {
1380                 shr->diff[0]+= accum[0]/div;
1381                 shr->diff[1]+= accum[1]/div;
1382                 shr->diff[2]+= accum[2]/div;
1383         }
1384         shr->alpha= 1.0f;
1385         
1386         only_one= 0;
1387         return 1;
1388 }
1389
1390 /* aolight: function to create random unit sphere vectors for total random sampling */
1391 static void RandomSpherical(float *v)
1392 {
1393         float r;
1394         v[2] = 2.f*BLI_frand()-1.f;
1395         if ((r = 1.f - v[2]*v[2])>0.f) {
1396                 float a = 6.283185307f*BLI_frand();
1397                 r = sqrt(r);
1398                 v[0] = r * cos(a);
1399                 v[1] = r * sin(a);
1400         }
1401         else v[2] = 1.f;
1402 }
1403
1404 /* calc distributed spherical energy */
1405 static void DS_energy(float *sphere, int tot, float *vec)
1406 {
1407         float *fp, fac, force[3], res[3];
1408         int a;
1409         
1410         res[0]= res[1]= res[2]= 0.0f;
1411         
1412         for(a=0, fp=sphere; a<tot; a++, fp+=3) {
1413                 VecSubf(force, vec, fp);
1414                 fac= force[0]*force[0] + force[1]*force[1] + force[2]*force[2];
1415                 if(fac!=0.0f) {
1416                         fac= 1.0f/fac;
1417                         res[0]+= fac*force[0];
1418                         res[1]+= fac*force[1];
1419                         res[2]+= fac*force[2];
1420                 }
1421         }
1422
1423         VecMulf(res, 0.5);
1424         VecAddf(vec, vec, res);
1425         Normalize(vec);
1426         
1427 }
1428
1429 /* called from convertBlenderScene.c */
1430 /* creates an equally distributed spherical sample pattern */
1431 /* and allocates threadsafe memory */
1432 void init_ao_sphere(World *wrld)
1433 {
1434         float *fp;
1435         int a, tot, iter= 16;
1436
1437         /* we make twice the amount of samples, because only a hemisphere is used */
1438         tot= 2*wrld->aosamp*wrld->aosamp;
1439         
1440         wrld->aosphere= MEM_mallocN(3*tot*sizeof(float), "AO sphere");
1441         
1442         /* fixed random */
1443         BLI_srandom(tot);
1444         
1445         /* init */
1446         fp= wrld->aosphere;
1447         for(a=0; a<tot; a++, fp+= 3) {
1448                 RandomSpherical(fp);
1449         }
1450         
1451         while(iter--) {
1452                 for(a=0, fp= wrld->aosphere; a<tot; a++, fp+= 3) {
1453                         DS_energy(wrld->aosphere, tot, fp);
1454                 }
1455         }
1456         
1457         /* tables */
1458         wrld->aotables= MEM_mallocN(BLENDER_MAX_THREADS*3*tot*sizeof(float), "AO tables");
1459 }
1460
1461 /* give per thread a table, we have to compare xs ys because of way OSA works... */
1462 static float *threadsafe_table_sphere(int test, int thread, int xs, int ys, int tot)
1463 {
1464         static int xso[BLENDER_MAX_THREADS], yso[BLENDER_MAX_THREADS];
1465         static int firsttime= 1;
1466         
1467         if(firsttime) {
1468                 memset(xso, 255, sizeof(xso));
1469                 memset(yso, 255, sizeof(yso));
1470                 firsttime= 0;
1471         }
1472         
1473         if(xs==xso[thread] && ys==yso[thread]) return R.wrld.aotables+ thread*tot*3;
1474         if(test) return NULL;
1475         xso[thread]= xs; yso[thread]= ys;
1476         return R.wrld.aotables+ thread*tot*3;
1477 }
1478
1479 static float *sphere_sampler(int type, int resol, int thread, int xs, int ys)
1480 {
1481         int tot;
1482         float *vec;
1483         
1484         tot= 2*resol*resol;
1485
1486         if (type & WO_AORNDSMP) {
1487                 float *sphere;
1488                 int a;
1489                 
1490                 // always returns table
1491                 sphere= threadsafe_table_sphere(0, thread, xs, ys, tot);
1492
1493                 /* total random sampling. NOT THREADSAFE! (should be removed, is not useful) */
1494                 vec= sphere;
1495                 for (a=0; a<tot; a++, vec+=3) {
1496                         RandomSpherical(vec);
1497                 }
1498                 
1499                 return sphere;
1500         } 
1501         else {
1502                 float *sphere;
1503                 float cosfi, sinfi, cost, sint;
1504                 float ang, *vec1;
1505                 int a;
1506                 
1507                 // returns table if xs and ys were equal to last call
1508                 sphere= threadsafe_table_sphere(1, thread, xs, ys, tot);
1509                 if(sphere==NULL) {
1510                         sphere= threadsafe_table_sphere(0, thread, xs, ys, tot);
1511                         
1512                         // random rotation
1513                         ang= BLI_thread_frand(thread);
1514                         sinfi= sin(ang); cosfi= cos(ang);
1515                         ang= BLI_thread_frand(thread);
1516                         sint= sin(ang); cost= cos(ang);
1517                         
1518                         vec= R.wrld.aosphere;
1519                         vec1= sphere;
1520                         for (a=0; a<tot; a++, vec+=3, vec1+=3) {
1521                                 vec1[0]= cost*cosfi*vec[0] - sinfi*vec[1] + sint*cosfi*vec[2];
1522                                 vec1[1]= cost*sinfi*vec[0] + cosfi*vec[1] + sint*sinfi*vec[2];
1523                                 vec1[2]= -sint*vec[0] + cost*vec[2];                    
1524                         }
1525                 }
1526                 return sphere;
1527         }
1528 }
1529
1530 static void ray_ao_qmc(ShadeInput *shi, float *shadfac)
1531 {
1532         Isect isec;
1533         QMCSampler *qsa=NULL;
1534         float samp3d[3];
1535         float up[3], side[3], dir[3], nrm[3];
1536         
1537         float maxdist = R.wrld.aodist;
1538         float fac=0.0f, prev=0.0f;
1539         float adapt_thresh = G.scene->world->ao_adapt_thresh;
1540         float adapt_speed_fac = G.scene->world->ao_adapt_speed_fac;
1541         
1542         int samples=0;
1543         int max_samples = R.wrld.aosamp*R.wrld.aosamp;
1544         
1545         float dxyview[3], skyadded=0, div;
1546         int aocolor;
1547         
1548         isec.faceorig= (RayFace*)shi->vlr;
1549         isec.oborig= RAY_OBJECT_SET(&R, shi->obi);
1550         isec.face_last= NULL;
1551         isec.ob_last= 0;
1552         isec.mode= (R.wrld.aomode & WO_AODIST)?RE_RAY_SHADOW_TRA:RE_RAY_SHADOW;
1553         isec.lay= -1;
1554         
1555         shadfac[0]= shadfac[1]= shadfac[2]= 0.0f;
1556         
1557         /* prevent sky colors to be added for only shadow (shadow becomes alpha) */
1558         aocolor= R.wrld.aocolor;
1559         if(shi->mat->mode & MA_ONLYSHADOW)
1560                 aocolor= WO_AOPLAIN;
1561         
1562         if(aocolor == WO_AOSKYTEX) {
1563                 dxyview[0]= 1.0f/(float)R.wrld.aosamp;
1564                 dxyview[1]= 1.0f/(float)R.wrld.aosamp;
1565                 dxyview[2]= 0.0f;
1566         }
1567         
1568         if(shi->vlr->flag & R_SMOOTH) {
1569                 VECCOPY(nrm, shi->vn);
1570         }
1571         else {
1572                 VECCOPY(nrm, shi->facenor);
1573         }
1574         
1575         VecOrthoBasisf(nrm, up, side);
1576         
1577         /* sampling init */
1578         if (R.wrld.ao_samp_method==WO_AOSAMP_HALTON) {
1579                 float speedfac;
1580                 
1581                 speedfac = get_avg_speed(shi) * adapt_speed_fac;
1582                 CLAMP(speedfac, 1.0, 1000.0);
1583                 max_samples /= speedfac;
1584                 if (max_samples < 5) max_samples = 5;
1585                 
1586                 qsa = get_thread_qmcsampler(&R, shi->thread, SAMP_TYPE_HALTON, max_samples);
1587         } else if (R.wrld.ao_samp_method==WO_AOSAMP_HAMMERSLEY)
1588                 qsa = get_thread_qmcsampler(&R, shi->thread, SAMP_TYPE_HAMMERSLEY, max_samples);
1589
1590         QMC_initPixel(qsa, shi->thread);
1591         
1592         while (samples < max_samples) {
1593
1594                 /* sampling, returns quasi-random vector in unit hemisphere */
1595                 QMC_sampleHemi(samp3d, qsa, shi->thread, samples);
1596
1597                 dir[0] = (samp3d[0]*up[0] + samp3d[1]*side[0] + samp3d[2]*nrm[0]);
1598                 dir[1] = (samp3d[0]*up[1] + samp3d[1]*side[1] + samp3d[2]*nrm[1]);
1599                 dir[2] = (samp3d[0]*up[2] + samp3d[1]*side[2] + samp3d[2]*nrm[2]);
1600                 
1601                 Normalize(dir);
1602                         
1603                 VECCOPY(isec.start, shi->co);
1604                 isec.end[0] = shi->co[0] - maxdist*dir[0];
1605                 isec.end[1] = shi->co[1] - maxdist*dir[1];
1606                 isec.end[2] = shi->co[2] - maxdist*dir[2];
1607                 
1608                 prev = fac;
1609                 
1610                 if(RE_ray_tree_intersect(R.raytree, &isec)) {
1611                         if (R.wrld.aomode & WO_AODIST) fac+= exp(-isec.labda*R.wrld.aodistfac); 
1612                         else fac+= 1.0f;
1613                 }
1614                 else if(aocolor!=WO_AOPLAIN) {
1615                         float skycol[4];
1616                         float skyfac, view[3];
1617                         
1618                         view[0]= -dir[0];
1619                         view[1]= -dir[1];
1620                         view[2]= -dir[2];
1621                         Normalize(view);
1622                         
1623                         if(aocolor==WO_AOSKYCOL) {
1624                                 skyfac= 0.5*(1.0f+view[0]*R.grvec[0]+ view[1]*R.grvec[1]+ view[2]*R.grvec[2]);
1625                                 shadfac[0]+= (1.0f-skyfac)*R.wrld.horr + skyfac*R.wrld.zenr;
1626                                 shadfac[1]+= (1.0f-skyfac)*R.wrld.horg + skyfac*R.wrld.zeng;
1627                                 shadfac[2]+= (1.0f-skyfac)*R.wrld.horb + skyfac*R.wrld.zenb;
1628                         }
1629                         else {  /* WO_AOSKYTEX */
1630                                 shadeSkyView(skycol, isec.start, view, dxyview);
1631                                 shadeSunView(skycol, shi->view);
1632                                 shadfac[0]+= skycol[0];
1633                                 shadfac[1]+= skycol[1];
1634                                 shadfac[2]+= skycol[2];
1635                         }
1636                         skyadded++;
1637                 }
1638                 
1639                 samples++;
1640                 
1641                 if (qsa->type == SAMP_TYPE_HALTON) {
1642                         /* adaptive sampling - consider samples below threshold as in shadow (or vice versa) and exit early */          
1643                         if (adapt_thresh > 0.0 && (samples > max_samples/2) ) {
1644                                 
1645                                 if (adaptive_sample_contrast_val(samples, prev, fac, adapt_thresh)) {
1646                                         break;
1647                                 }
1648                         }
1649                 }
1650         }
1651         
1652         if(aocolor!=WO_AOPLAIN && skyadded) {
1653                 div= (1.0f - fac/(float)samples)/((float)skyadded);
1654                 
1655                 shadfac[0]*= div;       // average color times distances/hits formula
1656                 shadfac[1]*= div;       // average color times distances/hits formula
1657                 shadfac[2]*= div;       // average color times distances/hits formula
1658         } else {
1659                 shadfac[0]= shadfac[1]= shadfac[2]= 1.0f - fac/(float)samples;
1660         }
1661         
1662         if (qsa)
1663                 release_thread_qmcsampler(&R, shi->thread, qsa);
1664 }
1665
1666 /* extern call from shade_lamp_loop, ambient occlusion calculus */
1667 static void ray_ao_spheresamp(ShadeInput *shi, float *shadfac)
1668 {
1669         Isect isec;
1670         float *vec, *nrm, div, bias, sh=0.0f;
1671         float maxdist = R.wrld.aodist;
1672         float dxyview[3];
1673         int j= -1, tot, actual=0, skyadded=0, aocolor, resol= R.wrld.aosamp;
1674         
1675         isec.faceorig= (RayFace*)shi->vlr;
1676         isec.oborig= RAY_OBJECT_SET(&R, shi->obi);
1677         isec.face_last= NULL;
1678         isec.ob_last= 0;
1679         isec.mode= (R.wrld.aomode & WO_AODIST)?RE_RAY_SHADOW_TRA:RE_RAY_SHADOW;
1680         isec.lay= -1;
1681
1682
1683         shadfac[0]= shadfac[1]= shadfac[2]= 0.0f;
1684
1685         /* bias prevents smoothed faces to appear flat */
1686         if(shi->vlr->flag & R_SMOOTH) {
1687                 bias= G.scene->world->aobias;
1688                 nrm= shi->vn;
1689         }
1690         else {
1691                 bias= 0.0f;
1692                 nrm= shi->facenor;
1693         }
1694
1695         /* prevent sky colors to be added for only shadow (shadow becomes alpha) */
1696         aocolor= R.wrld.aocolor;
1697         if(shi->mat->mode & MA_ONLYSHADOW)
1698                 aocolor= WO_AOPLAIN;
1699         
1700         if(resol>32) resol= 32;
1701         
1702         vec= sphere_sampler(R.wrld.aomode, resol, shi->thread, shi->xs, shi->ys);
1703         
1704         // warning: since we use full sphere now, and dotproduct is below, we do twice as much
1705         tot= 2*resol*resol;
1706
1707         if(aocolor == WO_AOSKYTEX) {
1708                 dxyview[0]= 1.0f/(float)resol;
1709                 dxyview[1]= 1.0f/(float)resol;
1710                 dxyview[2]= 0.0f;
1711         }
1712         
1713         while(tot--) {
1714                 
1715                 if ((vec[0]*nrm[0] + vec[1]*nrm[1] + vec[2]*nrm[2]) > bias) {
1716                         /* only ao samples for mask */
1717                         if(R.r.mode & R_OSA) {
1718                                 j++;
1719                                 if(j==R.osa) j= 0;
1720                                 if(!(shi->mask & (1<<j))) {
1721                                         vec+=3;
1722                                         continue;
1723                                 }
1724                         }
1725                         
1726                         actual++;
1727                         
1728                         /* always set start/end, RE_ray_tree_intersect clips it */
1729                         VECCOPY(isec.start, shi->co);
1730                         isec.end[0] = shi->co[0] - maxdist*vec[0];
1731                         isec.end[1] = shi->co[1] - maxdist*vec[1];
1732                         isec.end[2] = shi->co[2] - maxdist*vec[2];
1733                         
1734                         /* do the trace */
1735                         if(RE_ray_tree_intersect(R.raytree, &isec)) {
1736                                 if (R.wrld.aomode & WO_AODIST) sh+= exp(-isec.labda*R.wrld.aodistfac); 
1737                                 else sh+= 1.0f;
1738                         }
1739                         else if(aocolor!=WO_AOPLAIN) {
1740                                 float skycol[4];
1741                                 float fac, view[3];
1742                                 
1743                                 view[0]= -vec[0];
1744                                 view[1]= -vec[1];
1745                                 view[2]= -vec[2];
1746                                 Normalize(view);
1747                                 
1748                                 if(aocolor==WO_AOSKYCOL) {
1749                                         fac= 0.5*(1.0f+view[0]*R.grvec[0]+ view[1]*R.grvec[1]+ view[2]*R.grvec[2]);
1750                                         shadfac[0]+= (1.0f-fac)*R.wrld.horr + fac*R.wrld.zenr;
1751                                         shadfac[1]+= (1.0f-fac)*R.wrld.horg + fac*R.wrld.zeng;
1752                                         shadfac[2]+= (1.0f-fac)*R.wrld.horb + fac*R.wrld.zenb;
1753                                 }
1754                                 else {  /* WO_AOSKYTEX */
1755                                         shadeSkyView(skycol, isec.start, view, dxyview);
1756                                         shadeSunView(skycol, shi->view);
1757                                         shadfac[0]+= skycol[0];
1758                                         shadfac[1]+= skycol[1];
1759                                         shadfac[2]+= skycol[2];
1760                                 }
1761                                 skyadded++;
1762                         }
1763                 }
1764                 // samples
1765                 vec+= 3;
1766         }
1767         
1768         if(actual==0) sh= 1.0f;
1769         else sh = 1.0f - sh/((float)actual);
1770         
1771         if(aocolor!=WO_AOPLAIN && skyadded) {
1772                 div= sh/((float)skyadded);
1773                 
1774                 shadfac[0]*= div;       // average color times distances/hits formula
1775                 shadfac[1]*= div;       // average color times distances/hits formula
1776                 shadfac[2]*= div;       // average color times distances/hits formula
1777         }
1778         else {
1779                 shadfac[0]= shadfac[1]= shadfac[2]= sh;
1780         }
1781 }
1782
1783 void ray_ao(ShadeInput *shi, float *shadfac)
1784 {
1785         /* Unfortunately, the unusual way that the sphere sampler calculates roughly twice as many
1786          * samples as are actually traced, and skips them based on bias and OSA settings makes it very difficult
1787          * to reuse code between these two functions. This is the easiest way I can think of to do it
1788          * --broken */
1789         if (ELEM(R.wrld.ao_samp_method, WO_AOSAMP_HAMMERSLEY, WO_AOSAMP_HALTON))
1790                 ray_ao_qmc(shi, shadfac);
1791         else if (R.wrld.ao_samp_method == WO_AOSAMP_CONSTANT)
1792                 ray_ao_spheresamp(shi, shadfac);
1793 }
1794
1795 static void ray_shadow_jittered_coords(ShadeInput *shi, int max, float jitco[RE_MAX_OSA][3], int *totjitco)
1796 {
1797         /* magic numbers for reordering sample positions to give better
1798          * results with adaptive sample, when it usually only takes 4 samples */
1799         int order8[8] = {0, 1, 5, 6, 2, 3, 4, 7};
1800         int order11[11] = {1, 3, 8, 10, 0, 2, 4, 5, 6, 7, 9};
1801         int order16[16] = {1, 3, 9, 12, 0, 6, 7, 8, 13, 2, 4, 5, 10, 11, 14, 15};
1802         int count = count_mask(shi->mask);
1803
1804         /* for better antialising shadow samples are distributed over the subpixel
1805          * sample coordinates, this only works for raytracing depth 0 though */
1806         if(!shi->strand && shi->depth == 0 && count > 1 && count <= max) {
1807                 float xs, ys, zs, view[3];
1808                 int samp, ordsamp, tot= 0;
1809
1810                 for(samp=0; samp<R.osa; samp++) {
1811                         if(R.osa == 8) ordsamp = order8[samp];
1812                         else if(R.osa == 11) ordsamp = order11[samp];
1813                         else if(R.osa == 16) ordsamp = order16[samp];
1814                         else ordsamp = samp;
1815
1816                         if(shi->mask & (1<<ordsamp)) {
1817                                 /* zbuffer has this inverse corrected, ensures xs,ys are inside pixel */
1818                                 xs= (float)shi->scanco[0] + R.jit[ordsamp][0] + 0.5f;
1819                                 ys= (float)shi->scanco[1] + R.jit[ordsamp][1] + 0.5f;
1820                                 zs= shi->scanco[2];
1821
1822                                 shade_input_calc_viewco(shi, xs, ys, zs, view, NULL, jitco[tot], NULL, NULL);
1823                                 tot++;
1824                         }
1825                 }
1826
1827                 *totjitco= tot;
1828         }
1829         else {
1830                 VECCOPY(jitco[0], shi->co);
1831                 *totjitco= 1;
1832         }
1833 }
1834
1835 static void ray_shadow_qmc(ShadeInput *shi, LampRen *lar, float *lampco, float *shadfac, Isect *isec)
1836 {
1837         QMCSampler *qsa=NULL;
1838         int samples=0;
1839         float samp3d[3];
1840
1841         float fac=0.0f, vec[3];
1842         float colsq[4];
1843         float adapt_thresh = lar->adapt_thresh;
1844         int min_adapt_samples=4, max_samples = lar->ray_totsamp;
1845         float *co;
1846         int do_soft=1, full_osa=0;
1847
1848         float jitco[RE_MAX_OSA][3];
1849         int totjitco;
1850
1851         colsq[0] = colsq[1] = colsq[2] = 0.0;
1852         if(isec->mode==RE_RAY_SHADOW_TRA) {
1853                 shadfac[0]= shadfac[1]= shadfac[2]= shadfac[3]= 0.0f;
1854         } else
1855                 shadfac[3]= 1.0f;
1856         
1857         if (lar->ray_totsamp < 2) do_soft = 0;
1858         if ((R.r.mode & R_OSA) && (R.osa > 0) && (shi->vlr->flag & R_FULL_OSA)) full_osa = 1;
1859         
1860         if (full_osa) {
1861                 if (do_soft) max_samples  = max_samples/R.osa + 1;
1862                 else max_samples = 1;
1863         } else {
1864                 if (do_soft) max_samples = lar->ray_totsamp;
1865                 else max_samples = (R.osa > 4)?R.osa:5;
1866         }
1867         
1868         ray_shadow_jittered_coords(shi, max_samples, jitco, &totjitco);
1869
1870         /* sampling init */
1871         if (lar->ray_samp_method==LA_SAMP_HALTON)
1872                 qsa = get_thread_qmcsampler(&R, shi->thread, SAMP_TYPE_HALTON, max_samples);
1873         else if (lar->ray_samp_method==LA_SAMP_HAMMERSLEY)
1874                 qsa = get_thread_qmcsampler(&R, shi->thread, SAMP_TYPE_HAMMERSLEY, max_samples);
1875         
1876         QMC_initPixel(qsa, shi->thread);
1877         
1878         VECCOPY(vec, lampco);
1879         
1880         
1881         while (samples < max_samples) {
1882                 isec->faceorig= (RayFace*)shi->vlr;
1883                 isec->oborig= RAY_OBJECT_SET(&R, shi->obi);
1884
1885                 /* manually jitter the start shading co-ord per sample
1886                  * based on the pre-generated OSA texture sampling offsets, 
1887                  * for anti-aliasing sharp shadow edges. */
1888                 co = jitco[samples % totjitco];
1889
1890                 if (do_soft) {
1891                         /* sphere shadow source */
1892                         if (lar->type == LA_LOCAL) {
1893                                 float ru[3], rv[3], v[3], s[3];
1894                                 
1895                                 /* calc tangent plane vectors */
1896                                 v[0] = co[0] - lampco[0];
1897                                 v[1] = co[1] - lampco[1];
1898                                 v[2] = co[2] - lampco[2];
1899                                 Normalize(v);
1900                                 VecOrthoBasisf(v, ru, rv);
1901                                 
1902                                 /* sampling, returns quasi-random vector in area_size disc */
1903                                 QMC_sampleDisc(samp3d, qsa, shi->thread, samples,lar->area_size);
1904
1905                                 /* distribute disc samples across the tangent plane */
1906                                 s[0] = samp3d[0]*ru[0] + samp3d[1]*rv[0];
1907                                 s[1] = samp3d[0]*ru[1] + samp3d[1]*rv[1];
1908                                 s[2] = samp3d[0]*ru[2] + samp3d[1]*rv[2];
1909                                 
1910                                 VECCOPY(samp3d, s);
1911                         }
1912                         else {
1913                                 /* sampling, returns quasi-random vector in [sizex,sizey]^2 plane */
1914                                 QMC_sampleRect(samp3d, qsa, shi->thread, samples, lar->area_size, lar->area_sizey);
1915                                                                 
1916                                 /* align samples to lamp vector */
1917                                 Mat3MulVecfl(lar->mat, samp3d);
1918                         }
1919                         isec->end[0]= vec[0]+samp3d[0];
1920                         isec->end[1]= vec[1]+samp3d[1];
1921                         isec->end[2]= vec[2]+samp3d[2];
1922                 } else {
1923                         VECCOPY(isec->end, vec);
1924                 }
1925
1926                 if(shi->strand) {
1927                         /* bias away somewhat to avoid self intersection */
1928                         float jitbias= 0.5f*(VecLength(shi->dxco) + VecLength(shi->dyco));
1929                         float v[3];
1930
1931                         VECSUB(v, co, isec->end);
1932                         Normalize(v);
1933
1934                         co[0] -= jitbias*v[0];
1935                         co[1] -= jitbias*v[1];
1936                         co[2] -= jitbias*v[2];
1937                 }
1938
1939                 VECCOPY(isec->start, co);
1940                 
1941                 /* trace the ray */
1942                 if(isec->mode==RE_RAY_SHADOW_TRA) {
1943                         isec->col[0]= isec->col[1]= isec->col[2]=  1.0f;
1944                         isec->col[3]= 1.0f;
1945                         
1946                         ray_trace_shadow_tra(isec, DEPTH_SHADOW_TRA, 0);
1947                         shadfac[0] += isec->col[0];
1948                         shadfac[1] += isec->col[1];
1949                         shadfac[2] += isec->col[2];
1950                         shadfac[3] += isec->col[3];
1951                         
1952                         /* for variance calc */
1953                         colsq[0] += isec->col[0]*isec->col[0];
1954                         colsq[1] += isec->col[1]*isec->col[1];
1955                         colsq[2] += isec->col[2]*isec->col[2];
1956                 }
1957                 else {
1958                         if( RE_ray_tree_intersect(R.raytree, isec) ) fac+= 1.0f;
1959                 }
1960                 
1961                 samples++;
1962                 
1963                 if ((lar->ray_samp_method == LA_SAMP_HALTON)) {
1964                 
1965                         /* adaptive sampling - consider samples below threshold as in shadow (or vice versa) and exit early */
1966                         if ((max_samples > min_adapt_samples) && (adapt_thresh > 0.0) && (samples > max_samples / 3)) {
1967                                 if (isec->mode==RE_RAY_SHADOW_TRA) {
1968                                         if ((shadfac[3] / samples > (1.0-adapt_thresh)) || (shadfac[3] / samples < adapt_thresh))
1969                                                 break;
1970                                         else if (adaptive_sample_variance(samples, shadfac, colsq, adapt_thresh))
1971                                                 break;
1972                                 } else {
1973                                         if ((fac / samples > (1.0-adapt_thresh)) || (fac / samples < adapt_thresh))
1974                                                 break;
1975                                 }
1976                         }
1977                 }
1978         }
1979         
1980         if(isec->mode==RE_RAY_SHADOW_TRA) {
1981                 shadfac[0] /= samples;
1982                 shadfac[1] /= samples;
1983                 shadfac[2] /= samples;
1984                 shadfac[3] /= samples;
1985         } else
1986                 shadfac[3]= 1.0f-fac/samples;
1987
1988         if (qsa)
1989                 release_thread_qmcsampler(&R, shi->thread, qsa);
1990 }
1991
1992 static void ray_shadow_jitter(ShadeInput *shi, LampRen *lar, float *lampco, float *shadfac, Isect *isec)
1993 {
1994         /* area soft shadow */
1995         float *jitlamp;
1996         float fac=0.0f, div=0.0f, vec[3];
1997         int a, j= -1, mask;
1998         
1999         if(isec->mode==RE_RAY_SHADOW_TRA) {
2000                 shadfac[0]= shadfac[1]= shadfac[2]= shadfac[3]= 0.0f;
2001         }
2002         else shadfac[3]= 1.0f;
2003         
2004         fac= 0.0f;
2005         jitlamp= give_jitter_plane(lar, shi->thread, shi->xs, shi->ys);
2006
2007         a= lar->ray_totsamp;
2008         
2009         /* this correction to make sure we always take at least 1 sample */
2010         mask= shi->mask;
2011         if(a==4) mask |= (mask>>4)|(mask>>8);
2012         else if(a==9) mask |= (mask>>9);
2013         
2014         while(a--) {
2015                 
2016                 if(R.r.mode & R_OSA) {
2017                         j++;
2018                         if(j>=R.osa) j= 0;
2019                         if(!(mask & (1<<j))) {
2020                                 jitlamp+= 2;
2021                                 continue;
2022                         }
2023                 }
2024                 
2025                 isec->faceorig= (RayFace*)shi->vlr;
2026                 isec->oborig= RAY_OBJECT_SET(&R, shi->obi);
2027                 
2028                 vec[0]= jitlamp[0];
2029                 vec[1]= jitlamp[1];
2030                 vec[2]= 0.0f;
2031                 Mat3MulVecfl(lar->mat, vec);
2032                 
2033                 /* set start and end, RE_ray_tree_intersect clips it */
2034                 VECCOPY(isec->start, shi->co);
2035                 isec->end[0]= lampco[0]+vec[0];
2036                 isec->end[1]= lampco[1]+vec[1];
2037                 isec->end[2]= lampco[2]+vec[2];
2038                 
2039                 if(isec->mode==RE_RAY_SHADOW_TRA) {
2040                         /* isec.col is like shadfac, so defines amount of light (0.0 is full shadow) */
2041                         isec->col[0]= isec->col[1]= isec->col[2]=  1.0f;
2042                         isec->col[3]= 1.0f;
2043                         
2044                         ray_trace_shadow_tra(isec, DEPTH_SHADOW_TRA, 0);
2045                         shadfac[0] += isec->col[0];
2046                         shadfac[1] += isec->col[1];
2047                         shadfac[2] += isec->col[2];
2048                         shadfac[3] += isec->col[3];
2049                 }
2050                 else if( RE_ray_tree_intersect(R.raytree, isec) ) fac+= 1.0f;
2051                 
2052                 div+= 1.0f;
2053                 jitlamp+= 2;
2054         }
2055         
2056         if(isec->mode==RE_RAY_SHADOW_TRA) {
2057                 shadfac[0] /= div;
2058                 shadfac[1] /= div;
2059                 shadfac[2] /= div;
2060                 shadfac[3] /= div;
2061         }
2062         else {
2063                 // sqrt makes nice umbra effect
2064                 if(lar->ray_samp_type & LA_SAMP_UMBRA)
2065                         shadfac[3]= sqrt(1.0f-fac/div);
2066                 else
2067                         shadfac[3]= 1.0f-fac/div;
2068         }
2069 }
2070 /* extern call from shade_lamp_loop */
2071 void ray_shadow(ShadeInput *shi, LampRen *lar, float *shadfac)
2072 {
2073         Isect isec;
2074         float lampco[3], maxsize;
2075
2076         /* setup isec */
2077         if(shi->mat->mode & MA_SHADOW_TRA) isec.mode= RE_RAY_SHADOW_TRA;
2078         else isec.mode= RE_RAY_SHADOW;
2079         
2080         if(lar->mode & (LA_LAYER|LA_LAYER_SHADOW))
2081                 isec.lay= lar->lay;
2082         else
2083                 isec.lay= -1;
2084
2085         /* only when not mir tracing, first hit optimm */
2086         if(shi->depth==0) {
2087                 isec.face_last= (RayFace*)lar->vlr_last[shi->thread];
2088                 isec.ob_last= RAY_OBJECT_SET(&R, lar->obi_last[shi->thread]);
2089         }
2090         else {
2091                 isec.face_last= NULL;
2092                 isec.ob_last= 0;
2093         }
2094         
2095         if(lar->type==LA_SUN || lar->type==LA_HEMI) {
2096                 maxsize= RE_ray_tree_max_size(R.raytree);
2097                 lampco[0]= shi->co[0] - maxsize*lar->vec[0];
2098                 lampco[1]= shi->co[1] - maxsize*lar->vec[1];
2099                 lampco[2]= shi->co[2] - maxsize*lar->vec[2];
2100         }
2101         else {
2102                 VECCOPY(lampco, lar->co);
2103         }
2104         
2105         if (ELEM(lar->ray_samp_method, LA_SAMP_HALTON, LA_SAMP_HAMMERSLEY)) {
2106                 
2107                 ray_shadow_qmc(shi, lar, lampco, shadfac, &isec);
2108                 
2109         } else {
2110                 if(lar->ray_totsamp<2) {
2111                         
2112                         isec.faceorig= (RayFace*)shi->vlr;
2113                         isec.oborig= RAY_OBJECT_SET(&R, shi->obi);
2114                         shadfac[3]= 1.0f; // 1.0=full light
2115                         
2116                         /* set up isec vec */
2117                         VECCOPY(isec.start, shi->co);
2118                         VECCOPY(isec.end, lampco);
2119
2120                         if(isec.mode==RE_RAY_SHADOW_TRA) {
2121                                 /* isec.col is like shadfac, so defines amount of light (0.0 is full shadow) */
2122                                 isec.col[0]= isec.col[1]= isec.col[2]=  1.0f;
2123                                 isec.col[3]= 1.0f;
2124
2125                                 ray_trace_shadow_tra(&isec, DEPTH_SHADOW_TRA, 0);
2126                                 QUATCOPY(shadfac, isec.col);
2127                         }
2128                         else if(RE_ray_tree_intersect(R.raytree, &isec)) shadfac[3]= 0.0f;
2129                 }
2130                 else {
2131                         ray_shadow_jitter(shi, lar, lampco, shadfac, &isec);
2132                 }
2133         }
2134                 
2135         /* for first hit optim, set last interesected shadow face */
2136         if(shi->depth==0) {
2137                 lar->vlr_last[shi->thread]= (VlakRen*)isec.face_last;
2138                 lar->obi_last[shi->thread]= RAY_OBJECT_GET(&R, isec.ob_last);
2139         }
2140
2141 }
2142
2143 /* only when face points away from lamp, in direction of lamp, trace ray and find first exit point */
2144 static void ray_translucent(ShadeInput *shi, LampRen *lar, float *distfac, float *co)
2145 {
2146         Isect isec;
2147         float lampco[3], maxsize;
2148         
2149         /* setup isec */
2150         isec.mode= RE_RAY_SHADOW_TRA;
2151         
2152         if(lar->mode & LA_LAYER) isec.lay= lar->lay; else isec.lay= -1;
2153         
2154         if(lar->type==LA_SUN || lar->type==LA_HEMI) {
2155                 maxsize= RE_ray_tree_max_size(R.raytree);
2156                 lampco[0]= shi->co[0] - maxsize*lar->vec[0];
2157                 lampco[1]= shi->co[1] - maxsize*lar->vec[1];
2158                 lampco[2]= shi->co[2] - maxsize*lar->vec[2];
2159         }
2160         else {
2161                 VECCOPY(lampco, lar->co);
2162         }
2163         
2164         isec.faceorig= (RayFace*)shi->vlr;
2165         isec.oborig= RAY_OBJECT_SET(&R, shi->obi);
2166         
2167         /* set up isec vec */
2168         VECCOPY(isec.start, shi->co);
2169         VECCOPY(isec.end, lampco);
2170         
2171         if(RE_ray_tree_intersect(R.raytree, &isec)) {
2172                 /* we got a face */
2173                 
2174                 /* render co */
2175                 co[0]= isec.start[0]+isec.labda*(isec.vec[0]);
2176                 co[1]= isec.start[1]+isec.labda*(isec.vec[1]);
2177                 co[2]= isec.start[2]+isec.labda*(isec.vec[2]);
2178                 
2179                 *distfac= VecLength(isec.vec);
2180         }
2181         else
2182                 *distfac= 0.0f;
2183 }
2184
2185