forked from melonDS-emu/melonDS
-
Notifications
You must be signed in to change notification settings - Fork 0
/
GPU3D_Soft.cpp
1811 lines (1479 loc) · 57.7 KB
/
GPU3D_Soft.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
Copyright 2016-2023 melonDS team
This file is part of melonDS.
melonDS is free software: you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation, either version 3 of the License, or (at your option)
any later version.
melonDS is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with melonDS. If not, see http://www.gnu.org/licenses/.
*/
#include "GPU3D_Soft.h"
#include <algorithm>
#include <stdio.h>
#include <string.h>
#include "NDS.h"
#include "GPU.h"
namespace melonDS
{
void RenderThreadFunc();
void SoftRenderer::StopRenderThread()
{
if (RenderThreadRunning.load(std::memory_order_relaxed))
{
// Tell the render thread to stop drawing new frames, and finish up the current one.
RenderThreadRunning = false;
Platform::Semaphore_Post(Sema_RenderStart);
Platform::Thread_Wait(RenderThread);
Platform::Thread_Free(RenderThread);
RenderThread = nullptr;
}
}
void SoftRenderer::SetupRenderThread(GPU& gpu)
{
if (Threaded)
{
if (!RenderThreadRunning.load(std::memory_order_relaxed))
{ // If the render thread isn't already running...
RenderThreadRunning = true; // "Time for work, render thread!"
RenderThread = Platform::Thread_Create([this, &gpu]() {
RenderThreadFunc(gpu);
});
}
// "Be on standby, but don't start rendering until I tell you to!"
Platform::Semaphore_Reset(Sema_RenderStart);
// "Oh, sorry, were you already in the middle of a frame from the last iteration?"
if (RenderThreadRendering)
// "Tell me when you're done, I'll wait here."
Platform::Semaphore_Wait(Sema_RenderDone);
// "All good? Okay, let me give you your training."
// "(Maybe you're still the same thread, but I have to tell you this stuff anyway.)"
// "This is the signal you'll send when you're done with a frame."
// "I'll listen for it when I need to show something to the frontend."
Platform::Semaphore_Reset(Sema_RenderDone);
// "This is the signal I'll send when I want you to start rendering."
// "Don't do anything until you get the message."
Platform::Semaphore_Reset(Sema_RenderStart);
// "This is the signal you'll send every time you finish drawing a line."
// "I might need some of your scanlines before you finish the whole buffer,"
// "so let me know as soon as you're done with each one."
Platform::Semaphore_Reset(Sema_ScanlineCount);
}
else
{
StopRenderThread();
}
}
void SoftRenderer::EnableRenderThread()
{
if (Threaded && Sema_RenderStart)
{
Platform::Semaphore_Post(Sema_RenderStart);
}
}
SoftRenderer::SoftRenderer() noexcept
: Renderer3D(false)
{
Sema_RenderStart = Platform::Semaphore_Create();
Sema_RenderDone = Platform::Semaphore_Create();
Sema_ScanlineCount = Platform::Semaphore_Create();
RenderThreadRunning = false;
RenderThreadRendering = false;
RenderThread = nullptr;
}
SoftRenderer::~SoftRenderer()
{
StopRenderThread();
Platform::Semaphore_Free(Sema_RenderStart);
Platform::Semaphore_Free(Sema_RenderDone);
Platform::Semaphore_Free(Sema_ScanlineCount);
}
void SoftRenderer::Reset(GPU& gpu)
{
memset(ColorBuffer, 0, BufferSize * 2 * 4);
memset(DepthBuffer, 0, BufferSize * 2 * 4);
memset(AttrBuffer, 0, BufferSize * 2 * 4);
PrevIsShadowMask = false;
SetupRenderThread(gpu);
EnableRenderThread();
}
void SoftRenderer::SetThreaded(bool threaded, GPU& gpu) noexcept
{
if (Threaded != threaded)
{
Threaded = threaded;
SetupRenderThread(gpu);
EnableRenderThread();
}
}
void SoftRenderer::TextureLookup(const GPU& gpu, u32 texparam, u32 texpal, s16 s, s16 t, u16* color, u8* alpha) const
{
u32 vramaddr = (texparam & 0xFFFF) << 3;
s32 width = 8 << ((texparam >> 20) & 0x7);
s32 height = 8 << ((texparam >> 23) & 0x7);
s >>= 4;
t >>= 4;
// texture wrapping
// TODO: optimize this somehow
// testing shows that it's hardly worth optimizing, actually
if (texparam & (1<<16))
{
if (texparam & (1<<18))
{
if (s & width) s = (width-1) - (s & (width-1));
else s = (s & (width-1));
}
else
s &= width-1;
}
else
{
if (s < 0) s = 0;
else if (s >= width) s = width-1;
}
if (texparam & (1<<17))
{
if (texparam & (1<<19))
{
if (t & height) t = (height-1) - (t & (height-1));
else t = (t & (height-1));
}
else
t &= height-1;
}
else
{
if (t < 0) t = 0;
else if (t >= height) t = height-1;
}
u8 alpha0;
if (texparam & (1<<29)) alpha0 = 0;
else alpha0 = 31;
switch ((texparam >> 26) & 0x7)
{
case 1: // A3I5
{
vramaddr += ((t * width) + s);
u8 pixel = ReadVRAM_Texture<u8>(vramaddr, gpu);
texpal <<= 4;
*color = ReadVRAM_TexPal<u16>(texpal + ((pixel&0x1F)<<1), gpu);
*alpha = ((pixel >> 3) & 0x1C) + (pixel >> 6);
}
break;
case 2: // 4-color
{
vramaddr += (((t * width) + s) >> 2);
u8 pixel = ReadVRAM_Texture<u8>(vramaddr, gpu);
pixel >>= ((s & 0x3) << 1);
pixel &= 0x3;
texpal <<= 3;
*color = ReadVRAM_TexPal<u16>(texpal + (pixel<<1), gpu);
*alpha = (pixel==0) ? alpha0 : 31;
}
break;
case 3: // 16-color
{
vramaddr += (((t * width) + s) >> 1);
u8 pixel = ReadVRAM_Texture<u8>(vramaddr, gpu);
if (s & 0x1) pixel >>= 4;
else pixel &= 0xF;
texpal <<= 4;
*color = ReadVRAM_TexPal<u16>(texpal + (pixel<<1), gpu);
*alpha = (pixel==0) ? alpha0 : 31;
}
break;
case 4: // 256-color
{
vramaddr += ((t * width) + s);
u8 pixel = ReadVRAM_Texture<u8>(vramaddr, gpu);
texpal <<= 4;
*color = ReadVRAM_TexPal<u16>(texpal + (pixel<<1), gpu);
*alpha = (pixel==0) ? alpha0 : 31;
}
break;
case 5: // compressed
{
vramaddr += ((t & 0x3FC) * (width>>2)) + (s & 0x3FC);
vramaddr += (t & 0x3);
u32 slot1addr = 0x20000 + ((vramaddr & 0x1FFFC) >> 1);
if (vramaddr >= 0x40000)
slot1addr += 0x10000;
u8 val = ReadVRAM_Texture<u8>(vramaddr, gpu);
val >>= (2 * (s & 0x3));
u16 palinfo = ReadVRAM_Texture<u16>(slot1addr, gpu);
u32 paloffset = (palinfo & 0x3FFF) << 2;
texpal <<= 4;
switch (val & 0x3)
{
case 0:
*color = ReadVRAM_TexPal<u16>(texpal + paloffset, gpu);
*alpha = 31;
break;
case 1:
*color = ReadVRAM_TexPal<u16>(texpal + paloffset + 2, gpu);
*alpha = 31;
break;
case 2:
if ((palinfo >> 14) == 1)
{
u16 color0 = ReadVRAM_TexPal<u16>(texpal + paloffset, gpu);
u16 color1 = ReadVRAM_TexPal<u16>(texpal + paloffset + 2, gpu);
u32 r0 = color0 & 0x001F;
u32 g0 = color0 & 0x03E0;
u32 b0 = color0 & 0x7C00;
u32 r1 = color1 & 0x001F;
u32 g1 = color1 & 0x03E0;
u32 b1 = color1 & 0x7C00;
u32 r = (r0 + r1) >> 1;
u32 g = ((g0 + g1) >> 1) & 0x03E0;
u32 b = ((b0 + b1) >> 1) & 0x7C00;
*color = r | g | b;
}
else if ((palinfo >> 14) == 3)
{
u16 color0 = ReadVRAM_TexPal<u16>(texpal + paloffset, gpu);
u16 color1 = ReadVRAM_TexPal<u16>(texpal + paloffset + 2, gpu);
u32 r0 = color0 & 0x001F;
u32 g0 = color0 & 0x03E0;
u32 b0 = color0 & 0x7C00;
u32 r1 = color1 & 0x001F;
u32 g1 = color1 & 0x03E0;
u32 b1 = color1 & 0x7C00;
u32 r = (r0*5 + r1*3) >> 3;
u32 g = ((g0*5 + g1*3) >> 3) & 0x03E0;
u32 b = ((b0*5 + b1*3) >> 3) & 0x7C00;
*color = r | g | b;
}
else
*color = ReadVRAM_TexPal<u16>(texpal + paloffset + 4, gpu);
*alpha = 31;
break;
case 3:
if ((palinfo >> 14) == 2)
{
*color = ReadVRAM_TexPal<u16>(texpal + paloffset + 6, gpu);
*alpha = 31;
}
else if ((palinfo >> 14) == 3)
{
u16 color0 = ReadVRAM_TexPal<u16>(texpal + paloffset, gpu);
u16 color1 = ReadVRAM_TexPal<u16>(texpal + paloffset + 2, gpu);
u32 r0 = color0 & 0x001F;
u32 g0 = color0 & 0x03E0;
u32 b0 = color0 & 0x7C00;
u32 r1 = color1 & 0x001F;
u32 g1 = color1 & 0x03E0;
u32 b1 = color1 & 0x7C00;
u32 r = (r0*3 + r1*5) >> 3;
u32 g = ((g0*3 + g1*5) >> 3) & 0x03E0;
u32 b = ((b0*3 + b1*5) >> 3) & 0x7C00;
*color = r | g | b;
*alpha = 31;
}
else
{
*color = 0;
*alpha = 0;
}
break;
}
}
break;
case 6: // A5I3
{
vramaddr += ((t * width) + s);
u8 pixel = ReadVRAM_Texture<u8>(vramaddr, gpu);
texpal <<= 4;
*color = ReadVRAM_TexPal<u16>(texpal + ((pixel&0x7)<<1), gpu);
*alpha = (pixel >> 3);
}
break;
case 7: // direct color
{
vramaddr += (((t * width) + s) << 1);
*color = ReadVRAM_Texture<u16>(vramaddr, gpu);
*alpha = (*color & 0x8000) ? 31 : 0;
}
break;
}
}
// depth test is 'less or equal' instead of 'less than' under the following conditions:
// * when drawing a front-facing pixel over an opaque back-facing pixel
// * when drawing wireframe edges, under certain conditions (TODO)
//
// range is different based on depth-buffering mode
// Z-buffering: +-0x200
// W-buffering: +-0xFF
bool DepthTest_Equal_Z(s32 dstz, s32 z, u32 dstattr)
{
s32 diff = dstz - z;
if ((u32)(diff + 0x200) <= 0x400)
return true;
return false;
}
bool DepthTest_Equal_W(s32 dstz, s32 z, u32 dstattr)
{
s32 diff = dstz - z;
if ((u32)(diff + 0xFF) <= 0x1FE)
return true;
return false;
}
bool DepthTest_LessThan(s32 dstz, s32 z, u32 dstattr)
{
if (z < dstz)
return true;
return false;
}
bool DepthTest_LessThan_FrontFacing(s32 dstz, s32 z, u32 dstattr)
{
if ((dstattr & 0x00400010) == 0x00000010) // opaque, back facing
{
if (z <= dstz)
return true;
}
else
{
if (z < dstz)
return true;
}
return false;
}
u32 SoftRenderer::AlphaBlend(const GPU3D& gpu3d, u32 srccolor, u32 dstcolor, u32 alpha) const noexcept
{
u32 dstalpha = dstcolor >> 24;
if (dstalpha == 0)
return srccolor;
u32 srcR = srccolor & 0x3F;
u32 srcG = (srccolor >> 8) & 0x3F;
u32 srcB = (srccolor >> 16) & 0x3F;
if (gpu3d.RenderDispCnt & (1<<3))
{
u32 dstR = dstcolor & 0x3F;
u32 dstG = (dstcolor >> 8) & 0x3F;
u32 dstB = (dstcolor >> 16) & 0x3F;
alpha++;
srcR = ((srcR * alpha) + (dstR * (32-alpha))) >> 5;
srcG = ((srcG * alpha) + (dstG * (32-alpha))) >> 5;
srcB = ((srcB * alpha) + (dstB * (32-alpha))) >> 5;
alpha--;
}
if (alpha > dstalpha)
dstalpha = alpha;
return srcR | (srcG << 8) | (srcB << 16) | (dstalpha << 24);
}
u32 SoftRenderer::RenderPixel(const GPU& gpu, const Polygon* polygon, u8 vr, u8 vg, u8 vb, s16 s, s16 t) const
{
u8 r, g, b, a;
u32 blendmode = (polygon->Attr >> 4) & 0x3;
u32 polyalpha = (polygon->Attr >> 16) & 0x1F;
bool wireframe = (polyalpha == 0);
if (blendmode == 2)
{
if (gpu.GPU3D.RenderDispCnt & (1<<1))
{
// highlight mode: color is calculated normally
// except all vertex color components are set
// to the red component
// the toon color is added to the final color
vg = vr;
vb = vr;
}
else
{
// toon mode: vertex color is replaced by toon color
u16 tooncolor = gpu.GPU3D.RenderToonTable[vr >> 1];
vr = (tooncolor << 1) & 0x3E; if (vr) vr++;
vg = (tooncolor >> 4) & 0x3E; if (vg) vg++;
vb = (tooncolor >> 9) & 0x3E; if (vb) vb++;
}
}
if ((gpu.GPU3D.RenderDispCnt & (1<<0)) && (((polygon->TexParam >> 26) & 0x7) != 0))
{
u8 tr, tg, tb;
u16 tcolor; u8 talpha;
TextureLookup(gpu, polygon->TexParam, polygon->TexPalette, s, t, &tcolor, &talpha);
tr = (tcolor << 1) & 0x3E; if (tr) tr++;
tg = (tcolor >> 4) & 0x3E; if (tg) tg++;
tb = (tcolor >> 9) & 0x3E; if (tb) tb++;
if (blendmode & 0x1)
{
// decal
if (talpha == 0)
{
r = vr;
g = vg;
b = vb;
}
else if (talpha == 31)
{
r = tr;
g = tg;
b = tb;
}
else
{
r = ((tr * talpha) + (vr * (31-talpha))) >> 5;
g = ((tg * talpha) + (vg * (31-talpha))) >> 5;
b = ((tb * talpha) + (vb * (31-talpha))) >> 5;
}
a = polyalpha;
}
else
{
// modulate
r = ((tr+1) * (vr+1) - 1) >> 6;
g = ((tg+1) * (vg+1) - 1) >> 6;
b = ((tb+1) * (vb+1) - 1) >> 6;
a = ((talpha+1) * (polyalpha+1) - 1) >> 5;
}
}
else
{
r = vr;
g = vg;
b = vb;
a = polyalpha;
}
if ((blendmode == 2) && (gpu.GPU3D.RenderDispCnt & (1<<1)))
{
u16 tooncolor = gpu.GPU3D.RenderToonTable[vr >> 1];
vr = (tooncolor << 1) & 0x3E; if (vr) vr++;
vg = (tooncolor >> 4) & 0x3E; if (vg) vg++;
vb = (tooncolor >> 9) & 0x3E; if (vb) vb++;
r += vr;
g += vg;
b += vb;
if (r > 63) r = 63;
if (g > 63) g = 63;
if (b > 63) b = 63;
}
// checkme: can wireframe polygons use texture alpha?
if (wireframe) a = 31;
return r | (g << 8) | (b << 16) | (a << 24);
}
void SoftRenderer::PlotTranslucentPixel(const GPU3D& gpu3d, u32 pixeladdr, u32 color, u32 z, u32 polyattr, u32 shadow)
{
u32 dstattr = AttrBuffer[pixeladdr];
u32 attr = (polyattr & 0xE0F0) | ((polyattr >> 8) & 0xFF0000) | (1<<22) | (dstattr & 0xFF001F0F);
if (shadow)
{
// for shadows, opaque pixels are also checked
if (dstattr & (1<<22))
{
if ((dstattr & 0x007F0000) == (attr & 0x007F0000))
return;
}
else
{
if ((dstattr & 0x3F000000) == (polyattr & 0x3F000000))
return;
}
}
else
{
// skip if translucent polygon IDs are equal
if ((dstattr & 0x007F0000) == (attr & 0x007F0000))
return;
}
// fog flag
if (!(dstattr & (1<<15)))
attr &= ~(1<<15);
color = AlphaBlend(gpu3d, color, ColorBuffer[pixeladdr], color>>24);
if (z != -1)
DepthBuffer[pixeladdr] = z;
ColorBuffer[pixeladdr] = color;
AttrBuffer[pixeladdr] = attr;
}
void SoftRenderer::SetupPolygonLeftEdge(SoftRenderer::RendererPolygon* rp, s32 y) const
{
Polygon* polygon = rp->PolyData;
while (y >= polygon->Vertices[rp->NextVL]->FinalPosition[1] && rp->CurVL != polygon->VBottom)
{
rp->CurVL = rp->NextVL;
if (polygon->FacingView)
{
rp->NextVL = rp->CurVL + 1;
if (rp->NextVL >= polygon->NumVertices)
rp->NextVL = 0;
}
else
{
rp->NextVL = rp->CurVL - 1;
if ((s32)rp->NextVL < 0)
rp->NextVL = polygon->NumVertices - 1;
}
}
rp->XL = rp->SlopeL.Setup(polygon->Vertices[rp->CurVL]->FinalPosition[0], polygon->Vertices[rp->NextVL]->FinalPosition[0],
polygon->Vertices[rp->CurVL]->FinalPosition[1], polygon->Vertices[rp->NextVL]->FinalPosition[1],
polygon->FinalW[rp->CurVL], polygon->FinalW[rp->NextVL], y);
}
void SoftRenderer::SetupPolygonRightEdge(SoftRenderer::RendererPolygon* rp, s32 y) const
{
Polygon* polygon = rp->PolyData;
while (y >= polygon->Vertices[rp->NextVR]->FinalPosition[1] && rp->CurVR != polygon->VBottom)
{
rp->CurVR = rp->NextVR;
if (polygon->FacingView)
{
rp->NextVR = rp->CurVR - 1;
if ((s32)rp->NextVR < 0)
rp->NextVR = polygon->NumVertices - 1;
}
else
{
rp->NextVR = rp->CurVR + 1;
if (rp->NextVR >= polygon->NumVertices)
rp->NextVR = 0;
}
}
rp->XR = rp->SlopeR.Setup(polygon->Vertices[rp->CurVR]->FinalPosition[0], polygon->Vertices[rp->NextVR]->FinalPosition[0],
polygon->Vertices[rp->CurVR]->FinalPosition[1], polygon->Vertices[rp->NextVR]->FinalPosition[1],
polygon->FinalW[rp->CurVR], polygon->FinalW[rp->NextVR], y);
}
void SoftRenderer::SetupPolygon(SoftRenderer::RendererPolygon* rp, Polygon* polygon) const
{
u32 nverts = polygon->NumVertices;
u32 vtop = polygon->VTop, vbot = polygon->VBottom;
s32 ytop = polygon->YTop, ybot = polygon->YBottom;
rp->PolyData = polygon;
rp->CurVL = vtop;
rp->CurVR = vtop;
if (polygon->FacingView)
{
rp->NextVL = rp->CurVL + 1;
if (rp->NextVL >= nverts) rp->NextVL = 0;
rp->NextVR = rp->CurVR - 1;
if ((s32)rp->NextVR < 0) rp->NextVR = nverts - 1;
}
else
{
rp->NextVL = rp->CurVL - 1;
if ((s32)rp->NextVL < 0) rp->NextVL = nverts - 1;
rp->NextVR = rp->CurVR + 1;
if (rp->NextVR >= nverts) rp->NextVR = 0;
}
if (ybot == ytop)
{
vtop = 0; vbot = 0;
int i;
i = 1;
if (polygon->Vertices[i]->FinalPosition[0] < polygon->Vertices[vtop]->FinalPosition[0]) vtop = i;
if (polygon->Vertices[i]->FinalPosition[0] > polygon->Vertices[vbot]->FinalPosition[0]) vbot = i;
i = nverts - 1;
if (polygon->Vertices[i]->FinalPosition[0] < polygon->Vertices[vtop]->FinalPosition[0]) vtop = i;
if (polygon->Vertices[i]->FinalPosition[0] > polygon->Vertices[vbot]->FinalPosition[0]) vbot = i;
rp->CurVL = vtop; rp->NextVL = vtop;
rp->CurVR = vbot; rp->NextVR = vbot;
rp->XL = rp->SlopeL.SetupDummy(polygon->Vertices[rp->CurVL]->FinalPosition[0]);
rp->XR = rp->SlopeR.SetupDummy(polygon->Vertices[rp->CurVR]->FinalPosition[0]);
}
else
{
SetupPolygonLeftEdge(rp, ytop);
SetupPolygonRightEdge(rp, ytop);
}
}
void SoftRenderer::RenderShadowMaskScanline(const GPU3D& gpu3d, RendererPolygon* rp, s32 y)
{
Polygon* polygon = rp->PolyData;
u32 polyattr = (polygon->Attr & 0x3F008000);
if (!polygon->FacingView) polyattr |= (1<<4);
u32 polyalpha = (polygon->Attr >> 16) & 0x1F;
bool wireframe = (polyalpha == 0);
bool (*fnDepthTest)(s32 dstz, s32 z, u32 dstattr);
if (polygon->Attr & (1<<14))
fnDepthTest = polygon->WBuffer ? DepthTest_Equal_W : DepthTest_Equal_Z;
else if (polygon->FacingView)
fnDepthTest = DepthTest_LessThan_FrontFacing;
else
fnDepthTest = DepthTest_LessThan;
if (!PrevIsShadowMask)
memset(&StencilBuffer[256 * (y&0x1)], 0, 256);
PrevIsShadowMask = true;
if (polygon->YTop != polygon->YBottom)
{
if (y >= polygon->Vertices[rp->NextVL]->FinalPosition[1] && rp->CurVL != polygon->VBottom)
{
SetupPolygonLeftEdge(rp, y);
}
if (y >= polygon->Vertices[rp->NextVR]->FinalPosition[1] && rp->CurVR != polygon->VBottom)
{
SetupPolygonRightEdge(rp, y);
}
}
Vertex *vlcur, *vlnext, *vrcur, *vrnext;
s32 xstart, xend;
bool l_filledge, r_filledge;
s32 l_edgelen, r_edgelen;
s32 l_edgecov, r_edgecov;
Interpolator<1>* interp_start;
Interpolator<1>* interp_end;
xstart = rp->XL;
xend = rp->XR;
s32 wl = rp->SlopeL.Interp.Interpolate(polygon->FinalW[rp->CurVL], polygon->FinalW[rp->NextVL]);
s32 wr = rp->SlopeR.Interp.Interpolate(polygon->FinalW[rp->CurVR], polygon->FinalW[rp->NextVR]);
s32 zl = rp->SlopeL.Interp.InterpolateZ(polygon->FinalZ[rp->CurVL], polygon->FinalZ[rp->NextVL], polygon->WBuffer);
s32 zr = rp->SlopeR.Interp.InterpolateZ(polygon->FinalZ[rp->CurVR], polygon->FinalZ[rp->NextVR], polygon->WBuffer);
// right vertical edges are pushed 1px to the left as long as either:
// the left edge slope is not 0, or the span is not 0 pixels wide, and it is not at the leftmost pixel of the screen
if (rp->SlopeR.Increment==0 && (rp->SlopeL.Increment!=0 || xstart != xend) && (xend != 0))
xend--;
// if the left and right edges are swapped, render backwards.
if (xstart > xend)
{
vlcur = polygon->Vertices[rp->CurVR];
vlnext = polygon->Vertices[rp->NextVR];
vrcur = polygon->Vertices[rp->CurVL];
vrnext = polygon->Vertices[rp->NextVL];
interp_start = &rp->SlopeR.Interp;
interp_end = &rp->SlopeL.Interp;
rp->SlopeR.EdgeParams<true>(&l_edgelen, &l_edgecov);
rp->SlopeL.EdgeParams<true>(&r_edgelen, &r_edgecov);
std::swap(xstart, xend);
std::swap(wl, wr);
std::swap(zl, zr);
// CHECKME: edge fill rules for swapped opaque shadow mask polygons
if ((gpu3d.RenderDispCnt & ((1<<4)|(1<<5))) || ((polyalpha < 31) && (gpu3d.RenderDispCnt & (1<<3))) || wireframe)
{
l_filledge = true;
r_filledge = true;
}
else
{
l_filledge = (rp->SlopeR.Negative || !rp->SlopeR.XMajor)
|| (y == polygon->YBottom-1) && rp->SlopeR.XMajor && (vlnext->FinalPosition[0] != vrnext->FinalPosition[0]);
r_filledge = (!rp->SlopeL.Negative && rp->SlopeL.XMajor)
|| (!(rp->SlopeL.Negative && rp->SlopeL.XMajor) && rp->SlopeR.Increment==0)
|| (y == polygon->YBottom-1) && rp->SlopeL.XMajor && (vlnext->FinalPosition[0] != vrnext->FinalPosition[0]);
}
}
else
{
vlcur = polygon->Vertices[rp->CurVL];
vlnext = polygon->Vertices[rp->NextVL];
vrcur = polygon->Vertices[rp->CurVR];
vrnext = polygon->Vertices[rp->NextVR];
interp_start = &rp->SlopeL.Interp;
interp_end = &rp->SlopeR.Interp;
rp->SlopeL.EdgeParams<false>(&l_edgelen, &l_edgecov);
rp->SlopeR.EdgeParams<false>(&r_edgelen, &r_edgecov);
// CHECKME: edge fill rules for unswapped opaque shadow mask polygons
if ((gpu3d.RenderDispCnt & ((1<<4)|(1<<5))) || ((polyalpha < 31) && (gpu3d.RenderDispCnt & (1<<3))) || wireframe)
{
l_filledge = true;
r_filledge = true;
}
else
{
l_filledge = ((rp->SlopeL.Negative || !rp->SlopeL.XMajor)
|| (y == polygon->YBottom-1) && rp->SlopeL.XMajor && (vlnext->FinalPosition[0] != vrnext->FinalPosition[0]))
|| (rp->SlopeL.Increment == rp->SlopeR.Increment) && (xstart+l_edgelen == xend+1);
r_filledge = (!rp->SlopeR.Negative && rp->SlopeR.XMajor) || (rp->SlopeR.Increment==0)
|| (y == polygon->YBottom-1) && rp->SlopeR.XMajor && (vlnext->FinalPosition[0] != vrnext->FinalPosition[0]);
}
}
// color/texcoord attributes aren't needed for shadow masks
// all the pixels are guaranteed to have the same alpha
// even if a texture is used (decal blending is used for shadows)
// similarly, we can perform alpha test early (checkme)
if (wireframe) polyalpha = 31;
if (polyalpha <= gpu3d.RenderAlphaRef) return;
// in wireframe mode, there are special rules for equal Z (TODO)
int yedge = 0;
if (y == polygon->YTop) yedge = 0x4;
else if (y == polygon->YBottom-1) yedge = 0x8;
int edge;
s32 x = xstart;
Interpolator<0> interpX(xstart, xend+1, wl, wr);
if (x < 0) x = 0;
s32 xlimit;
// for shadow masks: set stencil bits where the depth test fails.
// draw nothing.
// part 1: left edge
edge = yedge | 0x1;
xlimit = xstart+l_edgelen;
if (xlimit > xend+1) xlimit = xend+1;
if (xlimit > 256) xlimit = 256;
if (!l_filledge) x = xlimit;
else
for (; x < xlimit; x++)
{
u32 pixeladdr = FirstPixelOffset + (y*ScanlineWidth) + x;
interpX.SetX(x);
s32 z = interpX.InterpolateZ(zl, zr, polygon->WBuffer);
u32 dstattr = AttrBuffer[pixeladdr];
if (!fnDepthTest(DepthBuffer[pixeladdr], z, dstattr))
StencilBuffer[256*(y&0x1) + x] = 1;
if (dstattr & 0xF)
{
pixeladdr += BufferSize;
if (!fnDepthTest(DepthBuffer[pixeladdr], z, AttrBuffer[pixeladdr]))
StencilBuffer[256*(y&0x1) + x] |= 0x2;
}
}
// part 2: polygon inside
edge = yedge;
xlimit = xend-r_edgelen+1;
if (xlimit > xend+1) xlimit = xend+1;
if (xlimit > 256) xlimit = 256;
if (wireframe && !edge) x = std::max(x, xlimit);
else for (; x < xlimit; x++)
{
u32 pixeladdr = FirstPixelOffset + (y*ScanlineWidth) + x;
interpX.SetX(x);
s32 z = interpX.InterpolateZ(zl, zr, polygon->WBuffer);
u32 dstattr = AttrBuffer[pixeladdr];
if (!fnDepthTest(DepthBuffer[pixeladdr], z, dstattr))
StencilBuffer[256*(y&0x1) + x] = 1;
if (dstattr & 0xF)
{
pixeladdr += BufferSize;
if (!fnDepthTest(DepthBuffer[pixeladdr], z, AttrBuffer[pixeladdr]))
StencilBuffer[256*(y&0x1) + x] |= 0x2;
}
}
// part 3: right edge
edge = yedge | 0x2;
xlimit = xend+1;
if (xlimit > 256) xlimit = 256;
if (r_filledge)
for (; x < xlimit; x++)
{
u32 pixeladdr = FirstPixelOffset + (y*ScanlineWidth) + x;
interpX.SetX(x);
s32 z = interpX.InterpolateZ(zl, zr, polygon->WBuffer);
u32 dstattr = AttrBuffer[pixeladdr];
if (!fnDepthTest(DepthBuffer[pixeladdr], z, dstattr))
StencilBuffer[256*(y&0x1) + x] = 1;
if (dstattr & 0xF)
{
pixeladdr += BufferSize;
if (!fnDepthTest(DepthBuffer[pixeladdr], z, AttrBuffer[pixeladdr]))
StencilBuffer[256*(y&0x1) + x] |= 0x2;
}
}
rp->XL = rp->SlopeL.Step();
rp->XR = rp->SlopeR.Step();
}
void SoftRenderer::RenderPolygonScanline(const GPU& gpu, RendererPolygon* rp, s32 y)
{
Polygon* polygon = rp->PolyData;
u32 polyattr = (polygon->Attr & 0x3F008000);
if (!polygon->FacingView) polyattr |= (1<<4);
u32 polyalpha = (polygon->Attr >> 16) & 0x1F;
bool wireframe = (polyalpha == 0);
bool (*fnDepthTest)(s32 dstz, s32 z, u32 dstattr);
if (polygon->Attr & (1<<14))
fnDepthTest = polygon->WBuffer ? DepthTest_Equal_W : DepthTest_Equal_Z;
else if (polygon->FacingView)
fnDepthTest = DepthTest_LessThan_FrontFacing;
else
fnDepthTest = DepthTest_LessThan;
PrevIsShadowMask = false;
if (polygon->YTop != polygon->YBottom)
{
if (y >= polygon->Vertices[rp->NextVL]->FinalPosition[1] && rp->CurVL != polygon->VBottom)
{
SetupPolygonLeftEdge(rp, y);
}
if (y >= polygon->Vertices[rp->NextVR]->FinalPosition[1] && rp->CurVR != polygon->VBottom)
{
SetupPolygonRightEdge(rp, y);
}
}
Vertex *vlcur, *vlnext, *vrcur, *vrnext;
s32 xstart, xend;
bool l_filledge, r_filledge;
s32 l_edgelen, r_edgelen;
s32 l_edgecov, r_edgecov;
Interpolator<1>* interp_start;
Interpolator<1>* interp_end;
xstart = rp->XL;
xend = rp->XR;
s32 wl = rp->SlopeL.Interp.Interpolate(polygon->FinalW[rp->CurVL], polygon->FinalW[rp->NextVL]);
s32 wr = rp->SlopeR.Interp.Interpolate(polygon->FinalW[rp->CurVR], polygon->FinalW[rp->NextVR]);
s32 zl = rp->SlopeL.Interp.InterpolateZ(polygon->FinalZ[rp->CurVL], polygon->FinalZ[rp->NextVL], polygon->WBuffer);
s32 zr = rp->SlopeR.Interp.InterpolateZ(polygon->FinalZ[rp->CurVR], polygon->FinalZ[rp->NextVR], polygon->WBuffer);
// right vertical edges are pushed 1px to the left as long as either:
// the left edge slope is not 0, or the span is not 0 pixels wide, and it is not at the leftmost pixel of the screen
if (rp->SlopeR.Increment==0 && (rp->SlopeL.Increment!=0 || xstart != xend) && (xend != 0))
xend--;
// if the left and right edges are swapped, render backwards.
// on hardware, swapped edges seem to break edge length calculation,
// causing X-major edges to be rendered wrong when filled,
// and resulting in buggy looking anti-aliasing on X-major edges
if (xstart > xend)
{
vlcur = polygon->Vertices[rp->CurVR];
vlnext = polygon->Vertices[rp->NextVR];
vrcur = polygon->Vertices[rp->CurVL];
vrnext = polygon->Vertices[rp->NextVL];
interp_start = &rp->SlopeR.Interp;
interp_end = &rp->SlopeL.Interp;
rp->SlopeR.EdgeParams<true>(&l_edgelen, &l_edgecov);
rp->SlopeL.EdgeParams<true>(&r_edgelen, &r_edgecov);