ShuriZma
/
desmume
mirror of https://github.com/TASEmulators/desmume.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

gpu: fix sprite rendering order to be front to back

This commit is contained in:
zeromus 2010-01-28 04:30:58 +00:00
parent 136f3a8642
commit 45e95b1a8a
1 changed files with 57 additions and 103 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

160
desmume/src/GPU.cpp
View File

@ -1,10 +1,8 @@
/* Copyright (C) 2006 yopyop
yopyop156@ifrance.com
yopyop156.ifrance.com
/* GPU.cpp
Copyright (C) 2006 yopyop
Copyright (C) 2006-2007 Theo Berkau
Copyright (C) 2007 shash
Copyright (C) 2008-2009 DeSmuME team
Copyright (C) 2008-2010 DeSmuME team
This file is part of DeSmuME
@ -1302,17 +1300,13 @@ INLINE void render_sprite_BMP (GPU * gpu, u8 spriteNum, u16 l, u8 * dst, u16 * s
color = LE_TO_LOCAL_16(src[x]);
// alpha bit = invisible
if ((color&0x8000)&&(prio<=prioTab[sprX]))
if ((color&0x8000)&&(prio<prioTab[sprX]))
{
/* if we don't draw, do not set prio, or else */
// if (gpu->setFinalColorSpr(gpu, sprX << 1,4,dst, color, sprX))
{
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = alpha;
typeTab[sprX] = 3;
prioTab[sprX] = prio;
gpu->sprNum[sprX] = spriteNum;
}
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = alpha;
typeTab[sprX] = 3;
prioTab[sprX] = prio;
gpu->sprNum[sprX] = spriteNum;
}
}
}
@ -1330,17 +1324,13 @@ INLINE void render_sprite_256 ( GPU * gpu, u8 spriteNum, u16 l, u8 * dst, u8 * s
color = LE_TO_LOCAL_16(pal[palette_entry]);
// palette entry = 0 means backdrop
if ((palette_entry>0)&&(prio<=prioTab[sprX]))
if ((palette_entry>0)&&(prio<prioTab[sprX]))
{
/* if we don't draw, do not set prio, or else */
//if (gpu->setFinalColorSpr(gpu, sprX << 1,4,dst, color, sprX))
{
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = 16;
typeTab[sprX] = (alpha ? 1 : 0);
prioTab[sprX] = prio;
gpu->sprNum[sprX] = spriteNum;
}
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = 16;
typeTab[sprX] = (alpha ? 1 : 0);
prioTab[sprX] = prio;
gpu->sprNum[sprX] = spriteNum;
}
}
}
@ -1361,16 +1351,12 @@ INLINE void render_sprite_16 ( GPU * gpu, u16 l, u8 * dst, u8 * src, u16 * pal,
color = LE_TO_LOCAL_16(pal[palette_entry]);
// palette entry = 0 means backdrop
if ((palette_entry>0)&&(prio<=prioTab[sprX]))
if ((palette_entry>0)&&(prio<prioTab[sprX]))
{
/* if we don't draw, do not set prio, or else */
//if (gpu->setFinalColorSpr(gpu, sprX << 1,4,dst, color, sprX ))
{
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = 16;
typeTab[sprX] = (alpha ? 1 : 0);
prioTab[sprX] = prio;
}
T2WriteWord(dst, (sprX<<1), color);
dst_alpha[sprX] = 16;
typeTab[sprX] = (alpha ? 1 : 0);
prioTab[sprX] = prio;
}
}
}
@ -1489,27 +1475,41 @@ void GPU::_spriteRender(u8 * dst, u8 * dst_alpha, u8 * typeTab, u8 * prioTab)
u16 l = currLine;
GPU *gpu = this;
int cost = 0;
struct _DISPCNT * dispCnt = &(gpu->dispx_st)->dispx_DISPCNT.bits;
_OAM_ * spriteInfo = (_OAM_ *)(gpu->oam + (nbShow-1));// + 127;
_OAM_ * spriteInfo = (_OAM_ *)(gpu->oam);
u8 block = gpu->sprBoundary;
u8 i;
#ifdef WORDS_BIGENDIAN
*(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) >> 1) | *(((u16*)spriteInfo)+1) << 15;
*(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) >> 2) | *(((u16*)spriteInfo)+2) << 14;
#endif
//what the hell? why is all this here? the #ifdefs in the bitfields definition should take care of this.
//this needs to be fixed anyway since i changed the sprite render order
//better yet, just dont do it this way at all. _OAM_ is so small, why not just copy it and then twiddle it?
for(i = 0; i<nbShow; ++i, --spriteInfo
#ifdef WORDS_BIGENDIAN
,*(((u16*)(spriteInfo+1))+1) = (*(((u16*)(spriteInfo+1))+1) << 1) | *(((u16*)(spriteInfo+1))+1) >> 15
,*(((u16*)(spriteInfo+1))+2) = (*(((u16*)(spriteInfo+1))+2) << 2) | *(((u16*)(spriteInfo+1))+2) >> 14
,*(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) >> 1) | *(((u16*)spriteInfo)+1) << 15
,*(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) >> 2) | *(((u16*)spriteInfo)+2) << 14
#endif
//#ifdef WORDS_BIGENDIAN
// *(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) >> 1) | *(((u16*)spriteInfo)+1) << 15;
// *(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) >> 2) | *(((u16*)spriteInfo)+2) << 14;
//#endif
for(i = 0; i<nbShow; ++i, ++spriteInfo
//#ifdef WORDS_BIGENDIAN
// ,*(((u16*)(spriteInfo+1))+1) = (*(((u16*)(spriteInfo+1))+1) << 1) | *(((u16*)(spriteInfo+1))+1) >> 15
// ,*(((u16*)(spriteInfo+1))+2) = (*(((u16*)(spriteInfo+1))+2) << 2) | *(((u16*)(spriteInfo+1))+2) >> 14
// ,*(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) >> 1) | *(((u16*)spriteInfo)+1) << 15
// ,*(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) >> 2) | *(((u16*)spriteInfo)+2) << 14
//#endif
)
{
//for each sprite:
if(cost>=2130)
{
//out of sprite rendering time
//printf("sprite overflow!\n");
//return;
}
//do we incur a cost if a sprite is disabled?? we guess so.
cost += 2;
size sprSize;
s32 sprX, sprY, x, y, lg;
@ -1558,6 +1558,8 @@ void GPU::_spriteRender(u8 * dst, u8 * dst_alpha, u8 * typeTab, u8 * prioTab)
(sprX==256) || (sprX+fieldX<=0))
continue;
cost += sprSize.x*2 + 10;
y = l - sprY;
// Get which four parameter block is assigned to this sprite
@ -1743,11 +1745,12 @@ void GPU::_spriteRender(u8 * dst, u8 * dst_alpha, u8 * typeTab, u8 * prioTab)
else
{
u16 * pal;
if (!compute_sprite_vars(spriteInfo, l, sprSize, sprX, sprY, x, y, lg, xdir))
continue;
cost += sprSize.x;
if (spriteInfo->Mode == 2)
{
if(MODE == SPRITE_2D)
@ -1816,10 +1819,11 @@ void GPU::_spriteRender(u8 * dst, u8 * dst_alpha, u8 * typeTab, u8 * prioTab)
}
}
#ifdef WORDS_BIGENDIAN
*(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) << 1) | *(((u16*)spriteInfo)+1) >> 15;
*(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) << 2) | *(((u16*)spriteInfo)+2) >> 14;
#endif
//#ifdef WORDS_BIGENDIAN
// *(((u16*)spriteInfo)+1) = (*(((u16*)spriteInfo)+1) << 1) | *(((u16*)spriteInfo)+1) >> 15;
// *(((u16*)spriteInfo)+2) = (*(((u16*)spriteInfo)+2) << 2) | *(((u16*)spriteInfo)+2) >> 14;
//#endif
}
@ -2133,10 +2137,8 @@ static void GPU_RenderLine_layer(NDS_Screen * screen, u16 l)
//useful for debugging individual layers
//if(gpu->core == 1 || i16 != 2) continue;
if(gpu->core == 0 || i16 == 2)
{
int zzz=9;
}
#ifndef DISABLE_MOSAIC
if(gpu->curr_mosaic_enabled)
@ -2801,51 +2803,3 @@ void gpu_SetRotateScreen(u16 angle)
{
gpu_angle = angle;
}
//here is an old bg mosaic with some old code commented out. I am going to leave it here for a while to look at it
//sometimes in case I find a problem with the mosaic.
//static void __setFinalColorBck(GPU *gpu, u32 passing, u8 bgnum, u8 *dst, u16 color, u16 x, bool opaque)
//{
// struct _BGxCNT *bgCnt = &(gpu->dispx_st)->dispx_BGxCNT[bgnum].bits;
// bool enabled = bgCnt->Mosaic_Enable;
//
//// if(!opaque) color = 0xFFFF;
//// else color &= 0x7FFF;
// if(!opaque)
// return;
//
// //mosaic test hacks
// enabled = true;
//
// //due to this early out, we will get incorrect behavior in cases where
// //we enable mosaic in the middle of a frame. this is deemed unlikely.
// if(enabled)
// {
// u8 y = gpu->currLine;
//
// //I intend to cache all this at the beginning of line rendering
//
// u16 mosaic_control = T1ReadWord((u8 *)&gpu->dispx_st->dispx_MISC.MOSAIC, 0);
// u8 mw = (mosaic_control & 0xF);
// u8 mh = ((mosaic_control>>4) & 0xF);
//
// //mosaic test hacks
// mw = 3;
// mh = 3;
//
// MosaicLookup::TableEntry &te_x = mosaicLookup.table[mw][x];
// MosaicLookup::TableEntry &te_y = mosaicLookup.table[mh][y];
//
// //int x_int;
// //if(enabled)
// int x_int = te_x.trunc;
// //else x_int = x;
//
// if(te_x.begin && te_y.begin) {}
// else color = gpu->MosaicColors.bg[bgnum][x_int];
// gpu->MosaicColors.bg[bgnum][x] = color;
// }
//
//// if(color != 0xFFFF)
// gpu->setFinalColorBck(gpu,0,bgnum,dst,color,x);
//}