ShuriZma
/
bsnes
mirror of https://github.com/bsnes-emu/bsnes.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
bsnes/sfc/alt/ppu-performance/background/mode7.cpp

107 lines
3.0 KiB
C++
Raw Normal View History

Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
#ifdef PPU_CPP
#define Clip(x) (((x) & 0x2000) ? ((x) | ~0x03ff) : ((x) & 0x03ff))
void PPU::Background::render_mode7() {
signed px, py;
signed tx, ty, tile, palette;
signed a = sclip<16>(self.regs.m7a);
signed b = sclip<16>(self.regs.m7b);
signed c = sclip<16>(self.regs.m7c);
signed d = sclip<16>(self.regs.m7d);
signed cx = sclip<13>(self.regs.m7x);
signed cy = sclip<13>(self.regs.m7y);
signed hofs = sclip<13>(self.regs.mode7_hoffset);
signed vofs = sclip<13>(self.regs.mode7_voffset);
signed y = (self.regs.mode7_vflip == false ? self.vcounter() : 255 - self.vcounter());
Update to v092r09 release. byuu says: This will be another massive diff from the previous version. All of higan was updated to use the new foo& bar syntax, and I also updated switch statements to be consistent as well (but not in the disassemblers, was starting to get an RSI just from what I already did.) phoenix/{windows, cocoa, qt} need to be updated to use "string foo" instead of "const string& foo", and after that, the major diffs should be finished. This archive is the first time I'm posting my copy-on-write, size+capacity nall::string class, so any feedback on that is welcome as well.
2013-05-05 09:21:30 +00:00
uint16* mosaic_x;
uint16* mosaic_y;
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
if(id == ID::BG1) {
mosaic_x = mosaic_table[self.bg1.regs.mosaic];
mosaic_y = mosaic_table[self.bg1.regs.mosaic];
} else {
mosaic_x = mosaic_table[self.bg2.regs.mosaic];
mosaic_y = mosaic_table[self.bg1.regs.mosaic];
}
Update to v068r15 release. byuu says: This adds a new "Effect Toggle" window to the tools tab list. There's no menu entry for it yet, oversight, but you can go to another tool and tab to it. The effect toggle window lets you toggle background/OAM layers on a per-priority basis and DSP sound channels. This only works with the compatibility and performance cores, because I'm not going to allow accuracy-violating hacks like that into the core. Non-essential icons have been removed, there's six of them left at the moment. I am too much of a consistency guy to only have some scattered around. I know other apps do that, but I'm not going to do that, and I grow tired of trying to hammer in icons that don't really represent the actions. Anyway, it still looks pretty good I think.
2010-09-06 11:13:51 +00:00
unsigned priority0 = (priority0_enable ? regs.priority0 : 0);
unsigned priority1 = (priority1_enable ? regs.priority1 : 0);
if(priority0 + priority1 == 0) return;
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
signed psx = ((a * Clip(hofs - cx)) & ~63) + ((b * Clip(vofs - cy)) & ~63) + ((b * mosaic_y[y]) & ~63) + (cx << 8);
signed psy = ((c * Clip(hofs - cx)) & ~63) + ((d * Clip(vofs - cy)) & ~63) + ((d * mosaic_y[y]) & ~63) + (cy << 8);
for(signed x = 0; x < 256; x++) {
px = (psx + (a * mosaic_x[x])) >> 8;
py = (psy + (c * mosaic_x[x])) >> 8;
switch(self.regs.mode7_repeat) {
Update to v092r09 release. byuu says: This will be another massive diff from the previous version. All of higan was updated to use the new foo& bar syntax, and I also updated switch statements to be consistent as well (but not in the disassemblers, was starting to get an RSI just from what I already did.) phoenix/{windows, cocoa, qt} need to be updated to use "string foo" instead of "const string& foo", and after that, the major diffs should be finished. This archive is the first time I'm posting my copy-on-write, size+capacity nall::string class, so any feedback on that is welcome as well.
2013-05-05 09:21:30 +00:00
case 0: case 1: {
px &= 1023;
py &= 1023;
tx = ((px >> 3) & 127);
ty = ((py >> 3) & 127);
tile = ppu.vram[(ty * 128 + tx) << 1];
palette = ppu.vram[(((tile << 6) + ((py & 7) << 3) + (px & 7)) << 1) + 1];
break;
}
case 2: {
if((px | py) & ~1023) {
palette = 0;
} else {
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
px &= 1023;
py &= 1023;
tx = ((px >> 3) & 127);
ty = ((py >> 3) & 127);
Update to v074r11 release. byuu says: Changelog: - debugger compiles on all three profiles - libsnes compiles on all three platforms (no API changes to libsnes) - memory.cpp : namespace memory removed (wram -> cpu, apuram -> smp, vram, oam, cgram -> ppu) - sa1.cpp : namespace memory removed (SA-1 specific functions merged inline to SA1::bus_read,write) - GameBoy: added serial link support with interrupts and proper 8192hz timing, but obviously it acts as if no other GB is connected to it - GameBoy: added STAT OAM interrupt, and better STAT d1,d0 mode values - UI: since Qt is dead, I've renamed the config files back to bsnes.cfg and bsnes-geometry.cfg - SA1: IRAM was not syncing to CPU on SA-1 side - PPU/Accuracy and PPU/Performance needed Sprite oam renamed to Sprite sprite; so that I could add uint8 oam[544] - makes more sense anyway, OAM = object attribute memory, obj or sprite are better names for Sprite rendering class - more cleanup
2011-01-24 09:03:17 +00:00
tile = ppu.vram[(ty * 128 + tx) << 1];
palette = ppu.vram[(((tile << 6) + ((py & 7) << 3) + (px & 7)) << 1) + 1];
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
}
Update to v092r09 release. byuu says: This will be another massive diff from the previous version. All of higan was updated to use the new foo& bar syntax, and I also updated switch statements to be consistent as well (but not in the disassemblers, was starting to get an RSI just from what I already did.) phoenix/{windows, cocoa, qt} need to be updated to use "string foo" instead of "const string& foo", and after that, the major diffs should be finished. This archive is the first time I'm posting my copy-on-write, size+capacity nall::string class, so any feedback on that is welcome as well.
2013-05-05 09:21:30 +00:00
break;
}
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
Update to v092r09 release. byuu says: This will be another massive diff from the previous version. All of higan was updated to use the new foo& bar syntax, and I also updated switch statements to be consistent as well (but not in the disassemblers, was starting to get an RSI just from what I already did.) phoenix/{windows, cocoa, qt} need to be updated to use "string foo" instead of "const string& foo", and after that, the major diffs should be finished. This archive is the first time I'm posting my copy-on-write, size+capacity nall::string class, so any feedback on that is welcome as well.
2013-05-05 09:21:30 +00:00
case 3: {
if((px | py) & ~1023) {
tile = 0;
} else {
px &= 1023;
py &= 1023;
tx = ((px >> 3) & 127);
ty = ((py >> 3) & 127);
tile = ppu.vram[(ty * 128 + tx) << 1];
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
}
Update to v092r09 release. byuu says: This will be another massive diff from the previous version. All of higan was updated to use the new foo& bar syntax, and I also updated switch statements to be consistent as well (but not in the disassemblers, was starting to get an RSI just from what I already did.) phoenix/{windows, cocoa, qt} need to be updated to use "string foo" instead of "const string& foo", and after that, the major diffs should be finished. This archive is the first time I'm posting my copy-on-write, size+capacity nall::string class, so any feedback on that is welcome as well.
2013-05-05 09:21:30 +00:00
palette = ppu.vram[(((tile << 6) + ((py & 7) << 3) + (px & 7)) << 1) + 1];
break;
}
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
}
unsigned priority;
if(id == ID::BG1) {
Update to v068r15 release. byuu says: This adds a new "Effect Toggle" window to the tools tab list. There's no menu entry for it yet, oversight, but you can go to another tool and tab to it. The effect toggle window lets you toggle background/OAM layers on a per-priority basis and DSP sound channels. This only works with the compatibility and performance cores, because I'm not going to allow accuracy-violating hacks like that into the core. Non-essential icons have been removed, there's six of them left at the moment. I am too much of a consistency guy to only have some scattered around. I know other apps do that, but I'm not going to do that, and I grow tired of trying to hammer in icons that don't really represent the actions. Anyway, it still looks pretty good I think.
2010-09-06 11:13:51 +00:00
priority = priority0;
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
} else {
Update to v068r15 release. byuu says: This adds a new "Effect Toggle" window to the tools tab list. There's no menu entry for it yet, oversight, but you can go to another tool and tab to it. The effect toggle window lets you toggle background/OAM layers on a per-priority basis and DSP sound channels. This only works with the compatibility and performance cores, because I'm not going to allow accuracy-violating hacks like that into the core. Non-essential icons have been removed, there's six of them left at the moment. I am too much of a consistency guy to only have some scattered around. I know other apps do that, but I'm not going to do that, and I grow tired of trying to hammer in icons that don't really represent the actions. Anyway, it still looks pretty good I think.
2010-09-06 11:13:51 +00:00
priority = (palette & 0x80 ? priority1 : priority0);
Update to v068r10 release. (there was no r09 release posted to the WIP thread) byuu says: It is feature-complete, but horizontal mosaic is less accurate. I have an idea for a mosaic color ring buffer to get it equally accurate, but I haven't implemented it yet. For now it's just a simple x & ~(mosaic >> 1) trick that is passable. Hires blending was left out, as it's more processor intensive and blargg's NTSC does a better job with that anyway. There's some OPT vertical positioning issues in the SNES Test Program's character test; Goodbye, Anthrox has some sort of fast CPU DMA issue; etc. Total speedup is a mere 13.5%. Not quite the 50% I wanted in the best case, but I'll take what I can get. 254->289fps in Zelda 3 on my E8400 now. There's another 15% hiding with blargg's SMP and 5-10% with blargg's fast DSP, but they lose too much accuracy. It'd put me at or below Snes9X accuracy, while still being 50% slower. SSE2 was performing worse this time, both on x86 and amd64, so I left that optimization off. So, barring a miracle, this is about the best it's going to get.
2010-09-03 11:37:36 +00:00
palette &= 0x7f;
}
if(palette == 0) continue;
unsigned plot_x = (self.regs.mode7_hflip == false ? x : 255 - x);
unsigned color;
if(self.screen.regs.direct_color && id == ID::BG1) {
color = self.screen.get_direct_color(0, palette);
} else {
color = self.screen.get_palette(palette);
}
if(regs.main_enable && !window.main[plot_x]) self.screen.output.plot_main(plot_x, color, priority, id);
if(regs.sub_enable && !window.sub[plot_x]) self.screen.output.plot_sub(plot_x, color, priority, id);
}
}
#undef Clip
#endif