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

185 lines
5.9 KiB
C++
Raw Normal View History

Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
#ifdef PPU_CPP
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
#include "mode7.cpp"
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
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 PPU::Background::get_tile(unsigned hoffset, unsigned voffset) {
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
unsigned tile_x = (hoffset & mask_x) >> tile_width;
unsigned tile_y = (voffset & mask_y) >> tile_height;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
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 tile_pos = ((tile_y & 0x1f) << 5) + (tile_x & 0x1f);
if(tile_y & 0x20) tile_pos += scy;
if(tile_x & 0x20) tile_pos += scx;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
const uint16 tiledata_addr = regs.screen_addr + (tile_pos << 1);
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
return (ppu.vram[tiledata_addr + 0] << 0) + (ppu.vram[tiledata_addr + 1] << 8);
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 v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
void PPU::Background::offset_per_tile(unsigned x, unsigned y, unsigned &hoffset, unsigned &voffset) {
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 opt_x = (x + (hscroll & 7)), hval, vval;
if(opt_x >= 8) {
hval = self.bg3.get_tile((opt_x - 8) + (self.bg3.regs.hoffset & ~7), self.bg3.regs.voffset + 0);
if(self.regs.bgmode != 4)
vval = self.bg3.get_tile((opt_x - 8) + (self.bg3.regs.hoffset & ~7), self.bg3.regs.voffset + 8);
if(self.regs.bgmode == 4) {
if(hval & opt_valid_bit) {
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(!(hval & 0x8000)) {
hoffset = opt_x + (hval & ~7);
} else {
voffset = y + hval;
}
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 {
if(hval & opt_valid_bit) {
hoffset = opt_x + (hval & ~7);
}
if(vval & opt_valid_bit) {
voffset = y + vval;
}
}
}
}
void PPU::Background::scanline() {
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(self.vcounter() == 1) {
mosaic_vcounter = regs.mosaic + 1;
Update to v070r12 release. byuu says: - removed support for images with copier headers - phoenix/Windows: Label properly refreshes on text changes, fixes video settings sliders - alt/ppu-performance: fixed mosaic Voffset bug, fixes Super Bowling et al - alt/cpu: fixed CPU::joylatch() reporting, allows serial applications to work with performance profile - hooked up SNES::cartridge.basename, allows MSU1 and serial support with the phoenix UI - updated UPS patching code for bsnes/Qt, allowing it to compile again, hidden config option file.bypassPatchCrc32 was removed
2010-10-11 10:36:00 +00:00
mosaic_voffset = 1;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
} else if(--mosaic_vcounter == 0) {
mosaic_vcounter = regs.mosaic + 1;
Update to v070r12 release. byuu says: - removed support for images with copier headers - phoenix/Windows: Label properly refreshes on text changes, fixes video settings sliders - alt/ppu-performance: fixed mosaic Voffset bug, fixes Super Bowling et al - alt/cpu: fixed CPU::joylatch() reporting, allows serial applications to work with performance profile - hooked up SNES::cartridge.basename, allows MSU1 and serial support with the phoenix UI - updated UPS patching code for bsnes/Qt, allowing it to compile again, hidden config option file.bypassPatchCrc32 was removed
2010-10-11 10:36:00 +00:00
mosaic_voffset += regs.mosaic + 1;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
}
if(self.regs.display_disable) 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
hires = (self.regs.bgmode == 5 || self.regs.bgmode == 6);
width = !hires ? 256 : 512;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
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
tile_height = regs.tile_size ? 4 : 3;
tile_width = hires ? 4 : tile_height;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
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
mask_x = (tile_height == 4 ? width << 1 : width);
mask_y = mask_x;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
if(regs.screen_size & 1) mask_x <<= 1;
if(regs.screen_size & 2) mask_y <<= 1;
mask_x--;
mask_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
scx = (regs.screen_size & 1 ? 32 << 5 : 0);
scy = (regs.screen_size & 2 ? 32 << 5 : 0);
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
if(regs.screen_size == 3) scy <<= 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 v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
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
void PPU::Background::render() {
if(regs.mode == Mode::Inactive) return;
if(regs.main_enable == false && regs.sub_enable == false) return;
if(regs.main_enable) window.render(0);
if(regs.sub_enable) window.render(1);
if(regs.mode == Mode::Mode7) return render_mode7();
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 v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
unsigned mosaic_hcounter = 1;
unsigned mosaic_palette = 0;
unsigned mosaic_priority = 0;
unsigned mosaic_color = 0;
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
const unsigned bgpal_index = (self.regs.bgmode == 0 ? id << 5 : 0);
const unsigned pal_size = 2 << regs.mode;
const unsigned tile_mask = 0x0fff >> regs.mode;
const unsigned tiledata_index = regs.tiledata_addr >> (4 + regs.mode);
hscroll = regs.hoffset;
vscroll = regs.voffset;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
Update to v070r12 release. byuu says: - removed support for images with copier headers - phoenix/Windows: Label properly refreshes on text changes, fixes video settings sliders - alt/ppu-performance: fixed mosaic Voffset bug, fixes Super Bowling et al - alt/cpu: fixed CPU::joylatch() reporting, allows serial applications to work with performance profile - hooked up SNES::cartridge.basename, allows MSU1 and serial support with the phoenix UI - updated UPS patching code for bsnes/Qt, allowing it to compile again, hidden config option file.bypassPatchCrc32 was removed
2010-10-11 10:36:00 +00:00
unsigned y = (regs.mosaic == 0 ? self.vcounter() : mosaic_voffset);
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
if(hires) {
hscroll <<= 1;
if(self.regs.interlace) y = (y << 1) + self.field();
}
unsigned tile_pri, tile_num;
unsigned pal_index, pal_num;
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 hoffset, voffset, col;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
bool mirror_x, mirror_y;
const bool is_opt_mode = (self.regs.bgmode == 2 || self.regs.bgmode == 4 || self.regs.bgmode == 6);
const bool is_direct_color_mode = (self.screen.regs.direct_color == true && id == ID::BG1 && (self.regs.bgmode == 3 || self.regs.bgmode == 4));
signed x = 0 - (hscroll & 7);
while(x < width) {
hoffset = x + hscroll;
voffset = y + vscroll;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(is_opt_mode) offset_per_tile(x, y, hoffset, voffset);
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
hoffset &= mask_x;
voffset &= mask_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
tile_num = get_tile(hoffset, voffset);
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
mirror_y = tile_num & 0x8000;
mirror_x = tile_num & 0x4000;
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
tile_pri = tile_num & 0x2000 ? priority1 : priority0;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
pal_num = (tile_num >> 10) & 7;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
pal_index = (bgpal_index + (pal_num << pal_size)) & 0xff;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(tile_width == 4 && (bool)(hoffset & 8) != mirror_x) tile_num += 1;
if(tile_height == 4 && (bool)(voffset & 8) != mirror_y) tile_num += 16;
tile_num = ((tile_num & 0x03ff) + tiledata_index) & tile_mask;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
if(mirror_y) voffset ^= 7;
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
unsigned mirror_xmask = !mirror_x ? 0 : 7;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
uint8 *tiledata = self.cache.tile(regs.mode, tile_num);
tiledata += ((voffset & 7) * 8);
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
for(unsigned n = 0; n < 8; n++, x++) {
if(x & width) continue;
if(--mosaic_hcounter == 0) {
mosaic_hcounter = regs.mosaic + 1;
mosaic_palette = tiledata[n ^ mirror_xmask];
mosaic_priority = tile_pri;
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(is_direct_color_mode) {
Update to v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
mosaic_color = self.screen.get_direct_color(pal_num, mosaic_palette);
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 v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
mosaic_color = self.screen.get_palette(pal_index + mosaic_palette);
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 v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
}
if(mosaic_palette == 0) continue;
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 v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(hires == false) {
if(regs.main_enable && !window.main[x]) self.screen.output.plot_main(x, mosaic_color, mosaic_priority, id);
if(regs.sub_enable && !window.sub[x]) self.screen.output.plot_sub(x, mosaic_color, mosaic_priority, id);
} else {
signed half_x = x >> 1;
if(x & 1) {
if(regs.main_enable && !window.main[half_x]) self.screen.output.plot_main(half_x, mosaic_color, mosaic_priority, id);
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 v068r13 release. byuu says: Bug-fix night for the new PPUs. Accuracy: Fixed BG palette clamping, which fixes Taz-Mania. Added blocking for CGRAM writes during active display, to match the compatibility core. It really should override to the last fetched palette color, I'll probably try that out soon enough. Performance: Mosaic should match the other renderers. Unfortunately, as suspected, it murders speed. 290->275fps. It's now only 11fps faster, hardly worth it at all. But the old rendering code is really awful, so maybe it's for the best it gets refreshed. It's really tough to understand why this is such a performance hit, it's just a decrement+compare check four times per pixel. But yeah, it hits it really, really hard. Fixed a missing check in Mode4 offset-per-tile, fixes vertical alignment of a test image in the SNES Test Program.
2010-09-05 13:22:26 +00:00
if(regs.sub_enable && !window.sub[half_x]) self.screen.output.plot_sub(half_x, mosaic_color, mosaic_priority, id);
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 v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
}
}
}
}
PPU::Background::Background(PPU &self, unsigned id) : self(self), id(id) {
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
priority0_enable = true;
priority1_enable = true;
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
opt_valid_bit = (id == ID::BG1 ? 0x2000 : id == ID::BG2 ? 0x4000 : 0x0000);
mosaic_table = new uint16*[16];
for(unsigned m = 0; m < 16; m++) {
mosaic_table[m] = new uint16[4096];
for(unsigned x = 0; x < 4096; x++) {
mosaic_table[m][x] = (x / (m + 1)) * (m + 1);
}
}
}
PPU::Background::~Background() {
for(unsigned m = 0; m < 16; m++) delete[] mosaic_table[m];
delete[] mosaic_table;
Update to v068r07 release. (there was no r06 release posted to the WIP thread) byuu says: New PPU renderer is coming along. Lots of new ideas, especially with regards to the way the background renders in tiles rather than in pixels. That skips the need for tile caching and compares, and it even gets scrolling right. The sprite item+tile lists are computing, but they are not rendering yet. It's a good deal faster for now, obviously, because 90% of the PPU features are missing.
2010-09-01 13:20:05 +00:00
}
#endif