mirror of
https://github.com/TomHarte/CLK.git
synced 2024-11-26 23:52:26 +00:00
1037 lines
34 KiB
C++
1037 lines
34 KiB
C++
//
|
|
// 9918.cpp
|
|
// Clock Signal
|
|
//
|
|
// Created by Thomas Harte on 25/11/2017.
|
|
// Copyright 2017 Thomas Harte. All rights reserved.
|
|
//
|
|
|
|
#include "9918.hpp"
|
|
|
|
#include <cassert>
|
|
#include <cstring>
|
|
#include <cstdlib>
|
|
#include "../../Outputs/Log.hpp"
|
|
|
|
using namespace TI::TMS;
|
|
|
|
namespace {
|
|
|
|
constexpr uint8_t StatusInterrupt = 0x80;
|
|
constexpr uint8_t StatusSpriteOverflow = 0x40;
|
|
|
|
constexpr int StatusSpriteCollisionShift = 5;
|
|
constexpr uint8_t StatusSpriteCollision = 0x20;
|
|
|
|
// 342 internal cycles are 228/227.5ths of a line, so 341.25 cycles should be a whole
|
|
// line. Therefore multiply everything by four, but set line length to 1365 rather than 342*4 = 1368.
|
|
constexpr unsigned int CRTCyclesPerLine = 1365;
|
|
constexpr unsigned int CRTCyclesDivider = 4;
|
|
|
|
struct ReverseTable {
|
|
std::uint8_t map[256];
|
|
|
|
ReverseTable() {
|
|
for(int c = 0; c < 256; ++c) {
|
|
map[c] = static_cast<uint8_t>(
|
|
((c & 0x80) >> 7) |
|
|
((c & 0x40) >> 5) |
|
|
((c & 0x20) >> 3) |
|
|
((c & 0x10) >> 1) |
|
|
((c & 0x08) << 1) |
|
|
((c & 0x04) << 3) |
|
|
((c & 0x02) << 5) |
|
|
((c & 0x01) << 7)
|
|
);
|
|
}
|
|
}
|
|
} reverse_table;
|
|
|
|
}
|
|
|
|
Base::Base(Personality p) :
|
|
personality_(p),
|
|
crt_(CRTCyclesPerLine, CRTCyclesDivider, Outputs::Display::Type::NTSC60, Outputs::Display::InputDataType::Red8Green8Blue8) {
|
|
// Unimaginatively, this class just passes RGB through to the shader. Investigation is needed
|
|
// into whether there's a more natural form. It feels unlikely given the diversity of chips modelled.
|
|
|
|
switch(p) {
|
|
case TI::TMS::TMS9918A:
|
|
case TI::TMS::SMSVDP:
|
|
case TI::TMS::SMS2VDP:
|
|
case TI::TMS::GGVDP:
|
|
ram_.resize(16 * 1024);
|
|
break;
|
|
case TI::TMS::V9938:
|
|
ram_.resize(128 * 1024);
|
|
break;
|
|
case TI::TMS::V9958:
|
|
ram_.resize(192 * 1024);
|
|
break;
|
|
}
|
|
|
|
if(is_sega_vdp(personality_)) {
|
|
mode_timing_.line_interrupt_position = 64;
|
|
|
|
mode_timing_.end_of_frame_interrupt_position.column = 63;
|
|
mode_timing_.end_of_frame_interrupt_position.row = 193;
|
|
}
|
|
|
|
// Establish that output is delayed after reading by `output_lag` cycles; start
|
|
// at a random position.
|
|
read_pointer_.row = rand() % 262;
|
|
read_pointer_.column = rand() % (342 - output_lag);
|
|
write_pointer_.row = read_pointer_.row;
|
|
write_pointer_.column = read_pointer_.column + output_lag;
|
|
}
|
|
|
|
TMS9918::TMS9918(Personality p):
|
|
Base(p) {
|
|
crt_.set_display_type(Outputs::Display::DisplayType::RGB);
|
|
crt_.set_visible_area(Outputs::Display::Rect(0.07f, 0.0375f, 0.875f, 0.875f));
|
|
|
|
// The TMS remains in-phase with the NTSC colour clock; this is an empirical measurement
|
|
// intended to produce the correct relationship between the hard edges between pixels and
|
|
// the colour clock. It was eyeballed rather than derived from any knowledge of the TMS
|
|
// colour burst generator because I've yet to find any.
|
|
crt_.set_immediate_default_phase(0.85f);
|
|
}
|
|
|
|
void TMS9918::set_tv_standard(TVStandard standard) {
|
|
tv_standard_ = standard;
|
|
switch(standard) {
|
|
case TVStandard::PAL:
|
|
mode_timing_.total_lines = 313;
|
|
mode_timing_.first_vsync_line = 253;
|
|
crt_.set_new_display_type(CRTCyclesPerLine, Outputs::Display::Type::PAL50);
|
|
break;
|
|
default:
|
|
mode_timing_.total_lines = 262;
|
|
mode_timing_.first_vsync_line = 227;
|
|
crt_.set_new_display_type(CRTCyclesPerLine, Outputs::Display::Type::NTSC60);
|
|
break;
|
|
}
|
|
}
|
|
|
|
void TMS9918::set_scan_target(Outputs::Display::ScanTarget *scan_target) {
|
|
crt_.set_scan_target(scan_target);
|
|
}
|
|
|
|
void TMS9918::set_display_type(Outputs::Display::DisplayType display_type) {
|
|
crt_.set_display_type(display_type);
|
|
}
|
|
|
|
void Base::LineBuffer::reset_sprite_collection() {
|
|
sprites_stopped = false;
|
|
active_sprite_slot = 0;
|
|
|
|
for(int c = 0; c < 8; ++c) {
|
|
active_sprites[c].shift_position = 0;
|
|
}
|
|
}
|
|
|
|
void Base::posit_sprite(LineBuffer &buffer, int sprite_number, int sprite_position, int screen_row) {
|
|
if(!(status_ & StatusSpriteOverflow)) {
|
|
status_ = static_cast<uint8_t>((status_ & ~0x1f) | (sprite_number & 0x1f));
|
|
}
|
|
if(buffer.sprites_stopped)
|
|
return;
|
|
|
|
// A sprite Y of 208 means "don't scan the list any further".
|
|
if(mode_timing_.allow_sprite_terminator && sprite_position == mode_timing_.sprite_terminator) {
|
|
buffer.sprites_stopped = true;
|
|
return;
|
|
}
|
|
|
|
const int sprite_row = (((screen_row + 1) % mode_timing_.total_lines) - ((sprite_position + 1) & 255)) & 255;
|
|
if(sprite_row < 0 || sprite_row >= sprite_height_) return;
|
|
|
|
if(buffer.active_sprite_slot == mode_timing_.maximum_visible_sprites) {
|
|
status_ |= StatusSpriteOverflow;
|
|
return;
|
|
}
|
|
|
|
LineBuffer::ActiveSprite &sprite = buffer.active_sprites[buffer.active_sprite_slot];
|
|
sprite.index = sprite_number;
|
|
sprite.row = sprite_row >> (sprites_magnified_ ? 1 : 0);
|
|
++buffer.active_sprite_slot;
|
|
}
|
|
|
|
void TMS9918::run_for(const HalfCycles cycles) {
|
|
// As specific as I've been able to get:
|
|
// Scanline time is always 228 cycles.
|
|
// PAL output is 313 lines total. NTSC output is 262 lines total.
|
|
// Interrupt is signalled upon entering the lower border.
|
|
|
|
// Convert 456 clocked half cycles per line to 342 internal cycles per line;
|
|
// the internal clock is 1.5 times the nominal 3.579545 Mhz that I've advertised
|
|
// for this part. So multiply by three quarters.
|
|
int int_cycles = int(cycles.as_integral() * 3) + cycles_error_;
|
|
cycles_error_ = int_cycles & 3;
|
|
int_cycles >>= 2;
|
|
if(!int_cycles) return;
|
|
|
|
// There are two intertwined processes here, 'writing' (which means writing to the
|
|
// line buffers, i.e. it's everything to do with collecting a line) and 'reading'
|
|
// (which means reading from the line buffers and generating video).
|
|
int write_cycles_pool = int_cycles;
|
|
int read_cycles_pool = int_cycles;
|
|
|
|
while(write_cycles_pool || read_cycles_pool) {
|
|
LineBufferPointer backup = read_pointer_;
|
|
|
|
if(write_cycles_pool) {
|
|
// Determine how much writing to do.
|
|
const int write_cycles = std::min(342 - write_pointer_.column, write_cycles_pool);
|
|
const int end_column = write_pointer_.column + write_cycles;
|
|
LineBuffer &line_buffer = line_buffers_[write_pointer_.row];
|
|
|
|
// Determine what this does to any enqueued VRAM access.
|
|
minimum_access_column_ = write_pointer_.column + cycles_until_access_;
|
|
cycles_until_access_ -= write_cycles;
|
|
|
|
|
|
// ---------------------------------------
|
|
// Latch scrolling position, if necessary.
|
|
// ---------------------------------------
|
|
if(is_sega_vdp(personality_)) {
|
|
if(write_pointer_.column < 61 && end_column >= 61) {
|
|
if(!write_pointer_.row) {
|
|
master_system_.latched_vertical_scroll = master_system_.vertical_scroll;
|
|
|
|
if(master_system_.mode4_enable) {
|
|
mode_timing_.pixel_lines = 192;
|
|
if(mode2_enable_ && mode1_enable_) mode_timing_.pixel_lines = 224;
|
|
if(mode2_enable_ && mode3_enable_) mode_timing_.pixel_lines = 240;
|
|
|
|
mode_timing_.allow_sprite_terminator = mode_timing_.pixel_lines == 192;
|
|
mode_timing_.first_vsync_line = (mode_timing_.total_lines + mode_timing_.pixel_lines) >> 1;
|
|
|
|
mode_timing_.end_of_frame_interrupt_position.row = mode_timing_.pixel_lines + 1;
|
|
}
|
|
}
|
|
line_buffer.latched_horizontal_scroll = master_system_.horizontal_scroll;
|
|
}
|
|
}
|
|
|
|
|
|
|
|
// ------------------------
|
|
// Perform memory accesses.
|
|
// ------------------------
|
|
#define fetch(function) \
|
|
if(final_window != 171) { \
|
|
function<true>(first_window, final_window);\
|
|
} else {\
|
|
function<false>(first_window, final_window);\
|
|
}
|
|
|
|
// column_ and end_column are in 342-per-line cycles;
|
|
// adjust them to a count of windows.
|
|
const int first_window = write_pointer_.column >> 1;
|
|
const int final_window = end_column >> 1;
|
|
if(first_window != final_window) {
|
|
switch(line_buffer.line_mode) {
|
|
case LineMode::Text: fetch(fetch_tms_text); break;
|
|
case LineMode::Character: fetch(fetch_tms_character); break;
|
|
case LineMode::SMS: fetch(fetch_sms); break;
|
|
case LineMode::Refresh: fetch(fetch_tms_refresh); break;
|
|
}
|
|
}
|
|
|
|
#undef fetch
|
|
|
|
|
|
|
|
// -------------------------------
|
|
// Check for interrupt conditions.
|
|
// -------------------------------
|
|
if(write_pointer_.column < mode_timing_.line_interrupt_position && end_column >= mode_timing_.line_interrupt_position) {
|
|
// The Sega VDP offers a decrementing counter for triggering line interrupts;
|
|
// it is reloaded either when it overflows or upon every non-pixel line after the first.
|
|
// It is otherwise decremented.
|
|
if(is_sega_vdp(personality_)) {
|
|
if(write_pointer_.row >= 0 && write_pointer_.row <= mode_timing_.pixel_lines) {
|
|
--line_interrupt_counter;
|
|
if(line_interrupt_counter == 0xff) {
|
|
line_interrupt_pending_ = true;
|
|
line_interrupt_counter = line_interrupt_target;
|
|
}
|
|
} else {
|
|
line_interrupt_counter = line_interrupt_target;
|
|
}
|
|
}
|
|
|
|
// TODO: the V9938 provides line interrupts from direct specification of the target line.
|
|
// So life is easy.
|
|
}
|
|
|
|
if(
|
|
write_pointer_.row == mode_timing_.end_of_frame_interrupt_position.row &&
|
|
write_pointer_.column < mode_timing_.end_of_frame_interrupt_position.column &&
|
|
end_column >= mode_timing_.end_of_frame_interrupt_position.column
|
|
) {
|
|
status_ |= StatusInterrupt;
|
|
}
|
|
|
|
|
|
|
|
// -------------
|
|
// Advance time.
|
|
// -------------
|
|
write_pointer_.column = end_column;
|
|
write_cycles_pool -= write_cycles;
|
|
|
|
if(write_pointer_.column == 342) {
|
|
write_pointer_.column = 0;
|
|
write_pointer_.row = (write_pointer_.row + 1) % mode_timing_.total_lines;
|
|
LineBuffer &next_line_buffer = line_buffers_[write_pointer_.row];
|
|
|
|
// Establish the output mode for the next line.
|
|
set_current_screen_mode();
|
|
|
|
// Based on the output mode, pick a line mode.
|
|
next_line_buffer.first_pixel_output_column = 86;
|
|
next_line_buffer.next_border_column = 342;
|
|
mode_timing_.maximum_visible_sprites = 4;
|
|
switch(screen_mode_) {
|
|
case ScreenMode::Text:
|
|
next_line_buffer.line_mode = LineMode::Text;
|
|
next_line_buffer.first_pixel_output_column = 94;
|
|
next_line_buffer.next_border_column = 334;
|
|
break;
|
|
case ScreenMode::SMSMode4:
|
|
next_line_buffer.line_mode = LineMode::SMS;
|
|
mode_timing_.maximum_visible_sprites = 8;
|
|
break;
|
|
default:
|
|
next_line_buffer.line_mode = LineMode::Character;
|
|
break;
|
|
}
|
|
|
|
if(
|
|
(screen_mode_ == ScreenMode::Blank) ||
|
|
(write_pointer_.row >= mode_timing_.pixel_lines && write_pointer_.row != mode_timing_.total_lines-1))
|
|
next_line_buffer.line_mode = LineMode::Refresh;
|
|
}
|
|
}
|
|
|
|
|
|
assert(backup.row == read_pointer_.row && backup.column == read_pointer_.column);
|
|
backup = write_pointer_;
|
|
|
|
|
|
if(read_cycles_pool) {
|
|
// Determine how much time has passed in the remainder of this line, and proceed.
|
|
const int target_read_cycles = std::min(342 - read_pointer_.column, read_cycles_pool);
|
|
int read_cycles_performed = 0;
|
|
uint32_t next_cram_value = 0;
|
|
|
|
while(read_cycles_performed < target_read_cycles) {
|
|
const uint32_t cram_value = next_cram_value;
|
|
next_cram_value = 0;
|
|
int read_cycles = target_read_cycles - read_cycles_performed;
|
|
if(!upcoming_cram_dots_.empty() && upcoming_cram_dots_.front().location.row == read_pointer_.row) {
|
|
int time_until_dot = upcoming_cram_dots_.front().location.column - read_pointer_.column;
|
|
|
|
if(time_until_dot < read_cycles) {
|
|
read_cycles = time_until_dot;
|
|
next_cram_value = upcoming_cram_dots_.front().value;
|
|
upcoming_cram_dots_.erase(upcoming_cram_dots_.begin());
|
|
}
|
|
}
|
|
|
|
if(!read_cycles) continue;
|
|
read_cycles_performed += read_cycles;
|
|
|
|
const int end_column = read_pointer_.column + read_cycles;
|
|
LineBuffer &line_buffer = line_buffers_[read_pointer_.row];
|
|
|
|
|
|
// --------------------
|
|
// Output video stream.
|
|
// --------------------
|
|
|
|
#define intersect(left, right, code) { \
|
|
const int start = std::max(read_pointer_.column, left); \
|
|
const int end = std::min(end_column, right); \
|
|
if(end > start) {\
|
|
code;\
|
|
}\
|
|
}
|
|
|
|
#define border(left, right) intersect(left, right, output_border(end - start, cram_value))
|
|
|
|
if(line_buffer.line_mode == LineMode::Refresh || read_pointer_.row > mode_timing_.pixel_lines) {
|
|
if(read_pointer_.row >= mode_timing_.first_vsync_line && read_pointer_.row < mode_timing_.first_vsync_line+4) {
|
|
// Vertical sync.
|
|
if(end_column == 342) {
|
|
crt_.output_sync(342 * 4);
|
|
}
|
|
} else {
|
|
// Right border.
|
|
border(0, 15);
|
|
|
|
// Blanking region; total length is 58 cycles,
|
|
// and 58+15 = 73. So output the lot when the
|
|
// cursor passes 73.
|
|
if(read_pointer_.column < 73 && end_column >= 73) {
|
|
crt_.output_blank(8*4);
|
|
crt_.output_sync(26*4);
|
|
crt_.output_blank(2*4);
|
|
crt_.output_default_colour_burst(14*4);
|
|
crt_.output_blank(8*4);
|
|
}
|
|
|
|
// Border colour for the rest of the line.
|
|
border(73, 342);
|
|
}
|
|
} else {
|
|
// Right border.
|
|
border(0, 15);
|
|
|
|
// Blanking region.
|
|
if(read_pointer_.column < 73 && end_column >= 73) {
|
|
crt_.output_blank(8*4);
|
|
crt_.output_sync(26*4);
|
|
crt_.output_blank(2*4);
|
|
crt_.output_default_colour_burst(14*4);
|
|
crt_.output_blank(8*4);
|
|
}
|
|
|
|
// Left border.
|
|
border(73, line_buffer.first_pixel_output_column);
|
|
|
|
// Pixel region.
|
|
intersect(
|
|
line_buffer.first_pixel_output_column,
|
|
line_buffer.next_border_column,
|
|
if(!asked_for_write_area_) {
|
|
asked_for_write_area_ = true;
|
|
pixel_origin_ = pixel_target_ = reinterpret_cast<uint32_t *>(
|
|
crt_.begin_data(size_t(line_buffer.next_border_column - line_buffer.first_pixel_output_column))
|
|
);
|
|
}
|
|
|
|
if(pixel_target_) {
|
|
const int relative_start = start - line_buffer.first_pixel_output_column;
|
|
const int relative_end = end - line_buffer.first_pixel_output_column;
|
|
switch(line_buffer.line_mode) {
|
|
case LineMode::SMS: draw_sms(relative_start, relative_end, cram_value); break;
|
|
case LineMode::Character: draw_tms_character(relative_start, relative_end); break;
|
|
case LineMode::Text: draw_tms_text(relative_start, relative_end); break;
|
|
|
|
case LineMode::Refresh: break; /* Dealt with elsewhere. */
|
|
}
|
|
}
|
|
|
|
if(end == line_buffer.next_border_column) {
|
|
const int length = line_buffer.next_border_column - line_buffer.first_pixel_output_column;
|
|
crt_.output_data(length * 4, size_t(length));
|
|
pixel_origin_ = pixel_target_ = nullptr;
|
|
asked_for_write_area_ = false;
|
|
}
|
|
);
|
|
|
|
// Additional right border, if called for.
|
|
if(line_buffer.next_border_column != 342) {
|
|
border(line_buffer.next_border_column, 342);
|
|
}
|
|
}
|
|
|
|
#undef border
|
|
#undef intersect
|
|
|
|
|
|
|
|
// -------------
|
|
// Advance time.
|
|
// -------------
|
|
read_pointer_.column = end_column;
|
|
}
|
|
|
|
read_cycles_pool -= target_read_cycles;
|
|
if(read_pointer_.column == 342) {
|
|
read_pointer_.column = 0;
|
|
read_pointer_.row = (read_pointer_.row + 1) % mode_timing_.total_lines;
|
|
}
|
|
}
|
|
|
|
assert(backup.row == write_pointer_.row && backup.column == write_pointer_.column);
|
|
}
|
|
}
|
|
|
|
void Base::output_border(int cycles, uint32_t cram_dot) {
|
|
cycles *= 4;
|
|
uint32_t border_colour =
|
|
is_sega_vdp(personality_) ?
|
|
master_system_.colour_ram[16 + background_colour_] :
|
|
palette[background_colour_];
|
|
|
|
if(cram_dot) {
|
|
uint32_t *const pixel_target = reinterpret_cast<uint32_t *>(crt_.begin_data(1));
|
|
if(pixel_target) {
|
|
*pixel_target = border_colour | cram_dot;
|
|
}
|
|
crt_.output_level(4);
|
|
cycles -= 4;
|
|
}
|
|
|
|
if(cycles) {
|
|
// If the border colour is 0, that can be communicated
|
|
// more efficiently as an explicit blank.
|
|
if(border_colour) {
|
|
uint32_t *const pixel_target = reinterpret_cast<uint32_t *>(crt_.begin_data(1));
|
|
if(pixel_target) {
|
|
*pixel_target = border_colour;
|
|
}
|
|
crt_.output_level(cycles);
|
|
} else {
|
|
crt_.output_blank(cycles);
|
|
}
|
|
}
|
|
}
|
|
|
|
void TMS9918::set_register(int address, uint8_t value) {
|
|
// Writes to address 0 are writes to the video RAM. Store
|
|
// the value and return.
|
|
if(!(address & 1)) {
|
|
write_phase_ = false;
|
|
|
|
// Enqueue the write to occur at the next available slot.
|
|
read_ahead_buffer_ = value;
|
|
queued_access_ = MemoryAccess::Write;
|
|
cycles_until_access_ = vram_access_delay();
|
|
|
|
return;
|
|
}
|
|
|
|
// Writes to address 1 are performed in pairs; if this is the
|
|
// low byte of a value, store it and wait for the high byte.
|
|
if(!write_phase_) {
|
|
low_write_ = value;
|
|
write_phase_ = true;
|
|
|
|
// The initial write should half update the access pointer.
|
|
ram_pointer_ = (ram_pointer_ & 0xff00) | low_write_;
|
|
return;
|
|
}
|
|
|
|
// The RAM pointer is always set on a second write, regardless of
|
|
// whether the caller is intending to enqueue a VDP operation.
|
|
ram_pointer_ = (ram_pointer_ & 0x00ff) | static_cast<uint16_t>(value << 8);
|
|
|
|
write_phase_ = false;
|
|
if(value & 0x80) {
|
|
if(is_sega_vdp(personality_)) {
|
|
if(value & 0x40) {
|
|
master_system_.cram_is_selected = true;
|
|
return;
|
|
}
|
|
value &= 0xf;
|
|
} else {
|
|
value &= 0x7;
|
|
}
|
|
|
|
// This is a write to a register.
|
|
switch(value) {
|
|
case 0:
|
|
if(is_sega_vdp(personality_)) {
|
|
master_system_.vertical_scroll_lock = !!(low_write_ & 0x80);
|
|
master_system_.horizontal_scroll_lock = !!(low_write_ & 0x40);
|
|
master_system_.hide_left_column = !!(low_write_ & 0x20);
|
|
enable_line_interrupts_ = !!(low_write_ & 0x10);
|
|
master_system_.shift_sprites_8px_left = !!(low_write_ & 0x08);
|
|
master_system_.mode4_enable = !!(low_write_ & 0x04);
|
|
}
|
|
mode2_enable_ = !!(low_write_ & 0x02);
|
|
break;
|
|
|
|
case 1:
|
|
blank_display_ = !(low_write_ & 0x40);
|
|
generate_interrupts_ = !!(low_write_ & 0x20);
|
|
mode1_enable_ = !!(low_write_ & 0x10);
|
|
mode3_enable_ = !!(low_write_ & 0x08);
|
|
sprites_16x16_ = !!(low_write_ & 0x02);
|
|
sprites_magnified_ = !!(low_write_ & 0x01);
|
|
|
|
sprite_height_ = 8;
|
|
if(sprites_16x16_) sprite_height_ <<= 1;
|
|
if(sprites_magnified_) sprite_height_ <<= 1;
|
|
break;
|
|
|
|
case 2:
|
|
pattern_name_address_ = size_t((low_write_ & 0xf) << 10) | 0x3ff;
|
|
master_system_.pattern_name_address = pattern_name_address_ | ((personality_ == TMS::SMSVDP) ? 0x000 : 0x400);
|
|
break;
|
|
|
|
case 3:
|
|
colour_table_address_ = size_t(low_write_ << 6) | 0x3f;
|
|
break;
|
|
|
|
case 4:
|
|
pattern_generator_table_address_ = size_t((low_write_ & 0x07) << 11) | 0x7ff;
|
|
break;
|
|
|
|
case 5:
|
|
sprite_attribute_table_address_ = size_t((low_write_ & 0x7f) << 7) | 0x7f;
|
|
master_system_.sprite_attribute_table_address = sprite_attribute_table_address_ | ((personality_ == TMS::SMSVDP) ? 0x00 : 0x80);
|
|
break;
|
|
|
|
case 6:
|
|
sprite_generator_table_address_ = size_t((low_write_ & 0x07) << 11) | 0x7ff;
|
|
master_system_.sprite_generator_table_address = sprite_generator_table_address_ | ((personality_ == TMS::SMSVDP) ? 0x0000 : 0x1800);
|
|
break;
|
|
|
|
case 7:
|
|
text_colour_ = low_write_ >> 4;
|
|
background_colour_ = low_write_ & 0xf;
|
|
break;
|
|
|
|
case 8:
|
|
if(is_sega_vdp(personality_)) {
|
|
master_system_.horizontal_scroll = low_write_;
|
|
}
|
|
break;
|
|
|
|
case 9:
|
|
if(is_sega_vdp(personality_)) {
|
|
master_system_.vertical_scroll = low_write_;
|
|
}
|
|
break;
|
|
|
|
case 10:
|
|
if(is_sega_vdp(personality_)) {
|
|
line_interrupt_target = low_write_;
|
|
}
|
|
break;
|
|
|
|
default:
|
|
LOG("Unknown TMS write: " << int(low_write_) << " to " << int(value));
|
|
break;
|
|
}
|
|
} else {
|
|
// This is an access via the RAM pointer.
|
|
if(!(value & 0x40)) {
|
|
// A read request is enqueued upon setting the address; conversely a write
|
|
// won't be enqueued unless and until some actual data is supplied.
|
|
queued_access_ = MemoryAccess::Read;
|
|
cycles_until_access_ = vram_access_delay();
|
|
}
|
|
master_system_.cram_is_selected = false;
|
|
}
|
|
}
|
|
|
|
uint8_t TMS9918::get_current_line() {
|
|
// Determine the row to return.
|
|
constexpr int row_change_position = 63; // This is the proper Master System value; substitute if any other VDPs turn out to have this functionality.
|
|
int source_row =
|
|
(write_pointer_.column < row_change_position)
|
|
? (write_pointer_.row + mode_timing_.total_lines - 1)%mode_timing_.total_lines
|
|
: write_pointer_.row;
|
|
|
|
if(tv_standard_ == TVStandard::NTSC) {
|
|
if(mode_timing_.pixel_lines == 240) {
|
|
// NTSC 256x240: 00-FF, 00-06
|
|
} else if(mode_timing_.pixel_lines == 224) {
|
|
// NTSC 256x224: 00-EA, E5-FF
|
|
if(source_row >= 0xeb) source_row -= 6;
|
|
} else {
|
|
// NTSC 256x192: 00-DA, D5-FF
|
|
if(source_row >= 0xdb) source_row -= 6;
|
|
}
|
|
} else {
|
|
if(mode_timing_.pixel_lines == 240) {
|
|
// PAL 256x240: 00-FF, 00-0A, D2-FF
|
|
if(source_row >= 267) source_row -= 0x39;
|
|
} else if(mode_timing_.pixel_lines == 224) {
|
|
// PAL 256x224: 00-FF, 00-02, CA-FF
|
|
if(source_row >= 259) source_row -= 0x39;
|
|
} else {
|
|
// PAL 256x192: 00-F2, BA-FF
|
|
if(source_row >= 0xf3) source_row -= 0x39;
|
|
}
|
|
}
|
|
|
|
return static_cast<uint8_t>(source_row);
|
|
}
|
|
|
|
uint8_t TMS9918::get_latched_horizontal_counter() {
|
|
// Translate from internal numbering, which puts pixel output
|
|
// in the final 256 pixels of 342, to the public numbering,
|
|
// which makes the 256 pixels the first 256 spots, but starts
|
|
// counting at -48, and returns only the top 8 bits of the number.
|
|
int public_counter = latched_column_ - 86;
|
|
if(public_counter < -46) public_counter += 342;
|
|
return uint8_t(public_counter >> 1);
|
|
}
|
|
|
|
void TMS9918::latch_horizontal_counter() {
|
|
latched_column_ = write_pointer_.column;
|
|
}
|
|
|
|
uint8_t TMS9918::get_register(int address) {
|
|
write_phase_ = false;
|
|
|
|
// Reads from address 0 read video RAM, via the read-ahead buffer.
|
|
if(!(address & 1)) {
|
|
// Enqueue the write to occur at the next available slot.
|
|
uint8_t result = read_ahead_buffer_;
|
|
queued_access_ = MemoryAccess::Read;
|
|
return result;
|
|
}
|
|
|
|
// Reads from address 1 get the status register.
|
|
uint8_t result = status_;
|
|
status_ &= ~(StatusInterrupt | StatusSpriteOverflow | StatusSpriteCollision);
|
|
line_interrupt_pending_ = false;
|
|
return result;
|
|
}
|
|
|
|
HalfCycles Base::half_cycles_before_internal_cycles(int internal_cycles) {
|
|
return HalfCycles(((internal_cycles << 2) + (2 - cycles_error_)) / 3);
|
|
}
|
|
|
|
HalfCycles TMS9918::get_time_until_interrupt() {
|
|
if(!generate_interrupts_ && !enable_line_interrupts_) return HalfCycles(-1);
|
|
if(get_interrupt_line()) return HalfCycles(0);
|
|
|
|
// Calculate the amount of time until the next end-of-frame interrupt.
|
|
const int frame_length = 342 * mode_timing_.total_lines;
|
|
int time_until_frame_interrupt =
|
|
(
|
|
((mode_timing_.end_of_frame_interrupt_position.row * 342) + mode_timing_.end_of_frame_interrupt_position.column + frame_length) -
|
|
((write_pointer_.row * 342) + write_pointer_.column)
|
|
) % frame_length;
|
|
if(!time_until_frame_interrupt) time_until_frame_interrupt = frame_length;
|
|
|
|
if(!enable_line_interrupts_) return half_cycles_before_internal_cycles(time_until_frame_interrupt);
|
|
|
|
// Calculate when the next line interrupt will occur.
|
|
int next_line_interrupt_row = -1;
|
|
|
|
int cycles_to_next_interrupt_threshold = mode_timing_.line_interrupt_position - write_pointer_.column;
|
|
int line_of_next_interrupt_threshold = write_pointer_.row;
|
|
if(cycles_to_next_interrupt_threshold <= 0) {
|
|
cycles_to_next_interrupt_threshold += 342;
|
|
++line_of_next_interrupt_threshold;
|
|
}
|
|
|
|
if(is_sega_vdp(personality_)) {
|
|
// If there is still time for a line interrupt this frame, that'll be it;
|
|
// otherwise it'll be on the next frame, supposing there's ever time for
|
|
// it at all.
|
|
if(line_of_next_interrupt_threshold + line_interrupt_counter <= mode_timing_.pixel_lines) {
|
|
next_line_interrupt_row = line_of_next_interrupt_threshold + line_interrupt_counter;
|
|
} else {
|
|
if(line_interrupt_target <= mode_timing_.pixel_lines)
|
|
next_line_interrupt_row = mode_timing_.total_lines + line_interrupt_target;
|
|
}
|
|
}
|
|
|
|
// If there's actually no interrupt upcoming, despite being enabled, either return
|
|
// the frame end interrupt or no interrupt pending as appropriate.
|
|
if(next_line_interrupt_row == -1) {
|
|
return generate_interrupts_ ?
|
|
half_cycles_before_internal_cycles(time_until_frame_interrupt) :
|
|
HalfCycles(-1);
|
|
}
|
|
|
|
// Figure out the number of internal cycles until the next line interrupt, which is the amount
|
|
// of time to the next tick over and then next_line_interrupt_row - row_ lines further.
|
|
const int local_cycles_until_line_interrupt = cycles_to_next_interrupt_threshold + (next_line_interrupt_row - line_of_next_interrupt_threshold) * 342;
|
|
if(!generate_interrupts_) return half_cycles_before_internal_cycles(local_cycles_until_line_interrupt);
|
|
|
|
// Return whichever interrupt is closer.
|
|
return half_cycles_before_internal_cycles(std::min(local_cycles_until_line_interrupt, time_until_frame_interrupt));
|
|
}
|
|
|
|
HalfCycles TMS9918::get_time_until_line(int line) {
|
|
if(line < 0) line += mode_timing_.total_lines;
|
|
|
|
int cycles_to_next_interrupt_threshold = mode_timing_.line_interrupt_position - write_pointer_.column;
|
|
int line_of_next_interrupt_threshold = write_pointer_.row;
|
|
if(cycles_to_next_interrupt_threshold <= 0) {
|
|
cycles_to_next_interrupt_threshold += 342;
|
|
++line_of_next_interrupt_threshold;
|
|
}
|
|
|
|
if(line_of_next_interrupt_threshold > line) {
|
|
line += mode_timing_.total_lines;
|
|
}
|
|
|
|
return half_cycles_before_internal_cycles(cycles_to_next_interrupt_threshold + (line - line_of_next_interrupt_threshold)*342);
|
|
}
|
|
|
|
bool TMS9918::get_interrupt_line() {
|
|
return ((status_ & StatusInterrupt) && generate_interrupts_) || (enable_line_interrupts_ && line_interrupt_pending_);
|
|
}
|
|
|
|
// MARK: -
|
|
|
|
void Base::draw_tms_character(int start, int end) {
|
|
LineBuffer &line_buffer = line_buffers_[read_pointer_.row];
|
|
|
|
// Paint the background tiles.
|
|
const int pixels_left = end - start;
|
|
if(screen_mode_ == ScreenMode::MultiColour) {
|
|
for(int c = start; c < end; ++c) {
|
|
pixel_target_[c] = palette[
|
|
(line_buffer.patterns[c >> 3][0] >> (((c & 4)^4))) & 15
|
|
];
|
|
}
|
|
} else {
|
|
const int shift = start & 7;
|
|
int byte_column = start >> 3;
|
|
|
|
int length = std::min(pixels_left, 8 - shift);
|
|
|
|
int pattern = reverse_table.map[line_buffer.patterns[byte_column][0]] >> shift;
|
|
uint8_t colour = line_buffer.patterns[byte_column][1];
|
|
uint32_t colours[2] = {
|
|
palette[(colour & 15) ? (colour & 15) : background_colour_],
|
|
palette[(colour >> 4) ? (colour >> 4) : background_colour_]
|
|
};
|
|
|
|
int background_pixels_left = pixels_left;
|
|
while(true) {
|
|
background_pixels_left -= length;
|
|
for(int c = 0; c < length; ++c) {
|
|
pixel_target_[c] = colours[pattern&0x01];
|
|
pattern >>= 1;
|
|
}
|
|
pixel_target_ += length;
|
|
|
|
if(!background_pixels_left) break;
|
|
length = std::min(8, background_pixels_left);
|
|
byte_column++;
|
|
|
|
pattern = reverse_table.map[line_buffer.patterns[byte_column][0]];
|
|
colour = line_buffer.patterns[byte_column][1];
|
|
colours[0] = palette[(colour & 15) ? (colour & 15) : background_colour_];
|
|
colours[1] = palette[(colour >> 4) ? (colour >> 4) : background_colour_];
|
|
}
|
|
}
|
|
|
|
// Paint sprites and check for collisions, but only if at least one sprite is active
|
|
// on this line.
|
|
if(line_buffer.active_sprite_slot) {
|
|
const int shift_advance = sprites_magnified_ ? 1 : 2;
|
|
// If this is the start of the line clip any part of any sprites that is off to the left.
|
|
if(!start) {
|
|
for(int index = 0; index < line_buffer.active_sprite_slot; ++index) {
|
|
LineBuffer::ActiveSprite &sprite = line_buffer.active_sprites[index];
|
|
if(sprite.x < 0) sprite.shift_position -= shift_advance * sprite.x;
|
|
}
|
|
}
|
|
|
|
int sprite_buffer[256];
|
|
int sprite_collision = 0;
|
|
memset(&sprite_buffer[start], 0, size_t(end - start)*sizeof(sprite_buffer[0]));
|
|
|
|
constexpr uint32_t sprite_colour_selection_masks[2] = {0x00000000, 0xffffffff};
|
|
constexpr int colour_masks[16] = {0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1};
|
|
|
|
// Draw all sprites into the sprite buffer.
|
|
const int shifter_target = sprites_16x16_ ? 32 : 16;
|
|
for(int index = line_buffer.active_sprite_slot - 1; index >= 0; --index) {
|
|
LineBuffer::ActiveSprite &sprite = line_buffer.active_sprites[index];
|
|
if(sprite.shift_position < shifter_target) {
|
|
const int pixel_start = std::max(start, sprite.x);
|
|
for(int c = pixel_start; c < end && sprite.shift_position < shifter_target; ++c) {
|
|
const int shift = (sprite.shift_position >> 1) ^ 7;
|
|
int sprite_colour = (sprite.image[shift >> 3] >> (shift & 7)) & 1;
|
|
|
|
// A colision is detected regardless of sprite colour ...
|
|
sprite_collision |= sprite_buffer[c] & sprite_colour;
|
|
sprite_buffer[c] |= sprite_colour;
|
|
|
|
// ... but a sprite with the transparent colour won't actually be visible.
|
|
sprite_colour &= colour_masks[sprite.image[2]&15];
|
|
pixel_origin_[c] =
|
|
(pixel_origin_[c] & sprite_colour_selection_masks[sprite_colour^1]) |
|
|
(palette[sprite.image[2]&15] & sprite_colour_selection_masks[sprite_colour]);
|
|
|
|
sprite.shift_position += shift_advance;
|
|
}
|
|
}
|
|
}
|
|
|
|
status_ |= sprite_collision << StatusSpriteCollisionShift;
|
|
}
|
|
}
|
|
|
|
void Base::draw_tms_text(int start, int end) {
|
|
LineBuffer &line_buffer = line_buffers_[read_pointer_.row];
|
|
const uint32_t colours[2] = { palette[background_colour_], palette[text_colour_] };
|
|
|
|
const int shift = start % 6;
|
|
int byte_column = start / 6;
|
|
int pattern = reverse_table.map[line_buffer.patterns[byte_column][0]] >> shift;
|
|
int pixels_left = end - start;
|
|
int length = std::min(pixels_left, 6 - shift);
|
|
while(true) {
|
|
pixels_left -= length;
|
|
for(int c = 0; c < length; ++c) {
|
|
pixel_target_[c] = colours[pattern&0x01];
|
|
pattern >>= 1;
|
|
}
|
|
pixel_target_ += length;
|
|
|
|
if(!pixels_left) break;
|
|
length = std::min(6, pixels_left);
|
|
byte_column++;
|
|
pattern = reverse_table.map[line_buffer.patterns[byte_column][0]];
|
|
}
|
|
}
|
|
|
|
void Base::draw_sms(int start, int end, uint32_t cram_dot) {
|
|
LineBuffer &line_buffer = line_buffers_[read_pointer_.row];
|
|
int colour_buffer[256];
|
|
|
|
/*
|
|
Add extra border for any pixels that fall before the fine scroll.
|
|
*/
|
|
int tile_start = start, tile_end = end;
|
|
int tile_offset = start;
|
|
if(read_pointer_.row >= 16 || !master_system_.horizontal_scroll_lock) {
|
|
for(int c = start; c < (line_buffer.latched_horizontal_scroll & 7); ++c) {
|
|
colour_buffer[c] = 16 + background_colour_;
|
|
++tile_offset;
|
|
}
|
|
|
|
// Remove the border area from that to which tiles will be drawn.
|
|
tile_start = std::max(start - (line_buffer.latched_horizontal_scroll & 7), 0);
|
|
tile_end = std::max(end - (line_buffer.latched_horizontal_scroll & 7), 0);
|
|
}
|
|
|
|
|
|
uint32_t pattern;
|
|
uint8_t *const pattern_index = reinterpret_cast<uint8_t *>(&pattern);
|
|
|
|
/*
|
|
Add background tiles; these will fill the colour_buffer with values in which
|
|
the low five bits are a palette index, and bit six is set if this tile has
|
|
priority over sprites.
|
|
*/
|
|
if(tile_start < end) {
|
|
const int shift = tile_start & 7;
|
|
int byte_column = tile_start >> 3;
|
|
int pixels_left = tile_end - tile_start;
|
|
int length = std::min(pixels_left, 8 - shift);
|
|
|
|
pattern = *reinterpret_cast<const uint32_t *>(line_buffer.patterns[byte_column]);
|
|
if(line_buffer.names[byte_column].flags&2)
|
|
pattern >>= shift;
|
|
else
|
|
pattern <<= shift;
|
|
|
|
while(true) {
|
|
const int palette_offset = (line_buffer.names[byte_column].flags&0x18) << 1;
|
|
if(line_buffer.names[byte_column].flags&2) {
|
|
for(int c = 0; c < length; ++c) {
|
|
colour_buffer[tile_offset] =
|
|
((pattern_index[3] & 0x01) << 3) |
|
|
((pattern_index[2] & 0x01) << 2) |
|
|
((pattern_index[1] & 0x01) << 1) |
|
|
((pattern_index[0] & 0x01) << 0) |
|
|
palette_offset;
|
|
++tile_offset;
|
|
pattern >>= 1;
|
|
}
|
|
} else {
|
|
for(int c = 0; c < length; ++c) {
|
|
colour_buffer[tile_offset] =
|
|
((pattern_index[3] & 0x80) >> 4) |
|
|
((pattern_index[2] & 0x80) >> 5) |
|
|
((pattern_index[1] & 0x80) >> 6) |
|
|
((pattern_index[0] & 0x80) >> 7) |
|
|
palette_offset;
|
|
++tile_offset;
|
|
pattern <<= 1;
|
|
}
|
|
}
|
|
|
|
pixels_left -= length;
|
|
if(!pixels_left) break;
|
|
|
|
length = std::min(8, pixels_left);
|
|
byte_column++;
|
|
pattern = *reinterpret_cast<const uint32_t *>(line_buffer.patterns[byte_column]);
|
|
}
|
|
}
|
|
|
|
/*
|
|
Apply sprites (if any).
|
|
*/
|
|
if(line_buffer.active_sprite_slot) {
|
|
const int shift_advance = sprites_magnified_ ? 1 : 2;
|
|
|
|
// If this is the start of the line clip any part of any sprites that is off to the left.
|
|
if(!start) {
|
|
for(int index = 0; index < line_buffer.active_sprite_slot; ++index) {
|
|
LineBuffer::ActiveSprite &sprite = line_buffer.active_sprites[index];
|
|
if(sprite.x < 0) sprite.shift_position -= shift_advance * sprite.x;
|
|
}
|
|
}
|
|
|
|
int sprite_buffer[256];
|
|
int sprite_collision = 0;
|
|
memset(&sprite_buffer[start], 0, size_t(end - start)*sizeof(sprite_buffer[0]));
|
|
|
|
// Draw all sprites into the sprite buffer.
|
|
for(int index = line_buffer.active_sprite_slot - 1; index >= 0; --index) {
|
|
LineBuffer::ActiveSprite &sprite = line_buffer.active_sprites[index];
|
|
if(sprite.shift_position < 16) {
|
|
const int pixel_start = std::max(start, sprite.x);
|
|
|
|
// TODO: it feels like the work below should be simplifiable;
|
|
// the double shift in particular, and hopefully the variable shift.
|
|
for(int c = pixel_start; c < end && sprite.shift_position < 16; ++c) {
|
|
const int shift = (sprite.shift_position >> 1);
|
|
const int sprite_colour =
|
|
(((sprite.image[3] << shift) & 0x80) >> 4) |
|
|
(((sprite.image[2] << shift) & 0x80) >> 5) |
|
|
(((sprite.image[1] << shift) & 0x80) >> 6) |
|
|
(((sprite.image[0] << shift) & 0x80) >> 7);
|
|
|
|
if(sprite_colour) {
|
|
sprite_collision |= sprite_buffer[c];
|
|
sprite_buffer[c] = sprite_colour | 0x10;
|
|
}
|
|
|
|
sprite.shift_position += shift_advance;
|
|
}
|
|
}
|
|
}
|
|
|
|
// Draw the sprite buffer onto the colour buffer, wherever the tile map doesn't have
|
|
// priority (or is transparent).
|
|
for(int c = start; c < end; ++c) {
|
|
if(
|
|
sprite_buffer[c] &&
|
|
(!(colour_buffer[c]&0x20) || !(colour_buffer[c]&0xf))
|
|
) colour_buffer[c] = sprite_buffer[c];
|
|
}
|
|
|
|
if(sprite_collision)
|
|
status_ |= StatusSpriteCollision;
|
|
}
|
|
|
|
// Map from the 32-colour buffer to real output pixels, applying the specific CRAM dot if any.
|
|
pixel_target_[start] = master_system_.colour_ram[colour_buffer[start] & 0x1f] | cram_dot;
|
|
for(int c = start+1; c < end; ++c) {
|
|
pixel_target_[c] = master_system_.colour_ram[colour_buffer[c] & 0x1f];
|
|
}
|
|
|
|
// If the VDP is set to hide the left column and this is the final call that'll come
|
|
// this line, hide it.
|
|
if(end == 256) {
|
|
if(master_system_.hide_left_column) {
|
|
pixel_origin_[0] = pixel_origin_[1] = pixel_origin_[2] = pixel_origin_[3] =
|
|
pixel_origin_[4] = pixel_origin_[5] = pixel_origin_[6] = pixel_origin_[7] =
|
|
master_system_.colour_ram[16 + background_colour_];
|
|
}
|
|
}
|
|
}
|