mirror of
https://github.com/MiSTer-devel/Arcade-Dcon_MiSTer.git
synced 2026-05-31 03:02:21 +00:00
3ad6536176
D-Con (Success, 1992) — FPGA core for the MiSTer FPGA platform (Terasic DE10-Nano). Reimplements the original Seibu hardware in SystemVerilog from MAME (seibu/dcon.cpp, seibu/seibu_crtc.cpp), hardware documentation, main 68000 ROM disassembly and observation of real PCB behavior. Hardware emulated: - M68000 main CPU @ 10 MHz (FX68K cycle-accurate) - Z80 sound CPU @ 4 MHz (T80s) with audio-rate multicycle SDC - YM3812 OPL2 FM (jtopl2) + OKI M6295 ADPCM (jt6295) - 320x224 active video area (Seibu D-Con timings) - BG / MG / FG tilemaps + Text layer + 16x16 sprites (SEI0211) - Seibu CRTC registers (scroll, layer enable, flip screen) - xBGR_555 palette, 2048 entries Includes: - 16 SystemVerilog RTL files in rtl/DCon/ - FX68K core (rtl/fx68k/), audio cores (rtl/sound/jtopl, jt6295, t80) - JTFRAME framework subset (rtl/jtframe/), Sorgelig SDRAM bridge - MiSTer sys/ framework (HPS_IO, OSD, video scaler, audio) - Pause overlay with logo, supporters list and patron scroll - Quartus project files (DCon.qpf/qsf/sdc) - Prebuilt RBF (releases/DCon_20260530.rbf) and parent MRA - 6 in-game screenshots in docs/ Licensed under GNU GPL v3 or later. ROMs not included.
93 lines
2.5 KiB
Python
93 lines
2.5 KiB
Python
#!/usr/bin/env python3
|
|
"""
|
|
Genera file .mem con encoding tier+ASCII per modulo pause_text.
|
|
|
|
Formato output: 1 word 9-bit per char (2 bit tier + 7 bit ASCII), salvato
|
|
come 16-bit hex (1 word per riga, valore <= 0x1FF).
|
|
|
|
Sintassi sorgente .txt:
|
|
- Righe che iniziano con `@` sono separatori/commenti — IGNORATE.
|
|
- Righe che iniziano con `N:` (N=0..3) → tier=N, testo dopo i due punti.
|
|
- Righe normali → tier=0 (default, bianco).
|
|
- Righe vuote → riga vuota tier=0.
|
|
|
|
Char accettati: 0x20-0x5F (uppercase). Lowercase a..z viene convertito in A..Z.
|
|
Char fuori range diventano space (0x20).
|
|
|
|
Usage:
|
|
python gen_msg.py file.txt file.mem CHAR_PER_ROW NUM_ROWS
|
|
"""
|
|
|
|
import sys
|
|
import os
|
|
|
|
|
|
def normalize_char(c):
|
|
code = ord(c) if isinstance(c, str) else c
|
|
if 0x61 <= code <= 0x7A:
|
|
code -= 0x20
|
|
if 0x20 <= code <= 0x5F:
|
|
return code
|
|
return 0x20
|
|
|
|
|
|
def parse_line(line):
|
|
"""Ritorna (tier, text) — tier 0..3, text già normalizzato."""
|
|
if line.startswith("@"):
|
|
return None # commento, skip
|
|
tier = 0
|
|
if len(line) >= 2 and line[0] in "0123" and line[1] == ":":
|
|
tier = int(line[0])
|
|
line = line[2:]
|
|
return (tier, line)
|
|
|
|
|
|
def main():
|
|
if len(sys.argv) != 5:
|
|
print(__doc__)
|
|
sys.exit(1)
|
|
in_path = sys.argv[1]
|
|
out_path = sys.argv[2]
|
|
chars_per_row = int(sys.argv[3])
|
|
num_rows = int(sys.argv[4])
|
|
|
|
with open(in_path, "r", encoding="utf-8") as f:
|
|
raw_lines = [l.rstrip("\n") for l in f.readlines()]
|
|
|
|
parsed = []
|
|
for raw in raw_lines:
|
|
result = parse_line(raw)
|
|
if result is None:
|
|
continue
|
|
parsed.append(result)
|
|
|
|
# Trunca o pad righe
|
|
if len(parsed) < num_rows:
|
|
parsed += [(0, "")] * (num_rows - len(parsed))
|
|
else:
|
|
parsed = parsed[:num_rows]
|
|
|
|
words_out = []
|
|
for tier, line in parsed:
|
|
if len(line) < chars_per_row:
|
|
line = line.ljust(chars_per_row)
|
|
else:
|
|
line = line[:chars_per_row]
|
|
for c in line:
|
|
ascii_code = normalize_char(c)
|
|
word = (tier << 7) | (ascii_code & 0x7F)
|
|
words_out.append(word)
|
|
|
|
expected = chars_per_row * num_rows
|
|
assert len(words_out) == expected, f"size mismatch: {len(words_out)} != {expected}"
|
|
|
|
with open(out_path, "w") as f:
|
|
for w in words_out:
|
|
f.write(f"{w:03X}\n")
|
|
|
|
print(f"Saved {out_path} ({len(words_out)} words = {chars_per_row}x{num_rows})")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|