When extracting OPN/OPN2 voices from VGM files, vgm2x creates individual files (TFI, DMP, etc.) for every patch instance found in the data stream. This results in hundreds of thousands of files for a single game, most of which are duplicates.
For example, extracting from 25 Shadow Dancer (Sega System 18) VGZ files creates ~700,000 TFI files, when the actual unique patch count is probably ~100-200.
Current behavior:
- -m OPM → single .opm text file with deduplicated patches ✅
- -n TFI → one file per patch instance (no deduplication) ❌
Requested feature:
Add deduplication and/or bank-style output for OPN extraction, similar to how OPM extraction works. Options could include:
- A --nodupes flag for OPN extraction
- An OPN bank format output (single file with multiple patches)
- Built-in deduplication during extraction
This would make OPN extraction practical for games with YM2612/YM3438 chips.
When extracting OPN/OPN2 voices from VGM files, vgm2x creates individual files (TFI, DMP, etc.) for every patch instance found in the data stream. This results in hundreds of thousands of files for a single game, most of which are duplicates.
For example, extracting from 25 Shadow Dancer (Sega System 18) VGZ files creates ~700,000 TFI files, when the actual unique patch count is probably ~100-200.
Current behavior:
Requested feature:
Add deduplication and/or bank-style output for OPN extraction, similar to how OPM extraction works. Options could include:
This would make OPN extraction practical for games with YM2612/YM3438 chips.