it's great that vgm format supports a significant amount of sound chips, but there are still missing a quite bunch of them, like:
- d65010g31 (from Casio's PV1000 game console - 3-channel based, quite similar to sn76489, lacking volume registers and noise)
- mos6581 (sid, from Commodore 64)
- csg8364 (paula, from Commodore Amiga)
some questions:
- are there more sound chips missing, beyond these listed above?
- how soon will we get them implemented in the next vgm specification?
missing sound chips in the last vgm specification
Technical discussion which is not directly related to VGM files. Talk about Hardware and Software.
Moderator: Staff
-
- nitrofurano Offline
- Posts: 116
- Joined: 2016-02-16, 12:54:37
SID and Paula won't be implemented anytime soon.
For SID the memory limitation of the C64 + variants and the fact that it's the machine to use the chip, and that the .SID format already exists.
For Amiga, Delitracker custom formats (as supported by UADE) is currently a better option than VGM. Also, the chip reads from RAM which will make things harder for VGM logging as the log will snap up everything from code to sound driver state. This is also a problem with SNES.
As for the Casio chip, we probably won't support it unless there's an interest in logging files from that machine. In that case, you should do like eito and prepare a fork of VGMPlay + vgmtools to actually make a proof of concept first. In fact, you should probably do that for every chip you propose.
For SID the memory limitation of the C64 + variants and the fact that it's the machine to use the chip, and that the .SID format already exists.
For Amiga, Delitracker custom formats (as supported by UADE) is currently a better option than VGM. Also, the chip reads from RAM which will make things harder for VGM logging as the log will snap up everything from code to sound driver state. This is also a problem with SNES.
As for the Casio chip, we probably won't support it unless there's an interest in logging files from that machine. In that case, you should do like eito and prepare a fork of VGMPlay + vgmtools to actually make a proof of concept first. In fact, you should probably do that for every chip you propose.