People noticed that the music was more shallow because one of the melody channels (#3) wasn't playing at all. With the 1 byte code change, music comes out more energetic since you can hear the "background" instrument playing now.
That was something I noticed lacking from the Switch, but I still prefer the "cleaner" raw soundtrack from the hardware. I feel like the Switch soundtrack is "EQ-enhanced" somehow.
GG ALESTE 3 for Sega Dreamcast Port
- ateam
- Metallic
- Posts: 826
Re: GG ALESTE 3 for Sega Dreamcast Port
Gotcha.
However, I'm confused then about the instruction to write those FE bytes. You said:
So is it a one-byte change (3A to offset 0xC66A), or is it all of those?Put 0xFE back to these locations:
- C79A, C8A1, C8B0, C942, CA6E, CAD2, CB65, CB74, CC06
(matches noise channel, which uses correct untouched code)
Also, do you have a source/original post for this? I'd love to read more

Find me on...
• DreamcastForever.com
• GitHub
• Reddit
• SegaXtreme
• Twitter
• YouTube
• Discord: derek.ateam
• DreamcastForever.com
• GitHub
• SegaXtreme
• YouTube
• Discord: derek.ateam
-
- noob
- Posts: 4
Re: GG ALESTE 3 for Sega Dreamcast Port
I was chatting on private irc and someone posted the updated patch. It's a 1-byte change to all those locations, but the only critical one is the C66A change.
People were taking apart the code and noticed that M2 has 2 paths for sound, which are nearly identical copy-paste. The one that handles the noise channel was undamaged and used as a clear reference. Eventually they decided to revert (unpatch) all the modified bytes. These are the 0xFE bytes but don't seem to have much (any?) effect. Done for "correctness" as they put it, but optional.
The 0x3A one is used in the irq handler. It's supposed to be a "read memory" operation, which makes sense since ch1 and ch2 does the same. M2 put a "write memory" which borked the ch3 track. It looks like valid asm so I guess everyone overlooked it all these years.
Heard that M2's VS Puyo Puyo MD was something very similar - some missing 1-byte change that caused the music to fail there too.
No idea who did the original patches for Aleste3 or the Mini2 games.
People were taking apart the code and noticed that M2 has 2 paths for sound, which are nearly identical copy-paste. The one that handles the noise channel was undamaged and used as a clear reference. Eventually they decided to revert (unpatch) all the modified bytes. These are the 0xFE bytes but don't seem to have much (any?) effect. Done for "correctness" as they put it, but optional.
The 0x3A one is used in the irq handler. It's supposed to be a "read memory" operation, which makes sense since ch1 and ch2 does the same. M2 put a "write memory" which borked the ch3 track. It looks like valid asm so I guess everyone overlooked it all these years.
Heard that M2's VS Puyo Puyo MD was something very similar - some missing 1-byte change that caused the music to fail there too.
No idea who did the original patches for Aleste3 or the Mini2 games.
-
- Similar Topics
- Replies
- Views
- Last post
-
- 8 Replies
- 6507 Views
-
Last post by gamesreup
-
- 5 Replies
- 1672 Views
-
Last post by gameblabla
-
- 27 Replies
- 36108 Views
-
Last post by infinity
-
- 0 Replies
- 5579 Views
-
Last post by gamesreup
-
- 0 Replies
- 4370 Views
-
Last post by gamesreup