Show Posts
1
Misc Audio / Re: [FF6] [OPEN BETA] Dancing Mad - FF6 Audio Replacement using MSU-1
« on: 2019-02-25 19:45:28 »
Hi, and thanks for responding to this. I do get the sense from looking at your work that you're fitting a mechanism with a relatively simplistic interface over something that's ultimately much more complicated. With originally so many different types of events the SPC gets to handle with things like operas and phantom trains and dancing mads, it's quite an accomplishment that the switch to using the MSU-1 works so well.
That said, I've been giving some thought to at least the transitions in Dancing Mad case and I'm wondering if maybe there could be a way to get it to work a little smoother. My thought is do we really need this status information for this to work properly?
As originally implemented, Dancing mad 1-3 seems to act like one song with 3 separate start and stop loop points of course; and what we have are three tracks with one loop point each and a distinct end of the track. The way things are now the song can move backward properly, but we can think of the trigger that goes off when a part's boss is defeated as a gate being lifted to allow us to move forward and the repeat control might be able to effectively work as this gate.
What I'm wondering is looking at the MSU-1 spec, what if when this trigger goes off we clear the repeat flag in the audio state control while still leaving the current track playing. My hope is in order for this to work when the track is finished, the audio playing bit in the status port will go to zero. Then the instant that happens we have the MSU-1 start the next track. The concern is of course will there be an audible gap or pop in the music between when the track reaches its end, realizing with whatever polling or other method that the track has stopped, and then starting the next track?
It's just an idea I had and I understand after so many years working on this if your response is, "that won't work" or "I already tried that" or even "I'm not going to bother with that". Maybe someday I could be able to attempt to implement this myself, but at least now I would certainly be willing to help test this or any other changes or added features to the project.
That said, I've been giving some thought to at least the transitions in Dancing Mad case and I'm wondering if maybe there could be a way to get it to work a little smoother. My thought is do we really need this status information for this to work properly?
As originally implemented, Dancing mad 1-3 seems to act like one song with 3 separate start and stop loop points of course; and what we have are three tracks with one loop point each and a distinct end of the track. The way things are now the song can move backward properly, but we can think of the trigger that goes off when a part's boss is defeated as a gate being lifted to allow us to move forward and the repeat control might be able to effectively work as this gate.
What I'm wondering is looking at the MSU-1 spec, what if when this trigger goes off we clear the repeat flag in the audio state control while still leaving the current track playing. My hope is in order for this to work when the track is finished, the audio playing bit in the status port will go to zero. Then the instant that happens we have the MSU-1 start the next track. The concern is of course will there be an audible gap or pop in the music between when the track reaches its end, realizing with whatever polling or other method that the track has stopped, and then starting the next track?
It's just an idea I had and I understand after so many years working on this if your response is, "that won't work" or "I already tried that" or even "I'm not going to bother with that". Maybe someday I could be able to attempt to implement this myself, but at least now I would certainly be willing to help test this or any other changes or added features to the project.