I suspect Follin simply did everything by ear.
That is correct, and confirmed by the man himself, iirc.
]]>https://www.youtube.com/watch?v=B_SmzoQJBsQ
The issue is that the playback engine has got extremely ad-hoc timing. Every bit of it takes as long as it takes with no attempt to synchronise timings anywhere. So, depending on how much work it's doing, the pitch will change slightly. Because my new engine is using a different architecture, all the timings are different, and it's all fundamentally untunable. I suspect Follin simply did everything by ear. I've found transcriptions of this track, and the same note in different contexts will be represented by a different value in the data. To make it work I'm pretty sure I'm going to have to retranscribe the music from scratch, and, well, then it's not the same engine any more, so I'm going to leave it here.
Anyway, I thought people might be interested to see this.
Source: https://gist.github.com/davidgiven/ca16 … bef4504526
.prg: https://drive.google.com/file/d/1jRVbj3 … ?usp=drive
(I'd be interested to know what it sounds like on real hardware.)
]]>