Jim
it's true that if you want to create a good quality UT you need to record a lot of audio. That's out of the question. Also, tempo is not really the problem with UTs. The problem, as I see it, is that:

1) even if you create A LOT of audio, in many case the UT will still sound bad, when changing chords, because of the extreme timing precision requirement that Icelander was talking about. Now, for some UTs this is not too much of a problem (when you don't have legato between chords, for example - I did create some UTs that sound absolutely perfect!!). But for (many) other kinds of UTs this problem makes it almost impossible to have them sound good, no matter how much audio you record, no matter how precise you try to be. Anyway, on this issue Icelander is much more competent than me and I think he explained it already very well

2) the other problem is also not about the tempo, but about key transposition. If during playback you set the song so that key transposition is allowed, with UTs you get some weird transpositions that don't seem to make musical (or logical) sense. If I have a recorded audio with the C major chord, and BIAB needs to play a Db major chord, I would want BIAB to transpose my C chord by 1 semitone, which would be totally fine. Instead, what happens is that I hear transpositions of MANY semitones (so for example not from C to Db, but from G to Db, or so ... I'm just guessing here), which is not only sounds really bad, but it's also unnecessary, since BIAB can use my readily available C major to be transposed into Db!
So, at the moment, there's only one solution to this problem: to record a 12 keys UT, and forbid key transposition with the song setting parameter during playback. Which is ok, it does lead to good results, but it's VERY inefficient. Why not allow the user to control the maximum number of semitones that BIAB transposes during playback instead? It would be a much more efficient way to manage the issue. In this way, I could record audio not to (unnecessarily) cover all keys, but to make a better UT with more variations etc.

Jford,
I totally agree with you. I think that PG will address this, eventually. I don't see why not, really. But BIAB is a very complex program and I guess that UTs are not a top priority. I think it's understandable. Maybe many users don't care about UTs, I dont know. I guess time will tell.

Last edited by Jon Thomas; 12/05/17 01:44 PM.