Subject: Re: [linux-audio-dev] protux, stereo and interleaving
From: Iain Sandoe (iain_AT_sandoe.co.uk)
Date: Sun Feb 11 2001 - 16:58:42 EET
Hi David,
[...] nice ascii-art snipped ;-)
I may comment on the first part later - bit short of time RN - but this next
bit is important (at least to me):
>> For example, I quite often record "live" in the studio and then
>> want to relate the recording back to Old-Fashioned Western (tm)
>> bars/beats etc. So that other tracks can be added on a sequencer
>> approach.
>>
>> It is essential to lock the tracks exactly because one will be
>> fiddling with the "notational timebase" - this affects the start
>> time of the audio (in bars/beats) and it must affect _all_ the
>> audio the same...
>
> I'm sure I'm following here... Indeed, fiddling with the "time track"
> as they call it in some apps, should keep audio in the right places
> musically, but I'm having some trouble seeing how you could make sure
> that is always the case when changing the tempo. (Some audio/MIDI
> programs do it by time stretching/compressing the audio to fit the
> MIDI timing exactly, but that has a tendency of either severely
> degrading the quality, or take ages to perform - usually both!)
nope. I haven't expressed it well... that wasn't it - although it *is*
sometimes useful (and the quality issue is an algorithmic one rather than
fundamental - I think).
---OK - I will try and expand upon this - it is *very* important to bands like mine where you have a mixture of people some of whom deal most comfortably with "standard" musical notation and some of whom "play exclusively by ear".
Let us assume that there is an instance of what VST calls the "Master Track". This contains the relationship between real elapsed time and "musical notation time" - the latter expressed in terms of tempi, beats signatures and so on.
so, supposing we play a piece (we are prog. dinosaurs BTW) that has several time signature changes and maybe almost continuous tempo change. We play "live"/Jam - that is the way we compose. So we have a stream of audio in the machine about which the audio program currently knows nothing - just where it starts. OK the program has a notional tempo (but that it currently irrelevant).
now I want to create a "master Track" that reflects the correct tempo/signature/real time relationships - because then I can add 'sequenced' type racks that can be printed out in score form. (maybe these will just transcribe what was played - it depends).
****> I do not wish to *touch* the audio - all I wish to do is to build a "sequencer" type track that would have the correct logical information to relate back to the audio.
So I lock the audio (VST 'timelock" facility). I then proceed to enter the time/tempi/signature info - by matching key points in the audio - tedious but rewarding when done.
Now I can release the lock on the audio and we can play midi stuff into the machine which not only will match the audio in time - but can be used to score the music.
--- I can expand further if anyone wants to understand this better - it is vital to me (and I don't think VST would have it if I was the only one ;-)
ciao, Iain.
This archive was generated by hypermail 2b28 : Sun Feb 11 2001 - 17:19:06 EET