Expected behaviour:
The original G-5 is faded out to zero. When the subsequent C-5 is triggered, only the C-5 should sound because the previous note was not already audible and has now entered its release phase.
Actual behaviour:
As soon as the C-5 is encountered, the previous G-5 immediately becomes audible as it enters its release phase.
I imagine this is because the volume of the new note set the volume for the entire note column, thus overriding the previous fade out, but this makes no sense from a user perspective: once a note is faded out it should not re-appear when a NEW note is triggered.
I had a feeling someone would say this, to which I have two comments:
The example is illustrative, it is not what I am actually doing. What I was trying to do was fade the previous note slightly (but not to zero) before the new note sounds, only to find that it unpleasantly jumps back to full volume along with the new note.
I have just discovered that even if you use a note off instead of fading the previous note, the jump to full volume still occurs if the previous note is still fading when the new note starts.
As far as I can see there is no way of avoiding the volume jump except by using a separate note column for the new note.
And why don’t you want to use a second note column? It’s what they are there for! Note column volume is going to affect all notes in that column is you have NNA set up so more than one can be played. Now else do you expect it to work?
Because the note in question is part of a four-note chord which needs to fade as a whole, and I don’t want to have to use an 8-column track throughout the entire song when it seems that NNA’s should be sufficient here.
I expect that when a note off event is triggered, the note continues to fade until it gets to the end of its volume envelope, without suddenly jumping back to full volume because a NEW note is played.
It may make perfect sense from an architectural standpoint to have a single active volume per note column, but it makes no sense from a usability perspective: no user would expect that playing a new note will affect the volume of the previous note.
I have no problem using note-offs in general, they just don’t seem to be effective if you want to start fading a previous note before the new one is triggered, and have the old note continue to fade.
In the end I just gave up and kept the old notes at full volume until replaced by the new chord.
I suspect, if you were to watch the envelope window, that it would not jump back to maximum as you put it. This is clearly true as you have said it works when you leave it at full volume. You can not really not expect it to affect the overall volume of the note when you change the value of the volume in the note column. That is the whole point of it. It’s like having a fade-out at the end of a track and pushing up your fader level then complaining that it’s not getting quieter.
I tried to create a simple test song that I could upload to demonstrate the problem, but in a simple test I don’t get the issue. Maybe I am just doing something stupid. Further investigation needed.
EDIT: OK, never mind. I can’t really tell whether it is actually happening at all; possibly it was caused by having a 00 fadeout on the volume envelope (which I guess means the note plays forever even when silent), but it might just as well be my imagination. My simple two-note sinewave test works exactly as expected.
I can’t edit the topic title, I guess someone should change this to NOTABUG status.
I guess that is what we all tried to tell you with the “That is what note-off is for” .
It is our fault that we were somehow not clear enough in explaining this better.
If you use the volume envelope of the instrument editor to your likings, you would not need to use volume values or volume fade effects in the note-column at all.
Ofcourse there might be a problem if you desire more than two kinds of fade values for the same instrument.
And you also might have a problem when using VST instruments which cannot be affected by the volume envelope.
No, I understood the suggestion to use note-off but mistakenly believed that the volume jump was still occuring if note off was used instead of volume slide. I suspect it is because the sample in question is an orchestral string section, and the new chord probably has some harmonics that sound a bit like the old chord.
That was the original motivation: to fade the old chord slightly before bring in the new one. Fortunately triggering the note off a few lines before the new note seems to create the desired effect.
