SWAM
Vivaldi String Orchestra with SWAM solo strings

I've been using Vivaldi to experiment with various techniques to improve the ensemble sound of solo strings. The results are in this playlist.

https://soundcloud.com/bruno-degazio/sets/vivaldi-four-seasons-spring

The main techniques are:

1.) Time spreading the Note-onsets.
For example, the 6 first violins are individually delayed anywhere from -50 milliseconds to +50 milliseconds (negative delay means the note starts before the beat). Each of the 8 'players' gets a different delay on each note. The overall spread from earliest player to the last is greater for the first note after a rest, as it usually is in an ensemble.

2.) Detuning the Note-onsets.

Each 'player' begins their note anywhere from 30 cents flat to 30 cents sharp, and quickly slides into tune. The amount of detuning is small for small pitch intervals (2 or 3 semitones) and greater for big pitch intervals.

3.) Detuning the Note Sustains.

Even with the above two techniques in place I found that sustained notes still tended to fuse into a single sound. Adding vibrato helped (at a slightly different rate for each player) but even better was adding a degree of random detuning, -10 cents to +10 cents, separately for each player, changing constantly at a slow rate like a random LFO. The detuning amount is updated every second or so, again different for each player.

I've been using Vivaldi to experiment with various techniques to improve the ensemble sound of solo strings. The results are in this playlist. https://soundcloud.com/bruno-degazio/sets/vivaldi-four-seasons-spring The main techniques are: **1.) Time spreading the Note-onsets. **For example, the 6 first violins are individually delayed anywhere from -50 milliseconds to +50 milliseconds (negative delay means the note starts before the beat). Each of the 8 'players' gets a different delay on each note. The overall spread from earliest player to the last is greater for the first note after a rest, as it usually is in an ensemble. **2.) Detuning the Note-onsets.** Each 'player' begins their note anywhere from 30 cents flat to 30 cents sharp, and quickly slides into tune. The amount of detuning is small for small pitch intervals (2 or 3 semitones) and greater for big pitch intervals. **3.) Detuning the Note Sustains. ** Even with the above two techniques in place I found that sustained notes still tended to fuse into a single sound. Adding vibrato helped (at a slightly different rate for each player) but even better was adding a degree of random detuning, -10 cents to +10 cents, separately for each player, changing constantly at a slow rate like a random LFO. The detuning amount is updated every second or so, again different for each player.
edited Mar 17 at 12:45 pm

Sounds great and very natural in most places. (But I am not an expert on such things at all.)
Did you apply the irregularities manually or did you use some software tools ? (In fact I recently did such a tool that might - or might not - be of interest for you.

Sounds great and very natural in most places. (But I am not an expert on such things at all.) Did you apply the irregularities manually or did you use some software tools ? (In fact I recently did such a tool that might - or might not - be of interest for you.

@mschnell, I use software tools (scripts) for Detuning of Note onsets and sustains. These were fairly complicated because the detuning needed a fast envelope to sound musical. I found that the detuning at the note-onset had to be fairly strong, but the sustain much less so. If the strong detuning of the note-onset remains throughout the sustain part of the note, it just sounds like a bunch of machines playing out of tune. If the note begins quite out of tune, but quickly slides into better tuning, it sounds much more like a human player would do.

The time spreading also uses a script but it's very simple one because support for millisecond level time-delay is built into the MacOS MIDI API. All I need to do is specify a slightly different delay value to MIDI driver for each channel of the string section.

@mschnell, I use software tools (scripts) for Detuning of Note onsets and sustains. These were fairly complicated because the detuning needed a fast envelope to sound musical. I found that the detuning at the note-onset had to be fairly strong, but the sustain much less so. If the strong detuning of the note-onset remains throughout the sustain part of the note, it just sounds like a bunch of machines playing out of tune. If the note begins quite out of tune, but quickly slides into better tuning, it sounds much more like a human player would do. The time spreading also uses a script but it's very simple one because support for millisecond level time-delay is built into the MacOS MIDI API. All I need to do is specify a slightly different delay value to MIDI driver for each channel of the string section.
72
2
2
live preview
enter atleast 1 characters
WARNING: You mentioned %MENTIONS%, but they cannot see this message and will not be notified
Saving...
Saved
With selected deselect posts show selected posts
All posts under this topic will be deleted ?
Pending draft ... Click to resume editing
Discard draft