Exactly. Agree completely.Borogove wrote:If it's on disk, it can read ahead (effectively 'into the future') to accomodate speed-up, or it can read more slowly without risk of running out of RAM to accomodate slow-down.JCJR wrote:Its realtime, and its an effect. What diff does it make if the effect is acting on a disk file rather than live input?
The quoted statement was a quibble about semantics-- what to call a player (or sequencer, or sampler, whatever) that can stretch in realtime. All audio players, IMO, must at least have 'loose' realtime characteristics, to keep audio from dropping out. The audio player must have each buffer prepared in time for the soundcard's realtime constraints, or playback fails. That would be called realtime stretching in my book.
http://www.linuxdevices.com/articles/AT3524337625.html
=====
On another topic, joer's gated stretch idea might be an interesting 'live playthru' effect. Perhaps useful in 'outside' recordings or live performance, or at least a hoot late at nite after a few beers.
Maybe have a 30 second delay buffer. Run a gate detector on the live input. On each audio input phrase onset, the gate would reset the stretch process to output to 'now' in the buffer, mixing the new stretch output with whatever was already written to the delay buffer. It would gradually 'fade out' stale contents of the buffer. Might be tricky to configure correctly.
If stretching 2X, if you speak, "Twinkle, Twinkle, Little", the gate would move the stretcher output to 'now' at the onset of each word. The first 'slowed down' word would still be playing when the second 'slowed down' word gets mixed-in atop the tail of the first 'slowed down' word, etc.
If nothing else, speaking, singing, or playing thru such a plugin would be a wonderful way to get terminally confused (GRIN).
...) But he obviously could.
