Hi,
setting up a bunch of pre-rendered stems on the different tracks in the Octa.
I know I should leave time stretch settings off if I want to make 100% sure they are not run via any additional processing at all, but I would also like the option to switch tempo, for live performances.
Do you find that turning the time stretch on affects audio quality, even if the audio is played at it’s original tempo?
I am aware you will always run into artefacts once you start stretching, but not sure if just leaving it on in the background will already eat into fidelity.
A tinsy bit it does, yes. At least to my ears. But you should try it yourself to see if it would be a problem for you. You might be fine with the loss of it means your performance will go the way you want it.
With OT master I don’t remember noticing a quality loss at original tempo, but I don’t use timestretch except for drastic fx…I usually turn it off anyway, especially for pitch doubling speed.
A recording test to compare would be interesting…
Maybe depends on audio material.
Yes, I could hear tiny changes in timbre and clarity. But - as I think most people on here that have presumably also sunk decent amounts of money into equipment are aware of already - that subjective and perceived qualities doesn’t necessarily mean that it’s the truth.
Either way, as you say, if the difference is that small and it makes the performance better it’s a reasonable trade-off. If I can barely hear it, then no one else will notice. Thanks for snapping me back to reality, of course that is the way to go. Otherwise I might as well just playback the whole thing from the master recording.
@xidnpnlss and @valdemar, I just made a test, a track with timestrech on, Normal mode, **same tempo as original", and another one with the same sample, tempo off, inverted phase.
Result : no sound at all, due to phase cancelation, meaning exactly the same sound.
I think with an AB test it would
be impossible to recognize the timestreched sample. May be more noticeable with other types of samples (on another sample I heard an artifact at the very beginning of the sample, that’s all).
In Beat mode, I could hear sound, meaning the sound is different.
I prefer to disable timestrech most of the time and anyway, if you want to activate / deactivate timestretch quickly on several tracks you can use parts.
Come to think of it, I determined that there was a change based on turning it in and off On a playing same and hearing a change between them (negligible). That should be expected, I imagine. Your test - for phase - was clearly more accurate.
Wow, that is super cool. Thanks for actually taking the time out to research what I was wondering about.
As we all now, regardless of what I “heard”, the phase cancellation goes to show that hearing is not everything, as it’s so subjective.
I did hear it mainly in the kick drums which inspired me to make this thread, and they were on the beat algorithm, and as you say that is indeed affected. But as long as the general textures stay the same, I will be good to go. Very few will tell the difference in percussion samples in a live situation.
The first sample I used was a 4 bar rich stereo pad, with other synths sounds and drums.
The sample with a little artifact at the beginning was Amen Break loop.
It would be interesting to test on longer samples, other audio material.
Ideally I think you have to use bpm with natural number values. Attributes tempo can’t match all bpm values. 0.2, 0.4, 0.5, 0.6, 0.8 decimals should work though.
I wish OT tempo was more precise. You can change Attributes bpm by 0.25 increments with arrows, or 0.4 increments with Func+Arrows. Probably work with Level knob too.