Knowing which parameter changes to an audio event cause destructive editing and degradation after Bounce Selection and which don’t cause destructive editing is important if you are making multiple changes in stages. Multiple parameter changes followed by Bounce Selection in stages that are destruction can cause artifacts and distortion to accumulate.
When using Bounce Selection on one or multiple audio events, I’ve read that Bounce Selection is inherently non- destructive, but that seems in correct.
There are many parameters that can be baked into the file.
I am trying to sort out which parameters cause destructive editing and degradation which parameters don’t when using Bounce Selection.
Which of these cause destructive editing and which don’t when using Bounce Selection.
Positioning of multiple audio events on the time line for consolidation. (?)
The main purpose of the Bouce Selection is creating a new file(s) from the selected event(s). It’s like a local export to put all the real-time changes into the file.
If I google this topic: Cubase: “Bounce To Selection” I get conflicting information. I seem to be unable to get a definitive answer about whether “bounce selection” produces a destructively or non-destructively created file with that has no degradation of quality or minimal or imperceptible loss of quality.
These are some of the responses I get from google:
• In Cubase, bouncing a selection (with no effects or EQ applied) generally causes no noticeable audio degradation as long as you keep the project’s bit depth and sample rate the same as the original audio. The bounce process essentially creates a digital copy of the selected audio, and if the source material and bounce settings are identical, there’s no quality loss.
• No quality loss bouncing digital wav to digital wav unless the track has been compromised by an effect or digital distortion from overdriving.
Let’s break down the element of this discussion:
1. destructive or non-destructive
Is the resulting file (Audio Event) destructively created or non destructively created. I am thinking it must be destructive because a new file is created. But the original file is not deleted.
2. No loss of quality or minimal loss of quality
Is there:
• no loss of quality (a perfect digital copy of the selected audio) if no effects or EQ applied
or
• minimal or imperceptible loss of quality?
Some of the response to this question state that if no processing is applied, the bounced audio is an exact copy of the original file(s).
So the question is, if no effects or EQ are applied, is the resulting bounced audio
• an exact digital copy of the original audio file(s) even when multiple audio event are combined into new bounced event
or
• a audio event with minimal or imperceptible of loss of quality?
I have tested Google AI responses to a vast number of topics, and it has been completely wrong about 90% of the time in its technical, algorithmic, and scientific results. It is objectively worthless in these areas from empirical testing.
Take that how you wish, but the very first thing I would do is completely ignore anything Google AI says. There’s no reason to even read its responses in regard to any discussion in any manner of technical forum.
When you bounce a file, everything you done to the actual audio, volume changes, fades, pitch, time corrections, everything, will be flattened to the new file, that not recoverable, unless you undo the function. I would suggest duplicating the audio track and saving it , in case you have doubts that you might want to do further changes to the audio.
I would agree to a fellow musician here, don’t take what AI says too much to be the absolute guideline.
But still one of the original questions remains, if there are no EQ or effects, no volume changes etc applied is the resulting audio a perfect digital copy of the original audio or a second generation digital copy with very minimal loss of quality.
It’s not a “perfect digital copy,” but rather, another audio file created with the same settings. The end result is a logical “copy” if there really aren’t any other elements at all that could change audio over time (like free LFOs, etc).
Meaning, all other things being equal, bouncing the exact same mix will result in the same file. There’s no interaction between the “first copy” and the “second copy.”
I finally got the answer I was looking for and it came from AI. I actually asked 5 different chat bots the same question and they all gave the same answer
When bouncing audio with no effects, EQ or volume changes applied and assuming same bit depth and sample rate, the resulting file is a bit-perfect digital copy of the original audio.
There is no quality loss in this scenario.
This is true even when combining multiple audio events into a new bounced event. The combination of multiple events is essentially a concatenation or merging of the digital audio data, with no re-encoding or resampling involved, ensuring no loss of quality.
The process is essentially just reading the digital data from the original file(s) and writing that exact same data to a new file. Since it’s a direct digital-to-digital transfer with no processing, conversion, resampling or re-encoding. The information remains identical at the bit level.
Minimal or imperceptible loss: This would only occur if some form of processing (e.g., effects, resampling, or bit depth/sample rate conversion) is applied during the bounce, which is not the case in your scenario.
Haha yeah, at the bottom of every google AI result it says “Generative AI is experimental,” and indeed it is… experimenting with bad/misleading info! It’s gotten me in trouble a few times leading to some very incorrect assumptions, so lately I just ignore it.
If I remember correctly, the target bit depth for bouncing is defined by the recording bit depth in project settings, so theoretically, if you have a source file with a higher bit depth than the project settings, this will lead to truncation when bounced, i.e. a not bit identical file. Unless you set 16bit though (and who does that…), it’s pretty much irrelevant.
Yes. That’s a very important consideration if you are bouncing a lot in stages, to check bit depth of thre clip against the project bit depth, is that it.?
