2

I looked online and didn't find anything definitive, they all wanted to compare the source and changed data structures and listen for differences....which I would consider too subjective to yield any factually grounded answers.

I feel like there should be a straight yes or no answer to this question. If I make a recording of a track in audacity for example. And then I use amplification to any extent short of clipping, am I losing quality from the original recording upon export? (I'm recording at 48hz and 32bitfloat, so I understand that even without amplification my export will lose quality unless its a wav f32 or higher format export.)

What is amplification doing exactly? Is it simply stretching existing finite data or recalculating the data to fit a larger model? The former would, to me, imply quality loss.

1

Amplification is multiplying your samples with a constant. Since your samples are floating point numbers, statistically this adds a bit of quantization noise with a size of the order of one least significant floating point mantissa bit. Usually this is quite less than one bit of the final data, so one can do quite a number of such operations before the accumulated effects are an appreciable source of noise of their own. Of course, if your intermediate sample format is not floating point but already the final sample format, the effects of such manipulations accumulate much faster.

  • My situation is: I capture recordings in audacity in 32 bit float, then I export them as 16 bit signed .wav files that I then toss into my VSE and line up there to video. I amplify these pieces in the vse up or down. I then 'have to' export those files out of the vse, I can do so at 32f wav or 16 bit signed wav. Then I amplify a bit more and apply effects in audacity in a new 32f space. Then I have to choose a final format to mixdown all those channels of audio together as my final render. Should I be concerned about loss due to amplification given all this? any suggestions to improve final? – kite Nov 17 '18 at 8:28
0

the simple and short answer to your question is No. You are not losing any quality by performing an amplification (positive gain) function.

If on the other hand, you were to perform the opposite attenuation function (negative gain) then the answer would be different - yes - you would be losing quality.

Subjective "quality" of a signal can be measured in a number of ways, however the easiest way to understand this is to consider signal/noise ratio.

When digitally amplifying an already recorded signal, you increase all the recorded material by the same level. This includes :

  • Signal
  • Self-noise of recording equipment (preamps/amps/capsules etc)
  • Quantization noise through digitization process.

However the ratio of each component stays constant.

By reducing the overall gain of your signal, you are able to reduce the signal to the point where the quantization noise component becomes a significant component of the signal and becomes clearly audible. Dither (shaped noise) is often used to counteract the effects of quantization noise at low bitrates.

  • Is this a case of the degradation of negative gain being so small its not even worth caring about? Or is it significant enough given the kinda process I explained in my reply above? Also what was the issue people had with this answer? – kite Nov 17 '18 at 20:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.