Multi-Track Amplify

Special Interest Groups Audio Processing
1

This may not exist as a product or tool, but it should.

37 track multi-track show. Many tracks one above the other. Select all the tracks. Effect > Amplify > -1.

Amplify adjusts all the selected tracks so the Stereo Mixdown comes out without overloading or clipping. I picked Amplify because Normalize would be a nightmare.

Koz

2

Interesting idea.

I think I prefer the idea of a “Master Fader” on the Mixer Board that moves all selected faders (and so all track gain sliders) in unison. This way would be non-destructive.
No reason that I can think of for not having both options.

3

With long tracks it would be slow.

Audacity would need to calculate the peak amplitude of the mix (Audacity “knows” the peak amplitude of each track, but it does not know what the peak of the sum of multiple tracks will be without calculating the sum for each sample period), then from that work out the necessary amplification.

4

I think I prefer the idea of a “Master Fader”

I prefer the idea of making the computer do the scut work. Both tools are valuable, but the Master Fader demands recursive efforts to find the sweet spot.

“Still clipping. Let’s try it a little lower.”

Koz

5

With long tracks it would be slow.

I’m not planning on being here while it does it. I have to strike the studio and clean up the broken bottles and spilled chocolate brownie tins.

Koz

6

The way to do it manually (Windows/Linux shortcuts)

  1. Ctrl+A (select All)
  2. Ctrl+Shift+M (Mix to new track)
  3. Select the new track
  4. Open the Amplify effect and make a note of the default amplification amount (as an example, let’s say that it is -23.6 dB)
  5. Subtract 1 from that number (-24.6 dB)
  6. Ctrl+Z (undo the mix track - all other tracks should be selected)
  7. Amplify and use the number from step 5 (-24.6 dB).
7

So now all we need is scripting support for effects :wink:

8

make a note of the default amplification amount (as an example, let’s say that it is -23.6 dB)

You’re assuming I’m doing the whole thing in 32-floating?

Koz

9

Of course. With 37 tracks you wouldn’t want 37 instances of quantize error would you?

10

Oh, right.
Koz

11

Three very good reasons to use 32-bit float for multi-track projects:

  1. Safeguard against clipping when mixing.
  2. No dither noise when processing tracks.
  3. With a lot of simultaneous tracks, the tracks will probably need to be at a very low level (to avoid clipping the mix). 32-bit float preserves the dynamic range and SNR.