Hi again everyone. I tried one thing, maybe I’m not using Audacity correctly. I tried the “Normalize” feature, it was set to -1dB. I figured I’d include the DC offset function. My test tracks are a quantity of 16, about 4 minutes long. With my old Acer laptop, it took forever to accomplish, one track at a time. When finished, it boosted all waverforms/tracks to, I guess, -1dB. It was overpowering LOUD. Also, this is the same reason why I shy from the “Amplify” feature, it takes quite a while on a slow computer. Thank you.
The problem with “Normalise” is that it treats each track individually and you’ll lose the balance of the tracks–they are all -1 dB now.
In order to normalise the project, you must use “Amplify”.
You’ll first render all tracks into one.
It’s best to mix and render them to a new track (Ctrl+Shift+m).
Select all tracks and open amplify, add the -1 dB to the proposed factor (-11 dB → -12 dB), Ok.
You can now delete the mixed track.
The whole project has now -1 dB overall while the relationship is still the same.
The settings of the gain sliders and the exclusion of muted tracks are included in this method.
The normalise button that I’ve proposed above (next to the master slider) would just do that:
- Mix down the tracks in the background
- The master slider would now have the negative found maximum peak, that is, 0 dB minus x dB.
You would afterwards decrease the master gain by an additional 1 dB in order to have this head room.
Audacity wouldn’t of course mix down the track physically, there are efficient structures (Fennwick tree for instance) to keep track of the running peaks in each track on import/creation/modification/playback.
On a slowish 2.4 GHz 6 GB RAM (HDD) Windows 7 x64 computer I find Amplify a little quicker on 16 four-minute tracks than default Normalize (including DC offset).
You don’t say what version of Audacity you have now, but please try some direct comparisons between Normalize and Amplify using the latest 2.1.0 ( http://audacityteam.org/download/ ). However as Robert says you must use Amplify to retain the volume balance between multiple tracks.
I’m away from home, so I can’t give an accurate version number.
I think I saved the flack file of where I began to doubt the limits of Amplify and/or what I was doing. Not sure I can post a snippet link of what I heard.
Also, if I may, something strange about Normalize. It was set to -1dB. I let it do its thing to all 16 tracks. The waveforms looked normal; greater than what they were before. However, when I went to play all the tracks, I heard a VERY LARGE amount of tape hiss noise, and quickly stopped the playing, fearing my hearing would be lost from the extreme blast of music. Give me a chance, maybe I can help via screen shots and output audio.
As Robert explained, Normalize sets each track to the same level. If you then have 16 tracks each at -1 dB, their volume will combine and the result will be several dB over the maximum level of 0 dB. You have also lost whatever intentional difference in volume there may have been between the tracks.
So you will need to do what Robert explained, mix the tracks to a new track. This lets Amplify calculate the correct negative gain to apply to all the tracks to bring their combined volume to 0 dB, without altering the relative volume that the tracks have to each other.
If instead you wanted to bring the combined volume of the tracks to -1 dB, you would choose -1 dB in the “New Peak Amplitude” box of “Amplify”. The simplest thing to do rather than us try to explain it, is to just try it yourself then you will see it working.
…It was overpowering LOUD. Also, this is the same reason why I shy from the “Amplify” feature
…fearing my hearing would be lost from the extreme blast of music.
Strange… The complaint is usually that it’s not loud enough.
You can try Normalizing or the Amplify effect set for -6 or -9dB, or whatever sounds right to you. As the others have said, if this is an album and you want to maintain the loudness relationship between the tracks, you need to change the gain of each track by the same amount instead of to the same peak level.
Most commercial modern releases are normalized to 0dB. The peak level does not correlate well with perceived loudness, but most modern releases are also highly dynamically compressed to make them constantly loud ([u]The Loudness War[/u]).
…I heard a VERY LARGE amount of tape hiss noise
Of course, the background noise is no worse than if you were to turn-up (or down) the analog volume. Turning down the playback volume will reduce the signal and the noise together.
In technical terms - Increasing the volume digitally does not degrade the signal-to-noise ratio (as long as you don’t push the level into digital clipping by trying to go over 0dB).