Steps to mixing multitrack recordings

[TunnelThroughTime]

Need a little input on effective steps to mix some multitrack recordings.
The sequence that I've been using involves (1), applying Noise Reduction to all tracks to remove background noise,
(2) applying equalization to track based on timbre of instrument in each track,
(3) Compressing tracks to even out peaks which would latter cause clipping when tracks are mixed and rendered,
(4) Normalizing each track to -6 to -25 db,
(5) Mixing down tracks?

[cyrano]

That sounds like the proper procedure. What's the problem?

[kozikowski]

You're listening to all this on very good quality speakers or headphones, right?

When you Export the work, Audacity will mix it down for you to a stereo show. Use WAV (Microsoft). Do Not use MP3. Make the MP3 later if you need one, but the music archive/backup needs to be in WAV.

You can do Mix and Render to a stereo track before you export, but I'm a good deal less familiar with that tool.

http://manual.audacityteam.org/o/man/mixing.html

Koz

[TunnelThroughTime]

Before I export the the Audacity Projects (Each song I've recorded I've given it's own .AUP file titled with the name of the song) I"ve used the Mix and Render function to
check for clipping, and one question I've had in regards to this is whether Audacity itself changes any of the gain levels when to Exports files to .Wav or .Mp3 since
I've taken measures to prevent clipping when mixing the the projects myself. For simplicity's sake I've taken the liberty of working solely with mono waveforms since I've read that Audacity will produce a Stereo Mix-down when you finally decide to Export the project.

"When peaks or troughs in the waveform coincide, the waveforms will reinforce each other, leading to an increased signal level. In fact if you combined two identical tracks, the signal level would exactly double, leading to an increase in peak level of 6 dB. But when a peak in one track coincides with a trough in another track the waveforms will tend to cancel each other out, leading to a lower level in the mix at that point."

Hypothetically, this means that the dynamics of a project recorded with 8-tracks would be greater than that of a 4-track recording; In such a case would the track be more
radio friendly (i.e., have greater clarity with sub-par playback equipment) if it's mixed down single stereo track finished product where again compressed and amplified?

In answer to the your reply I own a pair of sony stereo headphones MDR-XD100 which I've been using when I can to listen to the projects as I mix them down.

[steve]

I"ve used the Mix and Render function to check for clipping,

That's generally a good idea, but remember that Audacity projects do not save the "Undo" history.
I would recommend that before you save the project, undo the "Mix and Render" so that your project still has separate tracks (just in case you need to adjust the mix later).
Alternatively, use "Mix and Render to New Track" to check the levels. Then you can either solo the mixed track when you export (to export only the mixed track) or mute the mixed track (to export everything except for the mixed track).

whether Audacity itself changes any of the gain levels when to Exports files to .Wav or .Mp3

Mute, Track Gain and Track Pan are applied to the exported file in the same way as they are applied to playing the project. The exported file should sound virtually identical to the project (depending on the quality of the format - MP3 and other compressed formats always lose a bit of sound quality).

For "lossy" formats (such as MP3), the peak amplitude may slightly different from the peak amplitude of the mixed track in Audacity. This is because lossy formats are inexact. The difference is usually too small to make a noticeable difference, but you may want to allow a little more "headroom" to your final mix if you intend to make an MP3 version so as to ensure that the peak level does not exceed 0 dB. I generally allow 1 or 2 dB of headroom. 3 dB of headroom (Amplify or Normalize the mix to -3 dB) is virtually guaranteed to be enough to not clip in any format, but is probably a bit excessive.

I've read that Audacity will produce a Stereo Mix-down when you finally decide to Export the project.

Export will mix down to mono or stereo, depending on the project.

If the project is entirely mono tracks, and all tracks are centre panned, then the show is mono (what plays through the left speaker is identical to what plays through the right speaker). In this case, the exported file will be mono. Similarly, "Mix and Render" will produce a mono track in this case.

If the project includes any stereo tracks, or if any of the tracks are panned off-centre, then the exported file will be stereo (provided that the file format supports stereo, which most formats do), Similarly, "Mix and Render" will produce a stereo track.

Hypothetically, this means that the dynamics of a project recorded with 8-tracks would be greater than that of a 4-track recording;

Probably not. The "dynamic range" is the difference between the maximum peak and the level of the noise floor. Increasing the number of tracks will often increase the noise floor, so you need to be extra-careful about minimising noise when using lots of tracks.