Following the Letter of the (Pan) Law
Have you ever moved an audio or video project from one application to another or performed a mix down of your tracks, only to have the volume be too loud or too quiet for no apparent reason?
Dan Ramirez from the After Effects team brought it to our attention that he has recently been fielding questions about differences in audio levels when moving a project from one NLE (or DAW) to another. We in the Audition team thought this was a very relevant and interesting topic as many of you may also encounter problems with your audio levels such as quiet/loud vocals/narration, possibly even clipping when you know the files are OK, all resulting in uneven audio mixes. These problems often arise because the new application uses a different “pan law” than the application you originally mixed your audio with.
The pan law (or sometimes called “pan rule” or “panning law”) simply determines how an audio signal is modified across the Left to Right stereo field. To illustrate, if you pan a track to the right, the sound comes out your right speaker only, then if you pan the track to the center, the sound comes out of both speakers. If the same volume level was used when panned right and also in the center, the center would be louder than the right because the same level is now coming out both speakers, not just one and yet you haven’t made any changes to your mix.
The pan law creates a consistent and smooth volume transition when panning from right or left to the center in a stereo field. This compensation is done by reducing the level of the audio channel the closer it’s panned to the center. Typically the signal is lowered as it approaches center using a sine/cosine curve ranging from -3 dB to -6 dB (this sometimes is done using a logarithmic curve as an alternative). There was research completed in the 1930’s by Disney which suggested that -3 dB was optimal for the listener, however a different study done by the BBC in the 1970’s suggest that -4.5 dB was better. What is “best” is up for debate and will depend largely on the listener, listening environment, speaker placement and the speakers themselves (including headphones). See the reference bullets at the end of the post for more detail on this.
I should note this does not just apply to software, but also to external mixing consoles (digital and analog) which have since the beginning of stereo audio employed the same pan law principals, for example SSL consoles use a -4.5 dB pan law and Yamaha digital consoles typically use a -3 dB pan law.
Applications like Adobe Audition, along with other NLE’s and DAW’s, allow you to change your pan law settings to resolve the differences and ensure you get quality mixes for your project. Getting familiar with these settings can help you get past the clipping audio channels and back onto making creative decisions.
Accessing and changing this option in Audition 3 is easy:
- Go to “View” on the top menu bar
- Choose “Advanced Session Properties…” (or you can get to this using the shortcut CTRL+P)
- Select the “Mixing” tab:
References for further reading…
- Gerzon, Michael A. 1992. Panpot laws for multispeaker stereo. 92nd Convention of the Audio Eng. Soc., Vienna. Preprint 3309.
- Holman, T. (2000) 5.1 Surround Sound Up and Running. Focal Press Rumsey, F. (2001) Spatial Audio. Focal Press.
Of course, surround sound mixing adds yet another dimension to this, but that’s a topic for another day…
Colin Stefani / Sr. Program Manager – Audio
The Adobe Audition Team