SamplecrazeStretch That Note

The Panning Law

So, what is Panning Law and does anyone really care?

Hell yeah!  

In fact, I can state that many producers/engineers that I meet have little or no clue about the panning law.
As far as they are concerned, you turn the knob and the sound moves from one side of the stereo field to the other.  

What is not understood is that certain things happen to the sound when it is panned from one side of the field, through the centre, and then to the other side of the field.

When you are dealing with monaural sounds you have to take into account the dynamics of the sound when it is moved and summed. This is very important when it comes to the mix stage as many people seem to complain about how the sound behaves when it is panned, to the point that most software developers have tried to accommodate for the panning law in their coding.  

The problem facing most newcomers to this industry is that once a project is mixed in certain software and the project is then imported in separate mix software, the panned levels go for walkies. This is down to how the softwares behave and process the panning law, compensating for the characteristics of the process. 

When a signal is panned centrally,  the same signal will be output (identically) on both the left and right channels. If you were to pan this signal from the extreme left channel through the centre and then onto the extreme right channel, it will sound as if the level rises as it passes through the centre. The panning law was integrated to introduce a 3dB level drop at the centre. If you were to sum the left and right channels in a mono situation, the centre gain would result in a 6dB rise, so attenuating by that amount became a must in the broadcast industry as mono compatibility is always a prime issue.  

So, what the hell is the panning law?  

The panning law determines the relationship between the sound's apparent image position and the pan knob control. This refers to the way the sound behaves when it is moved across the stereo field. The usual requirement is that it moves smoothly and linearly across the field. This is, of course, pertinent to log/anti-log laws. If there was a linear gain increase in one channel and a linear gain decrease in the other channel to change the stereo position, at the centre position the sum of the two channels sounded louder than if the signal was panned full left or full right.
Why do you think we had to attenuate the gain whenever we panned a sound central?

Digital consoles and the digital domain started to change this thinking and accommodate and compensate for this behaviour.  

It became necessary to attenuate the centre level by four common centre attenuation figures: 0, -3. -4.5 and -6dB. The -3dB figure is the most natural because it ensures that the total acoustic power output the studio monitors remains subjectively constant as the source is panned from one extreme of the stereo field to the other. However, it also produces a 3dB bulge in level for central sources if the stereo output is summed to mono, and that can cause a problem for peak level metering for mono signals. So, most broadcast desks employ a -6dB centre attenuation so that the derived mono signal is never louder than either channel of the stereo source. However, sounds panned centrally may end up sounding a little quieter than when they are panned to the edges.

Confusing huh?  

Well, the answer is to simply compromise and split the difference and this is what led to most modern analogue consoles working off a -4.5dB centre attenuation.  

So, what does this mean to you and how does it help you, particularly if you are working ITB (in the box) and with different softwares? The answer is quite simple: find out what the software panning preferences are and adjust to taste. Most of today’s softwares will allow for fine tuning the panning law preferences. Cubase, along with most of the big players, has a preference dialogue box for exactly this. Cubase defaults to -3dB (classic equal power), but has settings for all the standards and I tend to work off -4.5dB.  

If you stay with the old school days of 0dB, then you are in ‘loud centre channel land’, and a little bit of gain riding will have to come into play.  

Check your software project and make sure you set the right preference, depending on what the project entails in terms of the mix criteria.