In the United States, movie theaters have to operate under federal Occupational Health and Safety regulations (OSHA), which include sound limits. You as a filmmaker/copyright owner can't be held responsible if someone outside your control cranks up the volume in your creative work and injures someone, unless you specifically (and negligently) give instructions to do so.
The legal limits used by governments are all based on exposure time of constant noise (such as machinery in say a factory). I think the OP now realises he's extremely unlikely to kill any one with a sound mix in a cinema, although it might be possible under some circumstances (without turning up the cinema's volume)! More pertinent to most filmmakers is that it's very easy in a theatrical situation to cause considerable discomfort or even pain. I've witnesses this myself at a lower tier film festival where a film managed to completely empty the auditorium before the opening credits were over, the audience literally ran out of the cinema faster than if the fire alarm had gone off!
Nevermind... I figured it out.
Sound level doubles every 6 db, thus 24 bits means 6*24, thus 144 db range.
Although the linked article is simplified, it is essentially correct. Each additional "bit" of data results in a halving of the noise (-6dB) and as dynamic range is defined as the range between the noise floor and the peak value, halving the noise floor results in double (+6dB) the dynamic range.
Your previous assertion regarding dBA is not correct though!
The dB scale/s can be quite confusing even for experienced pros. The dB scale is just a logarithmic scale used to express a ratio, a relationship relative to some defined point/value. We use the dB scale as a unit of measurement for audio energy but as there are different kinds of audio energy and different aspects to energy, there ends up being a whole bunch of different dB scales. There's a group of dB scales called weighted or filtered scales. This group exist because of the non-linear nature of human hearing and therefore the disparity between energy measurements and what we hear. For example, doubling the sound level (+6dB) does not result in double the loudness. So, we've invented various dB scales to account for the various non-linearities of human hearing. dB-A, dB-B, dB-C, LUFS/LKFS and dBVU are all examples of scales which have been weighted to try and take account of human hearing and therefore give some indication of loudness. I say "some indication" because some are more accurate than others but most of them are rather inaccurate. The dB-A, B, C, D and Z scales are different weightings applied to the basic dBSPL scale, the scale used to measure the energy of actual sound waves in the air (Sound Pressure Level). Our response to frequency actually changes as the SPL increases, which is why there are 5 different weightings, in an attempt to account for this. The dBA weighting is designed for use on low levels signals and is poorly suited to measuring dynamic range. In fact, the dBA weighting has got to be about the most abused scale in the history of scales! Environmental and health agencies, government departments and international organisations all publish data, recommendations and limits in the dBA scale even though they and every other audio pro in the world knows that it gets more and more inaccurate the higher the SPL and not by a little bit but out by as much as 100% or so! That's why many audio manufacturers like to use dBA for noise and dynamic range specifications, it makes their equipment specs look much better than they actually are!!!
G
