On a slightly larger scale, a whisper can be around 20 to 30 dB, while a typical conversation is around 60 dB. A motorcycle drive would be around 90 dB and a really loud concert might be in the 110 dB region. Much higher than this and you are falling into a range where the sound becomes physically painful. However, if that is the case, why would any recording equipment need to go beyond the 144.5 dB dynamic range of 24-bit audio?
Level setting (or not)
The insane dynamic range of 32-bit stereo is where the claim that you never need to set levels, though that’s a bit more complicated than it seems. The highest level the device can record is called 0 dBFS (FS here stands for “full scale”). Anything higher than this will be clipped, that’s why the sound is distorted when YouTube users scream like that.
Now, normally you would set the sound level when setting up your device to avoid reaching that limit. Setting those levels involves applying gain to the signal from the microphone, which is an irreversible step that will disrupt the dynamic range of even 24-bit recording.
“When you’re recording on set, you’ll usually apply gain. And some recorders will apply increments of 30 to 90 decibels,” explains Pereira. “It’s good when you have a quiet setting where two people are whispering. So you can rotate the dial on the recorder, say plus 60 [decibels]. So now when someone decides to scream, between 60 and 145, that’s not a lot of dynamic range. “
On the other hand, with 32-bit float writing, applying a gain before writing is not necessary. “When you record in 32-bit, with no volume knob, it basically just creates a mathematical graph of the data that it can then interpolate in post-production,” says Pereira.
Setting levels on a 24-bit system can be complicated because of the noise layer. To oversimplify, no matter how quiet you make your recording space, there will always be some noise from background objects or even the electronic devices you record on. . Adding gain to the signal while recording amplifies that noise, as does your audio source, and once it’s included in the recording, it stays there just fine.
The 32-bit float record has more flexibility to make adjustments after the fact (and in some cases it can even help solve low-noise problems). That said, it’s important not to let this give filmmakers and audio producers a false sense of security. “It won’t fix the inherent problems of your filming location. For example, if there’s an AC fan or a large fan nearby, recording at 32-bit won’t make that noise go away,” says Pereira.
Managing noise on set and making sure the microphones are picking up the signal properly will always be important, but once your equipment is set up properly, the ability to pick up sound even past its peak is a big deal. Useful tools. But don’t expect to leave 24-bit audio behind forever.
Links in the chain
So if 32-bit is so great, why isn’t it the default? For starters, many production steps – including editing, mixing, and especially distribution – will use a 24-bit workflow, which means that at some point additional data will be lost. And an audio engineer at some stage will need to make adjustments to ensure that the audio signal doesn’t get clipped when downsampled to 24-bit, like it would if the levels weren’t set properly. during the initial recording.
Essentially, this means that work that would have been done initially on set will be moved to post-production. So you have a choice: Set the correct levels on the setup and record directly in 24-bit, or write in 32-bit and add the following extra step. One way or another, it’s a step you’ll have to take, and some would argue that you can do it too while you’re on set.
