VO Studio Tech: What do We Get With 32 Bit?

Last week I dove deeply into the way bit depth works when recording audio. There are concrete benefits to increased bit depth. Using 24 bits to capture our raw voiceover recordings provides plenty of accuracy and dynamic range for anything we need to deliver. As long as we don’t hit 0 dB in our recordings, our audio should be usable.

Assuming that we’ve taken the steps to isolate and tune our recording space, setting input gain becomes the only technical variable we must control. With its ability to represent a wide dynamic range, 24 bit allows us to set a conservative input gain on our interface. Whether the recording is peaking at -12 dB or -18 dB becomes less critical. We can always boost things later. It’s easy to adjust our final Loudness for the needs of our clients, and we can always downsample to 16 bit if that is part of our delivery spec.

Why then would we mess with 32 bit floating point recording? It has one really amazing trick up its sleeve, but first let’s understand what it doesn’t do.

32 bit won’t sound “better”

Recording in 32 bit won’t make things sound “better.” This higher bit depth will still accurately record all the flaws in our space, the noise in our environment, and any weird interference patterns that result from electronic devices. There won’t be lower levels of noise in any practical way. 24 bit recording already has a theoretical noise floor below perceptible levels. If you hear noise in your recordings, it is likely coming from recording hardware or the environment rather than the bit depth. We don’t gain better sound quality with 32 bit.

32 bit doesn’t fix space issues

When Rode announced that their new NT1 Generation 5 microphone contained a 32 bit recording option, some stated it would be an ideal travel microphone, which doesn’t quite follow. While you can plug the NT1 Gen 5 directly into your computer via a USB-C port, it remains a large diaphragm condenser with a cardioid pickup pattern. As with any sensitive studio microphone, it will easily capture background noise and echoes in an untreated recording environment. Since the quality of the recording space is usually more of a limiting factor than proper input gain when recording “on the road”, it doesn’t seem as though this offers a solution. 

I’ve been putting the new NT1 Gen5 through its paces recently. With good tone and a balanced response in the vocal range, it’s a solid option for a home voiceover recording setup. In USB mode, it does allow some inbound processing, but that won’t fix echo or excessive environmental sound. 

How 32 Bit Floating Point works

Every chunk of 32 bit data is called a “word” and each word has 32 bits of information. When recording digital audio, each word actually encodes the audio waveform information using only 23 of the available bits. This part of the data is called the “mantissa” and essentially represents the shape of the wave. Subtracting 23 from 32 leaves us with 9 more bits to use for other information. Those remaining bits manage volume information separately. That’s the important difference. 

You can think of the 32 bit data as recording the shape of the wave separately from the volume of the audio. Using only 23 bits to capture the waveform data is actually less precise*, though not in any way which we might notice. The waveform and the volume are separate data values, which makes clipping impossible. 

Gain Recovery: 32 Bit Floating Point Recording is indistinguishable from magic.

32 bit’s ability to “re-gain” seemingly clipped recordings is seriously mind-blowing. In the past, when you came out of the booth and found audio above 0 dB in your DAW, there was nothing to do but throw your hands up and say “Oh, well…” before turning down the input gain even further for the next take. We know audio distorted on the input is likely not fixable.

With 32 bit floating point recording, you would simply adjust the track’s clip gain or just apply a Normalize function. And, what to your wondering eyes should appear? A glorious audio wave with no distortion or clipping! You can simply adjust your “input” gain after the fact. 

The Rode NT1 Generation 5 microphone – a few caveats

There are a few things to keep in mind with the new Rode NT1 Generation 5 microphone. First, though it has both an XLR and a USB-C output, it works in only one mode at a time. You cannot simultaneously get both an XLR and USB-C output from this microphone. The 32 bit mode only operates when it is connected via the USB-C output. As of this writing, the 32 bit mode is not compatible with all recording software. 

There is also no direct headphone output from the microphone itself. To monitor yourself when in USB mode, you need to use the free Rode Connect app to do so. In my studio setup, I ran this back to my headphones, which were connected to my regular SSL interface. This provided direct input monitoring with a very minor latency. Note that things may not sound great if the input gain is high enough to appear to clip. You may distortions that will be corrected later when you drop the volume. 

Finally, since the microphone is supplying digital data directly to the computer, there’s a potential issue in running the USB-C cable. Rode supplies a 3 meter USB-C cable, and when asked, they specifically recommended against using a longer one. A shorter cable can make attaching the mic to your system a little frustrating – in my case, I would need a pass-through on the wall nearest my studio computer to keep it from getting closed in the booth door. It did not work through a multi-port hub connected to my Mac Mini (Intel) and needed to be connected directly to the computer’s USB-C port. 

I’m continuing to experiment with this microphone and look forward to putting it through its paces.

*Note: After sharing this to my email community, I’ve been told that my description of 23 bits being theoretically “less precise” is actually incorrect. There’s a bit of math involved in their explanation which I honestly don’t quite follow yet. But, I am looking forward to understanding that. More as I know more.

Want to receive these resources (and more!) the day they come out?
Each week I send out a new article to my email community focused on creatively using recording technology in your home voiceover studio, and how to balance those technical challenges with the need to be brilliant behind the microphone.
If you would like to join in to receive those emails the day they publish, please take a moment to share your contact information through this sign up form.
Thank you!

If this resource has been helpful to you, please consider sharing it with one of the buttons below!