Topic: What is 32bit audio?

[DSatz]

Point well taken, Petrus. Some people seem to have the remarkable idea that dynamic microphones (including ribbons) have no inherent noise just because they don't contain active circuitry.

In fact the inherent noise of a dynamic microphone tends to be audibly more disturbing than that of a condenser microphone, since the frequency spectrum of resistive noise is weighted toward the higher frequencies like pink noise, rising throughout the sensitive region between 2 and 8 kHz (as well as beyond, though we don't tend to hear that so much).

The preamp also has to be turned up much higher for a dynamic microphone, so in a complete recording system (as opposed to the microphone alone, out of context) the noise is higher for that reason as well.

--best regards

[KLowe]

digifish, this thread is getting sidetracked by needlessly provocative postings. Let's try to stay on the subject, or else start a new thread called "How can I argue about 24-bit recording?", OK?

As to the point of all this: Very few people listen to 24-bit recordings as such; much more often, 24-bit recording is used for "sound acquisition," and the result is then normalized in level and dithered down to 16-bit for listening and delivery.

If you record in a 16-bit medium for 16-bit delivery, you must set your recording levels as high as possible to get the best possible signal-to-noise ratio. But in live recording, this exposes you to the risk that unexpectedly loud sounds will ruin your recording. The risk can be reduced through experience but never eliminated. It makes live recording a tense affair at times, especially when--as with most recordings that I make--one is the only person recording the event. (I record classical concerts and recitals professionally.)

24-bit recording is extremely useful for live performance recording because it allows you to set recording levels conservatively without giving up the full signal-to-noise ratio of the eventual 16-bit delivery medium (CD or whatever). It takes away the fear, increases the reliability of the process and restores the fun. To me that's been worth a lot.

--best regards

Once again.  DSatz is in my brain.  Could not have said it any better myself.  At 16 bit recording I used to clip and stare at the levels....took the fun out of recording the show....stressful.  24bit recording....set the levels conservatively and forget it....time to ENJOY the show.

I (heart) 24 bit recording

[datbrad]

If you record in a 16-bit medium for 16-bit delivery, you must set your recording levels as high as possible to get the best possible signal-to-noise ratio. But in live recording, this exposes you to the risk that unexpectedly loud sounds will ruin your recording. The risk can be reduced through experience but never eliminated. It makes live recording a tense affair at times

This is one of the reasons that when I was looking to move to a 16 bit tapeless recorder, I wanted one with a margin reset. Using this tool, you can monitor the peaks without having to stare at the levels to make sure you are getting as close to 0db consistantly, without going over. I definately agree that there is still a good amount of "work" involved with making a good 16 bit master in the field, compared to 24 bit. Being able to have levels that are peaking just over -12 at the show, and being able to boost them to optimum level in post before dithering to 16 bit certainly would take some of the tension out of field recording IMO. More time to watch out for the drunks.

I think any person just getting into taping today should go with a 24 bit recorder right off the bat, so they will not have to go through the steep learning curve required for getting the best results with 16 bit in the field.

[trajhip2000]

I may be showing my ignorance here (and taking another tangent), but since it was mentioned, 32-bit and 32-bit float are two different beasts, no? Doesn't processing in 32-bit float have the very real advantage that you don't have to worry about an intermediate processing step taking your audio over 0 dB and causing clipping?

Steve

[DSatz]

trajhip, you're right in a general sense. When we use editing software that is not open-source, we don't really know whether the code for various program actions (such as filtering and equalization) was or wasn't written with the proper care to avoid internal numeric overflow or underflow (the digital equivalent of brickwalling).

However, floating point representation has its own problems, including a kind of modulation noise (or noise modulation, depending on how you look at it) which occurs as the exponent varies. So the cure can definitely be worse than the almost completely avoidable disease.

It should be possible to test the various processing routines in programs such as Audition or Sound Forge, for overflow/underflow problems, but I've never seen this done in any published review of such software; I don't even know of any systematic methods for carrying out such testing, though I would think that they must surely exist somewhere.

--best regards