Topic: Recording in 24-Bit vs. 16-Bit

[Dan33185]

This is probably a really dumb question, but I have been having issues with clipping and distortion with a few of my recordings recently, which I record at 16-Bit. I am wondering if recording at 24-Bit would reduce the risk of clipping and distortion, or would it not make a difference?

[rocksuitcase]

need more details:
1] internals or external mics?
2] You using a battery box or pre-amp if external?
3] From your profile sig, if you are using the Zoom by itself with internals, that is the clipping in the pre-amps, the bit rate will not change that. If you are using it with external mics, then the clipping could also be about sensitivity of said mics. but again, not bit rate related.

[Dan33185]

need more details:
1] internals or external mics?  (Internals with the Zoom ; XLR > SBD with the Tascam)
2] You using a battery box or pre-amp if external? (See above)
3] From your profile sig, if you are using the Zoom by itself with internals, that is the clipping in the pre-amps, the bit rate will not change that. If you are using it with external mics, then the clipping could also be about sensitivity of said mics. but again, not bit rate related.  (That's what I was afraid of, but thought it was worth asking...thanks!)

[Gutbucket]

As rocksuitcase mentions, choosing to record at a different bit-rate will not fix your problem.  It can be difficult to determine where overload distortion is occurring.  However in your case, since you are using two fully integrated stand-alone recorders, it can only be occurring either before the signal reaches your recorders, or inside of them.

Sometimes the sound through the PA (and potentially through the SBD) is distorted before it reaches your recording gear, yet due to hearing perception and loudness masking it might not be obvious during the performance, and only becomes more obviously apparent upon listening to the resulting recording. In such a situation you may be making a faithful high-fidelity recording of distortions that have already occurred, over which you have no control.  That might not be what's going on in this case, but is one possibility to be aware of.

Lets assume instead the signal reaching your recording rigs (through the air and/or through a SBD patch) was not overly distorted before it got to them.  In that case, the distortion must be occurring somewhere in your signal chains. And in the setups you describe, the entire signal chain other than the XLR cables from the SBD is located inside of the recorders.  There are several separate stages inside the recorders in which distortion is mostly likely to occur: The analog to digital converter stage, the analog input stage, and (on the Zoom at least) the internal microphones.

First thing to check is sufficient powering, the lack of which is a common cause of distortion. Are the batteries fresh and providing sufficient voltage and current?  If they are, you'll need to reduce signal level through whichever stage inside the recorder is distorting. Do so by setting input sensitivity lower (confirm low microphone sensitivity is selected instead of high, switch to line-in rather than mic-in if possible), and make sure your recording levels are set sufficiently low.  If not possible to reduce levels further via the settings of the recorders, you might use signal attenuators in the XLR connection between the SBD and Tascam. With the Zoom using its own internal microphones, it might be sufficient to move the recording position farther away to where the SPL is lower.  Yeah, the recording is likely to suffer from doing that, but is better than unacceptable distortion. Alternately you might switch to using external microphones capable of handling the SPL, the sensitivity of which are low enough so as to not overload the input stage of the Zoom. 

If none of that works or is doable, you'll need to switch your current recorder(s) out for something else capable of handling higher signal levels without distorting.

[rocksuitcase]

I like also what GB is telling you. THIS, in specific may be the case, especially SBD output distortion:

Sometimes the sound through the PA (and potentially through the SBD) is distorted before it reaches your recording gear, yet due to hearing perception and loudness masking it might not be obvious during the performance, and only becomes more obviously apparent upon listening to the resulting recording. In such a situation you may be making a faithful high-fidelity recording of distortions that have already occurred, over which you have no control.  That might not be what's going on in this case, but is one possibility to be aware of

.

[Gutbucket]

And to answer your actual question, rather than helping trouble shoot the cause of distortion, here is why recording in 24 bit rather than 16 won't help-

It is true that a 24 bit digital audio file is capable of storing a signal of wider dynamic range than a 16 bit file, so its reasonable to suspect it might help in this case.  But the overload point of a 24 bit file is exactly the same as a 16bit file. Both max out at 0dBFS. What changes is that the noise floor of a 24bit file at the quiet end of things is significantly lower than that of a 16-bit file. The point of clipping at the loud end is the same, but the level of noise (hiss) at the bottom is lower.

To take advantage of that for recording louder things, the signal chain ahead of the digital file being written to memory needs to be capable of handling the higher signal levels involved before overloading (that is the key in this case) while sending a lower signal level to the ADC which converts the analog signal to digital, thus keeping the the digital signal from exceeding 0dBFS. But that much applies to both 16 and 24 bit.  To take advantage of the greater dynamic range of the 24bit digital file format, the signal chain ALSO needs to have a low enough electrical noise-floor that the extra dynamic range provided by the 24bit file format at the quiet end of things doesn't just end up getting filled with a bunch of random noise.

Its sort of a Goldilocks thing - comfortably fitting the dynamic range of the analog signal into the digital file container without exceeding its limits at both the top and bottom.  With 24 bits, the limit a the top is the same as with 16 bits. The limit at the bottom is lower.  That extra range at the bottom may not be useful.  It won't be useful if the electrical noise-floor of the signal chain through the recorder is already higher than the noise-floor of the 16 bit file format, which is not unusual. And it won't really be useful if the acoustic noise-floor of the recording environment is higher than that as well, which is pretty common.

[Dan33185]

Thanks for all the information...I know it's not a power issue, batteries are freshly charged every time I head out to a show and replaced every so often because as they age the charge obviously doesn't hold as long. I'm thinking I might just need to lower the recording level, I was experiencing hiss with some recordings because I had to amplify it so much in post that I was trying to minimize the need to do that by raising the recording level, but I think I over estimated how much I could. I'll play with some levels as I go to shows and hopefully figure out a happy medium. I just figured if a higher bitrate took some of the guess work out of it, that could be helpful, but that won't make a difference as I'm discovering. The plan is to upgrade my recorder and get externals at some point when finances allow, but just trying to make the best of what I have at the moment.

[Gutbucket]

Was adding a bit to the bottom of my previous post when you replied which somewhat addresses that.

In short, you do need a file format that provides sufficient dynamic range in which to fit the music.  But 16bits provides a pretty wide range.  The dynamic range of the music itself is defined by the acoustic noise-floor of the recording environment at the bottom, and the loudest sound being recorded at the top of that range. Its almost always smaller than that of the 16 bit file format. But you do need to fit it in there comfortably, so 24bits makes fitting it in easier while requiring less adjustment of levels, yet only if everything else in the chain is capable of handing that additional range as well.

The actual thresholds of hiss and overload in your recordings are probably not defined by the bit depth of the file format but rather by the circuitry which precedes it.. or the recording environment itself at the quiet end.  The noise-floor at the bottom could be the acoustic noise-floor of the recording environment, which unless dominated by background music, folks mulling around talking and making noise, is most often dominated in otherwise quiet sounding recording environments by acoustic noise that sounds like hiss when amplified.  Sometimes hard to tell what is dominating the noise-floor of the recording - the self-noise of the mics, electrical noise of the gain-stages, or the acoustic noise floor of the room itself.  They can all sound like hiss.

[robgronotte]

To get this out of the way, look at the wave form in any music editing program while listening. If it's only sounding distorted when the levels are peaking at 0, you had the levels set too high.  If you're hearing it other times, look at what the guys said above.

[Dan33185]

To get this out of the way, look at the wave form in any music editing program while listening. If it's only sounding distorted when the levels are peaking at 0, you had the levels set too high.  If you're hearing it other times, look at what the guys said above.

Good tip, I'll check that on the next show I run through Audition.

[rocksuitcase]

To get this out of the way, look at the wave form in any music editing program while listening. If it's only sounding distorted when the levels are peaking at 0, you had the levels set too high.  If you're hearing it other times, look at what the guys said above.

Good tip, I'll check that on the next show I run through Audition.

I was going to add this IS a good step to go through before you decide what is causing the distortion.

[Gutbucket]

Yeah that!

If so all you may need to do is set levels somewhat lower to avoid exceeding 0dBFS, just not so much lower that the quiet parts begin to get swamped in noise.

[DSatz]

Digital audio recording is easiest to understand correctly if you think of the samples as fractions rather than integers. They're binary fractions, but that's a technicality. The number .25 has the same magnitude as the number .250, but when you see the second as opposed to the first, you have _less uncertainty_ about what's going on in the "thousandths' place". That corresponds to a lower "noise floor" in the purely digital part of each recorder channel--though not its analog circuitry or A/D converter (where any audible noise in a recording usually comes from, if not from the venue or the microphones).

24-bit recording offers you no greater capacity to record higher numeric values (representing greater sound pressure levels) than 16-bit; ±1 is still the absolute outer limit that you don't want to crash into. But up to a point, if your signal path is generally sensible, 24-bit recording can let you reduce your record levels enough so that you don't run out of room at the top, and still not pay a price in increased audible noise. Not always--your mikes themselves and/or your outboard preamp, if you use one, can be overloaded before the signals get to the recorder; again, the ceiling isn't being raised but the floor is being lowered, one may reasonably hope.

The amount that lowering is never anything like what the additional 8 bits would imply (48 dB!), though, because no A/D converter or analog input circuitry has that much dynamic range. One well-known recorder that I measured some years ago gave only <2 dB improvement when going from 16 to 24 under the particular conditions I was measuring it. It wasn't a typical scenario, but it does illustrate the real dependency that exists on all the other factors in your actual scenario. If someone had used that recorder with those signal levels and gain settings except that they thought, "With 24 bits I can reduce my levels by (say) 12 dB to be absolutely safe from overload", their 24-bit recording would then have been ~10 dB noisier than the 16-bit one. Again that is based on an atypical scenario in terms of signal levels and gain settings, but not impossible by any means.

Thus you should always use the input to the recorder and the menu and level settings, if any, that are most appropriate for the signal levels you're dealing with, and always try to get your absolute peak sample values somewhere in the top, say, 6 dB or so of the recorder's range without hitting 0. It's just that with 24 bits as opposed to 16 (and with everything else optimal, please note) there should be no audible "noise penalty" if your highest peaks are only at, say, -8 dB rather than -2. So by setting somewhat conservative levels when you don't know what's going to happen (i.e. life), you're better prepared in the event of the unpredictable. Murphy's Law says, after all, that the unpredictable is the most predictable thing there is.

[Gutbucket]

The amount that lowering is never anything like what the additional 8 bits would imply (48 dB!), though, because no A/D converter or analog input circuitry has that much dynamic range. One well-known recorder that I measured some years ago gave only <2 dB improvement when going from 16 to 24 under the particular conditions I was measuring it. It wasn't a typical scenario, but it does illustrate the real dependency that exists on all the other factors in your actual scenario. If someone had used that recorder with those signal levels and gain settings except that they thought, "With 24 bits I can reduce my levels by (say) 12 dB to be absolutely safe from overload", their 24-bit recording would then have been ~10 dB noisier than the 16-bit one. Again that is based on an atypical scenario in terms of signal levels and gain settings, but not impossible by any means.

This makes me think about the 32-bit floating point recorders using multiple A/D converters along with proprietary schemes of switching dynamically between them.  Rather than folks thinking, "With 24 bits I can reduce my levels by (say) 12 dB to be absolutely safe from overload", with 32bit fp the ability to adjust recording level is essentially removed entirely. 

Seems to me these recorders incorporate two separate innovations which needn't necessarily be linked.. yet both are present in every product I've seen.  Ever since these recorders became available it's stuck me that the more fundamental innovation is not the use of the 32-bit floating point file format, but rather the scheme of switching between multiple A/D converters to provide a significantly greater "real world" dynamic range capability.  That same real world improvement should apply regardless if the recording format is 24bit fixed or 32bit fp. A true 24 bits worth of dynamic range already exceeds the specified input range of the recorder.  The Sound Devices MixPre II recorders with this feature claim a dynamic range of 142 dB, slightly less than what can be mathematically represented by a 24bit file format.  The Zoom F6 claims 131dB.

What am I missing here?  Why did manufacturers feel the need add the 32bit fp format when no 32bit fp recorder is capable of capturing a dynamic range greater than what can be handled by a 24bit recording format? 

More specifically, does this mean folks using these 32bit fp recorders may be able to record 24bit files which represent the same real-word dynamic range as a 32bit fp file made on the same recorder?  That assumes the multiple A/D scheme remains in use when recording to the 24bit format as well as the 32bit format, of course.

[aaronji]

Seems to me these recorders incorporate two separate innovations which needn't necessarily be linked.. yet both are present in every product I've seen. 

The original versions of the MixPre (such as the MixPre-6) have multiple ADCs and record a maximum of 24-bit. They even cite their patent in the specifications in the user manual. This is also the case for the higher-end Sound Devices recorders, such as the Scorpio, 888 and 883. This was discussed quite a bit previously; with those recorders, it is generally possible to set the gain such that the entire dynamic range of most microphones, which far exceeds the dynamic range of a performance, can be recorded.