Jump to content
IGNORED

Is recording at high bit depth objectively better?


Recommended Posts

I've been recording some stuff using Audacity, it sounds better at 24 and 32bit than 16bit (sample rate 44.1 in all cases). Now I'm a believer that Redbook should be enough, however I'm not sure modern ADC's are capable of capturing accurately at 16bit. Some Objectivist members like @Archimago record at 24bit so the question is, is 16bit good enough? (forget studio requirement for extra processing bandwidth) 

Link to post
Share on other sites
50 minutes ago, Rexp said:

I've been recording some stuff using Audacity, it sounds better at 24 and 32bit than 16bit (sample rate 44.1 in all cases). Now I'm a believer that Redbook should be enough, however I'm not sure modern ADC's are capable of capturing accurately at 16bit. Some Objectivist members like @Archimago record at 24bit so the question is, is 16bit good enough? (forget studio requirement for extra processing bandwidth) 

Not the Objective proof you  asked for, but how can you be a believer that 16/44.1 should be enough given your recent 24 and 32 bit 44.1 recordings created  using Audacity sounding better ?

IF you are recording from Vinyl, you will find that most these days record their Vinyl sourced albums at 24/96, with a trend now to 24/192 as can be seen on UseNet.

 Note also, that there are now also quite a few 24/44.1 recordings available from HDTracks etc. that you can compare with the RBCD.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to post
Share on other sites
7 minutes ago, sandyk said:

Not the Objective proof you  asked for, but how can you be a believer that 16/44.1 should be enough given your recent 24 and 32 bit 44.1 recordings created  using Audacity sounding better ?

I do have some good 16/44.1 recordings but most are bad so was wondering if the problem lies with ADC's used by most studios.

Link to post
Share on other sites
1 hour ago, Rexp said:

I do have some good 16/44.1 recordings but most are bad so was wondering if the problem lies with ADC's used by most studios.

 Perhaps too much track mixing used ,and limitations of the earlier ADC's and earlier mixers, especially the early opamp based mixers, where in more recent times they could start out with 32bit to reduce losses ?

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to post
Share on other sites
59 minutes ago, sandyk said:

 Perhaps too much track mixing used ,and limitations of the earlier ADC's and earlier mixers, especially the early opamp based mixers, where in more recent times they could start out with 32bit to reduce losses ?

I don't know, I would like to know from Objectivists why they use 24/32bit for needle drops and are there any measurable differences versus 16bit.

Link to post
Share on other sites
7 hours ago, Rexp said:

however I'm not sure modern ADC's are capable of capturing accurately at 16bit.

 

In my view you should rephrase that to "I'm not sure that all early ADC's were capable of capturing accurately at 16bit"

"Science draws the wave, poetry fills it with water" Teixeira de Pascoaes

Link to post
Share on other sites

Satisfied that PCM digital was an improvement over analog tape, Denon engineers set out to develop their own VTR-based system.

Their goals were improved audio quality and multi-track recording capability, which then made the system viable in most commercial recording settings of the 1970s.

In 1972, Denon unveiled the DN-023R, an 8-channel system featuring 13-bit resolution and a sampling rate of 47.25kHz. The system used a Hitachi (then called Shiba Electronics) 4-head open-reel broadcast VTR as its storage format.

Anazawa noted: "We used the low-band mode of the VTR, for black and white (video). The reasons were stronger (performance) than color mode for tape drop-out and less cost."

Anazawa said, with the DN-023R, "we could edit music recordings and cut (LP) discs using advanced (preview) head" to control lathe-automation.

 

Denon deployed the DN-023R system immediately and used it to make commercial-release recordings throughout the 1970s.

The first LP made with this system was Nippon Columbia NCC-8501, Mozart: String Quartets K. 458 and K. 421 by the Smetana Quartet.

This album was recorded 24-26 April 1972, at Aoyama Tower, Tokyo, and released in October 1972.

Denon also released at least six other digital-recording LPs in October 1972, including classical, jazz and traditional Japanese music selections.

Denon also made the first commercial digital recording in Western Europe, at Notre Dam de Rose outside of Paris on 2-3 December 1974, Bach "Musical Offering" BWV 1079, by the Paillard Chamber Orchestra.

The LP was released in May 1975.

 

In The Dawn of Commercial Digital Recording by Thomas Fine (attached)

Fine_Dawn-of-Digital.pdf

"Science draws the wave, poetry fills it with water" Teixeira de Pascoaes

Link to post
Share on other sites

--

Krzysztof Maj

http://mkrzych.wordpress.com/

"Music is the highest form of art. It is also the most noble. It is human emotion, captured, crystallised, encased… and then passed on to others." - By Ken Ishiwata

Link to post
Share on other sites
6 hours ago, Rexp said:

I don't know, I would like to know from Objectivists why they use 24/32bit for needle drops and are there any measurable differences versus 16bit.

 

There are measurable differences, mostly at the edge or well below of what is likely audible. But, if you want to be sure you have captured every crackle and pop in full glory (🤷‍♂️) then 24/48 or 24/96 is what I'd normally use. 32 bits instead of 24 doesn't make a difference. None of the existing ADCs or DACs have enough resolution/linearity to recover even the full 24 bits, much less 32.

Link to post
Share on other sites
11 hours ago, pkane2001 said:

 

There are measurable differences, mostly at the edge or well below of what is likely audible. But, if you want to be sure you have captured every crackle and pop in full glory (🤷‍♂️) then 24/48 or 24/96 is what I'd normally use. 32 bits instead of 24 doesn't make a difference. None of the existing ADCs or DACs have enough resolution/linearity to recover even the full 24 bits, much less 32.

Thanks, with my recordings I don't find sample rates above 44.1 make a huge improvement but find 16bit has distortion that hurts my 'golden' ears in the same way most commercial Redbook stuff does. If I could measure this distortion, I could vet downloads prior to purchase. (I have wasted alot of money on unlistenable downloads) 

Link to post
Share on other sites
39 minutes ago, Rexp said:

Thanks, with my recordings I don't find sample rates above 44.1 make a huge improvement but find 16bit has distortion that hurts my 'golden' ears in the same way most commercial Redbook stuff does. If I could measure this distortion, I could vet downloads prior to purchase. (I have wasted alot of money on unlistenable downloads) 

 

Don't know, I have found plenty of amazing-sounding 44.1/16 recordings, and plenty not so good. 

 

Make sure that you have dither applied when reproducing 16 bits, this helps with masking quantization noise. The difference between 16 and 24 bit should be really minor, unless you (or your DAC) are also applying some sort of DSP processing to the signal. In that case, it's better to keep everything in 24 bits to leave some room for loss of precision during calculations.

 

Take a 16 bit recording that sounds bad and convert it to 24 bits, then try to see if you can still hear the same problems. Could be something specific to how your DAC handles 16 bit data.

Link to post
Share on other sites
7 minutes ago, pkane2001 said:

 

Don't know, I have found plenty of amazing-sounding 44.1/16 recordings, and plenty not so good. 

 

Make sure that you have dither applied when reproducing 16 bits, this helps with masking quantization noise. The difference between 16 and 24 bit should be really minor, unless you (or your DAC) are also applying some sort of DSP processing to the signal. In that case, it's better to keep everything in 24 bits to leave some room for loss of precision during calculations.

 

Take a 16 bit recording that sounds bad and convert it to 24 bits, then try to see if you can still hear the same problems. Could be something specific to how your DAC handles 16 bit data.

Is there less quantization noise at 24bit?

Link to post
Share on other sites
14 minutes ago, Rexp said:

Is there less quantization noise at 24bit?

 

Quantization noise is generated when converting between analog and digital or doing any kind of numerical calculations with precision down-conversions in a digital computer/DSP. Whatever precision the DAC is capturing, the quantization noise will be at the level of the least significant bit. For a 16 bit capture, it'll be down below -90dB. For 22 bit capture (about the best you could expect in a real DAC or ADC), it'll be down below -125dB. So, while you can certainly apply dither to a 24 bit file, it's really, really unlikely it'll improve anything audible.

Link to post
Share on other sites
7 hours ago, Rexp said:

Thanks, with my recordings I don't find sample rates above 44.1 make a huge improvement but find 16bit has distortion that hurts my 'golden' ears in the same way most commercial Redbook stuff does. If I could measure this distortion, I could vet downloads prior to purchase. (I have wasted alot of money on unlistenable downloads) 

 

Are you adding dither when recording at 16-bit?

"Science draws the wave, poetry fills it with water" Teixeira de Pascoaes

Link to post
Share on other sites
3 hours ago, adamdea said:

ADCs are not the limitation. The limitation is in the mics, ambient noise floor and your ears. But there is no point recording editing etc at 16 bits. The ADC will almost certainly be sigma delta so there is nothing native about the 16 bits or even 24: any form of PCM will be the result of a conversion, and you might as well make that 24 or 32 bits in the first instance.

16 bits is not enough for editing.  To take a trival example, you will want to have some headroom in the recording, which you will then reduce to make the peak at or near full scale: if you started out at 16 bits you would immediately lose valuable DR.

As for the final distribution format, 16 bits is fine.

I am not doing any editing, just recording analog tapes or single voice. 16/44.1 sounds bad, 24/44.1 sounds good. I previousy used a USB audio interface which made all my recordings sound bad so figured the ADC and bit depth were key. 

Link to post
Share on other sites
2 hours ago, Rexp said:

I am not doing any editing, just recording analog tapes or single voice. 16/44.1 sounds bad, 24/44.1 sounds good. I previousy used a USB audio interface which made all my recordings sound bad so figured the ADC and bit depth were key. 

The ADC itself should not be any different if you are using the same device. It is possible that it's making a mess of the conversion from the ADC itelf to 16/44 (possibly by not dithering?) but it seems odd. Have you tried converting one of your 24/44 files to 16/44 using the HQ settings in audacity or dbpoweramp using triangular pdf dither? if so do they sound very different? Either way I guess it doesn't really matter provided that your 24/44 recordings sound good to you.

You are not a sound quality measurement device

Link to post
Share on other sites

If recording source is LP or other repeatable source, it is possible to play the same sound twice : on the first playback, find maximum input level and adjust recording gain to, say -6 dBFS on max input and on the second time, play with recording. Recorded signal is most probably free from harsh integer overflow or digital clipping. On this case, 16bit recording is sufficient, 90dB dynamic range can be achieved.

 

But on live recording, maximum input level is unknown before recording, therefore more conservative recording gain is selected. Something like, set VU meter 0 VU to -18 dBFS or -20 dBFS and adjust recording gain to not exceed 0 VU. In this case, with 16bit, available SNR is reduced to 76 ~ 78 dB on the loudest part. SNR of second movement becomes worse and this can be problematic. With 24bit, after recording, it is possible to increase gain by 20dB to get 16bit bit depth / dynamic range sound.

Sunday programmer since 1985

Developer of PlayPcmWin

Link to post
Share on other sites
9 hours ago, adamdea said:

The ADC itself should not be any different if you are using the same device. It is possible that it's making a mess of the conversion from the ADC itelf to 16/44 (possibly by not dithering?) but it seems odd. Have you tried converting one of your 24/44 files to 16/44 using the HQ settings in audacity or dbpoweramp using triangular pdf dither? if so do they sound very different? Either way I guess it doesn't really matter provided that your 24/44 recordings sound good to you.

Yes I've tried down converting using dppoweramp/Audacity and could not get the 16bit to sound as good as 24bit. Hey ho.. 

Link to post
Share on other sites
  • 4 months later...
On 2/21/2021 at 10:53 PM, erin said:

The objective reply is that 16 bit recordings have 96dB dynamic range.

 

With dithering, they can achieve 100dB dynamic range.

 

Subjective comment: surely this is enough dynamic range to capture 99% of music recordings and definitely enough dynamic range to capture recordings from vinyl records.

 

Subjective comment:

In my experience, there is nothing wrong with 16 bit recordings. I have never heard any dynamic compression due to 16 bits being used.  Any differences in sound between 16 bit and 24 bit recordings are due to the way the DAC chip manages the audio conversion from D to A. Any sonic difference is caused by the hardware, not the actual 16 bit data.

 

The real value for higher bit depth or higher sample rates is about signal processing, not really just listening.

 

When you start with 16 bits and work only in 16bits, any gain less than 0dB will start affecting the dither.  Also, pure 16bit processing can create zipper noise with certain situations.   Likewise, higher sample rates are useful for dynamics processing -- that is, any fast gain control creates sidebands - it is best to  do processing at double the normal sample rate (4X the maximum signal frequency) because of modulation effects.   You can often get by with 3X the maximum signal frequency (I know, off topic, but trying explain the real reasons for higher rate/bit depth.)

 

Of course, with pure 16bit processing, there are dynamic range issues other than truncating the dither in extreme cases also.

 

This is why, when reading in a 16bit file, if there is any processing, then upconvert to 24bits/32bits or floating point first.   Convert back down when done.   Same goes for sample rate -- if you are doing dynamics processing, then do an immediate sample rate upconversion from 44.1k (or even 48k) to at least 66k for audio (88.2k/96k are better.)

 

Doing signal processing purely at the delivery rate/bit depth (44.1k/16 bits), without due consideration, can really cause troubles.   Delivering at the 44.1k/16bit  rate is probably okay though.  (I prefer 48k or higher even for delivery, but dont really have a strong argument for it.)

 

 

 

 

 

Link to post
Share on other sites
14 hours ago, John Dyson said:

The real value for higher bit depth or higher sample rates is about signal processing, not really just listening.

 

When you start with 16 bits and work only in 16bits, any gain less than 0dB will start affecting the dither.  Also, pure 16bit processing can create zipper noise with certain situations.   Likewise, higher sample rates are useful for dynamics processing -- that is, any fast gain control creates sidebands - it is best to  do processing at double the normal sample rate (4X the maximum signal frequency) because of modulation effects.   You can often get by with 3X the maximum signal frequency (I know, off topic, but trying explain the real reasons for higher rate/bit depth.)

 

Of course, with pure 16bit processing, there are dynamic range issues other than truncating the dither in extreme cases also.

 

This is why, when reading in a 16bit file, if there is any processing, then upconvert to 24bits/32bits or floating point first.   Convert back down when done.   Same goes for sample rate -- if you are doing dynamics processing, then do an immediate sample rate upconversion from 44.1k (or even 48k) to at least 66k for audio (88.2k/96k are better.)

 

Doing signal processing purely at the delivery rate/bit depth (44.1k/16 bits), without due consideration, can really cause troubles.   Delivering at the 44.1k/16bit  rate is probably okay though.  (I prefer 48k or higher even for delivery, but dont really have a strong argument for it.)

 

 

 

 

 

I find the majority of commercial Redbook stuff unlistenable and was thinking its either the ADC used by some studios or the DSP/mastering at fault. Guess it could be either but recording at higher bit rates doesn't seem essential. 

Link to post
Share on other sites
1 hour ago, Rexp said:

I find the majority of commercial Redbook stuff unlistenable and was thinking its either the ADC used by some studios or the DSP/mastering at fault. Guess it could be either but recording at higher bit rates doesn't seem essential. 

 

Which is not how it can work out in practice ... 🙂.

 

Use any and all of the numerous tweaking methods, that have been mentioned hundreds of times, to improve the integrity of the replay path - and all those "unlistenable" anomalies disappear. That's how one knows one has reached a certain standard of competence - all the weaknesses in the playback chain which cause the unlistenability have been sorted; and you have a free run from then on, 😉.

Frank

 

http://artofaudioconjuring.blogspot.com/

 

 

Over and out.

.

 

Link to post
Share on other sites
1 hour ago, Rexp said:

I find the majority of commercial Redbook stuff unlistenable and was thinking its either the ADC used by some studios or the DSP/mastering at fault. Guess it could be either but recording at higher bit rates doesn't seem essential. 

My new decoder (with my new headphones -- the old ones were giving me too much bass, therefore telling me to make the output thin.)


The decoder WILL demonstrate the cause of bad sound.   (I am again hearing what I had been happy with, but with working headphones.)

I don't know when the old headphones failed -- but must have been 6mos ago.

 

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...