• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

16-bit... It really is enough!

Pluto

Addicted to Fun and Learning
Forum Donor
Joined
Sep 2, 2018
Messages
990
Likes
1,631
Location
Harrow, UK
Oddly enough, bottles and cans are 500 ml
As I said, it is indeed odd that the public desire to retain the pint only applies to draft beer, as served in bars. There is something rather nice and reassuring about the pint; it's a good thirst-quenching quantity without the beer going warm and dull by the time you get to the bottom. And a pint glass fits the average hand quite well.

The 1L steins commonly found in southern Germany are rather too large (IMHO) as the beer has typically lost its “liveliness” by the time you've reached the bottom. And before anyone comments, even “flat” British bitter has a certain liveliness when in good condition, vital to the enjoyment of the perfect pint!
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
I've compared a whole lot of higher res music to 16/44 versions of the same masters. I've never been able to tell the difference. Probably if I did Amir's trick and compared fadeouts with the volume cranked all the way, up I'd hear the higher noise floor on some pristine recordings. But in my mind this does not count as a real difference.

So I look for red book quality when I can find it. I understand the sentiment of "if the information is there, then why not hang onto it?" But already my music collection is approaching a terabyte. At 24/96 it would be 3 terabytes. And that cascades through all my backups. I don't feel like paying rent for all those extra bytes.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
Yeah, 16/44 is somewhere between 95% and 100% of all the fidelity needed to equal human hearing audibility. It probably is more like 98% for anyone, and 100% for a majority of the people.

It's not just a question of who's listening, and the circumstances of the listening test. It's about the recordings themselves. How many recordings in your music collection have 96db dynamic range, or anything even approaching this? I'm guessing zero. How many are made from masters with a noise floor that's at 96db or lower? I'm guessing very few. How many are made from masters with significant musical or ambient acoustic information above 22khz? I'm guessing very few.

If you look at a spectral analysis of the ultrasonic frequencies in high res music, you see lots of stuff that looks like noise spikes—possibly from electronics problems, or HVAC systems, or squeaky tape rollers. Things that would really mess with your dog's enjoyment of the recording. They got onto the master because all the engineers involved are humans and so can't hear any of it.
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,348
Likes
3,462
Location
San Diego
I've compared a whole lot of higher res music to 16/44 versions of the same masters.

I think the easiest and best way to prove to your self whether or not you can hear a difference is to take a 24/192 "real" hi-res file and dither it down to 16/44 and then ABX them in Foobar2000 and see if you can tell a difference. I have tried this and I can not. There is one trick though... if you turn the volume up super high before the music starts playing or between tracks you can sometimes hear extra noise in the 16/44 but it has no audibility when music is playing.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
I think the easiest and best way to prove to your self whether or not you can hear a difference is to take a 24/192 "real" hi-res file and dither it down to 16/44 and then ABX them in Foobar2000 and see if you can tell a difference. I have tried this and I can not. There is one trick though... if you turn the volume up super high before the music starts playing or between tracks you can sometimes hear extra noise in the 16/44 but it has no audibility when music is playing.

I've done this using Logic Pro. I can only even hear the added noise (with the volume cranked) if it's an especially low-noise recording. Quantizing noise is always much quieter than any tape hiss, for example.
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
I would say that the 10 dB SINAD rule between devices from the audio chain should also apply to recordings too, example:
- Recording should have at least 130 dB of SINAD
- DAC >= 120 dB SINAD
- Amplifier >= 110 dB SINAD at speakers/headphones outputs
- Headset < 0.01 dB THD (at least at 1 kHz and a SPL > 90 dB)
- Speakers < 0.1 dB THD (at least at 1 kHz and a SPL > 90 dB)

Aiming the recording to a SINAD of 90 dB may not be the best idea if we really want to get 90 dB SINAD out of our speakers or headphones, especially in an era when DAC chips are having a SINAD better than 120 dB. So, from a performance factor, definitely 16 bits resolution is not enough, although our hearing may not perceive any audible differences between 16 vs. 24 bits recordings.

P.S.: Yes, mics are all analogic and their SINAD is probably worse than our headset, I know that. :)
 
OP
M

MusicNBeer

Senior Member
Forum Donor
Joined
Jan 20, 2019
Messages
333
Likes
478
@sal, as a signal processing engineer, your assertion of more bit depth aiding in frequency accuracy upon playback really makes me shake my head. Please don't post complete dis-information like that.

For me, the main thing that angers me is not the storage space, but the myth that more than 16 bits sounds better, completely irrespective of dynamic range. This has led to the whole hi-rez market, totally driven on a false narrative.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
It's not just a question of who's listening, and the circumstances of the listening test. It's about the recordings themselves. How many recordings in your music collection have 96db dynamic range, or anything even approaching this? I'm guessing zero. How many are made from masters with a noise floor that's at 96db or lower? I'm guessing very few. How many are made from masters with significant musical or ambient acoustic information above 22khz? I'm guessing very few.

If you look at a spectral analysis of the ultrasonic frequencies in high res music, you see lots of stuff that looks like noise spikes—possibly from electronics problems, or HVAC systems, or squeaky tape rollers. Things that would really mess with your dog's enjoyment of the recording. They got onto the master because all the engineers involved are humans and so can't hear any of it.

Agree. There is far too much "........... 99,100, change hands" on audio forums.
hide.gif
 

Kegemusha

Senior Member
Joined
Jul 16, 2020
Messages
488
Likes
462
Indeed it is the quality of the recording that really matter.
have any of you make a double blind test between
16/44
24/48
24/96
...
?

I did with speakers and headphones and couldn’t find appreciable results.
for sure much less rewarding than add rug carpet to absorb floor reflections.
Really curious to know, I have 46 years old ears.
My Best

yes, and I can't hear any differnece between 16b or 24b, and even if you do, how much effort in listening you need to do, can you hear 20 to 20khz, because many people can not.

And at the end is the music more enjoyable?

I listen to CDs at home, I can spot if the source is spotify or CD, that I can.
I can also hear if the record is bad recorded,there are CDs that sound amazing and many that not. So all depends of the quality of the source and not so much of bit depth,just my 2 cents.
If you take any CD of The Fall for example (a band that I like a lot) and listen in 16 or 24, it will not mater :)
 

nimar

Active Member
Forum Donor
Joined
Jan 25, 2021
Messages
213
Likes
216
Location
Ontario, Canada
This should be _the_ argument against MQA, the hopeful (perhaps Canadian) side of me wanted to give the benefit of the doubt to the MQA folk as I couldn't fathom that so much effort would go into something that would provide absolutely zero value. Not to mention that I had to live with Tidal as the only "hi-res" streaming service, so wanted to get the most of what was available.

But if there's absolutely no reason to have above 16/44 or 16/48 then there's even less reason for MQA to exist.
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,348
Likes
3,462
Location
San Diego
I couldn't fathom that so much effort would go into something that would provide absolutely zero value.

There is value for the record companies and hardware manufacturers but not for consumers. Don't want to derail this thread with MQA but couldn't resist :)
 

nimar

Active Member
Forum Donor
Joined
Jan 25, 2021
Messages
213
Likes
216
Location
Ontario, Canada
There is value for the record companies and hardware manufacturers but not for consumers. Don't want to derail this thread with MQA but couldn't resist :)

Don't want to go down an MQA rabbit hole either, but it is very much in line with the original intention / title. 16bit really is enough, we don't need DSD or MQA or SACD or HDCD or any other hi-res nonsense because it adds nothing to the listening experience.

Only wish I had come to this realisation a few months back and just bought an RME DAC instead of the"better measuring" Topping stack. For better or worse a lot of what is on ASR promotes this more is better mentality and I would rather have more features, rather than higher numbers.
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
So your premiss is faulty and everything else is non-sequitur ;)
We are testing here on ASR audio equipment having a SINAD of 120 dB and I see no reason to promote an obsolete 16-bits audio recording technology that might provide audio transparency only under perfect conditions (near 0dB recording peaks, a CD player with a SINAD higher than 90 dB etc.).

I personally see no reasons for feeding a 120 dB SINAD DAC with a 16/44 track if the same recording can be found at 24-bit resolution (although I am against 24/192, but I am pro 24/48).

Don't get me wrong, I'm not against 16/44, but I will definitely not promote it, because I see no reasons to continue using an audio format that is technologically inferior to our current DACs and amps.
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
We are testing here on ASR audio equipment having a SINAD of 120 dB and I see no reason to promote an obsolete 16-bits audio recording technology that might provide audio transparency only under perfect conditions (near 0dB recording peaks, a CD player with a SINAD higher than 90 dB etc.).

I personally see no reasons for feeding a 120 dB SINAD DAC with a 16/44 track if the same recording can be found at 24-bit resolution (although I am against 24/192, but I am pro 24/48).

Don't get me wrong, I'm not against 16/44, but I will definitely not promote it, because I see no reasons to continue using an audio format that is technologically inferior to our current DACs and amps.
I am an engineer and I learned decades ago making something better than needed is stupid.
It always has been and the only excuse for DACs which have a massively greater dynamic range than any music is that it doesn't cost any more.
The age of a format is irrelevant if it does the job 100% faultlessly, which is the case with 16/44.1 - the first to do so.

Quoting the distance to the moon in millimeterss rather than kilometers does NOT give any more useful informtion.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
We are testing here on ASR audio equipment having a SINAD of 120 dB and I see no reason to promote an obsolete 16-bits audio recording technology that might provide audio transparency only under perfect conditions (near 0dB recording peaks, a CD player with a SINAD higher than 90 dB etc.).

Only under perfect conditions? Has anyone been able to tell a difference under ANY actual listening conditions in an ABX test? I know Amir has been able to tell the difference consistently, but as he describes it he didn't do it under real listening conditions. He looped the tail end of a fade-out, and listened with the volume turned up so high that a 0db signal would have blown everything to bits.
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
Don't get me wrong, I'm not against 16/44, but I will definitely not promote it, because I see no reasons to continue using an audio format that is technologically inferior to our current DACs and amps.

This argument completely ignores the limited dynamic range of the content and the limited senses of audience - 16/44 is good enough.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
I can understand someone saying that on general principle they want all the bits. But it makes no sense to consider this an imperative, when it can be shown that they'll never hear the difference. it's also important to ackowledge the costs. Storage has gotten cheaper, but not free. And music collections have a way of growing and growing. And don't forget the cost of backup space! I have a main hard drive, plus an internal Apple time machine drive, plus a couple of external backups, plus rented cloud backup space. Doubling the storage requirements of the main drive means doubling all the others.

To me it's about buying audio differences I won't ever hear for storage costs I will absolutely feel.
 
Top Bottom