[VocaloidDude]

So I know that I can hear about 17000khz. I can sort of "feel" some of the higher frequencies in my head when I go to the higher frequencies, but not really the actual tone of the sound. The highest the human ear can apparently hear is 20khz. This degrades as you get older. I'm not sure what the khps that you get as you go higher and higher on kbps, but I've done some blind tests between 128kbps (this one is audibly different), and variable bit rate and 320 kbps. Around this range it becomes difficult to detect a difference, if any at all. I don't believe that the human ear can actually hear the difference of a 16bit flac recording. I've heard it said again and again, I believe once from an audio myth buster video on youtube, people talking about how your ears actually cannot hear that much of a difference, because it's not humanly possible.
 
I know that there was recently a thread about the difference between 16bit and 24bit, but I am more concerned about what the thresh hold for sound quality is, in knowing exactly what bit rate the human ear can actually hear. 
 
I believe in the other thread someone said something like if you had the equipment to reproduce 16bit sound quality on your system it would actually make you deaf, because of something like the decibels increase as the bit rate increases, or something like that. I don't know.

 

[RRod]

  So I know that I can hear about 17000khz. I can sort of "feel" some of the higher frequencies in my head when I go to the higher frequencies, but not really the actual tone of the sound. The highest the human ear can apparently hear is 20khz. This degrades as you get older. I'm not sure what the khps that you get as you go higher and higher on kbps, but I've done some blind tests between 120kbps (this one is audibly different), and variable bit rate and 320 kbps. Around this range it becomes difficult to detect a difference, if any at all. I don't believe that the human ear can actually hear the difference of a 16bit flac recording. I've heard it said again and again, I believe once from an audio myth buster video on youtube, people talking about how your ears actually cannot hear that much of a difference, because it's not humanly possible.
 
I know that there was recently a thread about the difference between 16bit and 24bit, but I am more concerned about what the thresh hold for sound quality is, in knowing exactly what bit rate the human ear can actually hear. 
 
I believe in the other thread someone said something like if you had the equipment to reproduce 16bit sound quality on your system it would actually make you deaf, because of something like the decibels increase as the bit rate increases, or something like that. I don't know.

 
If you're talking in terms of information, the latest codecs tend to provide transparency below 256kbps. Depending on the material that can be much lower. I haven't been able to pick up any obvious problems at 96kbps Opus with my classical collection, with some exceptions needing 128k. Other genres with more electronic blips might need a bit more, as would listeners with keener ears than I.

 

[spruce music]

   
If you're talking in terms of information, the latest codecs tend to provide transparency below 256kbps. Depending on the material that can be much lower. I haven't been able to pick up any obvious problems at 96kbps Opus with my classical collection, with some exceptions needing 128k. Other genres with more electronic blips might need a bit more, as would listeners with keener ears than I.

Opus seems oriented for high quality at low bit rates.  Still it begins to merge with Vorbis, AAC and MP3 at around 128 kbps and higher according to some formal listening tests.  I believe AAC tests out slightly ahead of the others.  I would be surprised if 128 kbps were inaudible on music.  I haven't tried it. 
 
When I have tried Vorbis or MP3 320 kbps and usually 256 kbps is fine.  Lower and I can detect it.  Of course with age everyone's hearing worsens.  Losing not just high frequencies, but also the ability to hear details into complex sounds because the filtering done by our basilar membrane gets less sharp with age.
 
As for 16 and 24 bit, there might be some highly unusual situations where one might hear 16 vs 24 bit.  Theoretically one might need somewhere around 18-20 bits for complete transparency for those with good hearing and training of their hearing.   Those are uncommon edge cases.  Little and usually nothing is being lost with a well put together 16 bit medium. 

 

[VocaloidDude]

 
Opus seems oriented for high quality at low bit rates.  Still it begins to merge with Vorbis, AAC and MP3 at around 128 kbps and higher according to some formal listening tests.  I believe AAC tests out slightly ahead of the others.  I would be surprised if 128 kbps were inaudible on music.  I haven't tried it. 
 
When I have tried Vorbis or MP3 320 kbps and usually 256 kbps is fine.  Lower and I can detect it.  Of course with age everyone's hearing worsens.  Loosing not just high frequencies, but also the ability to hear details into complex sounds because the filtering done by our basilar membrane gets less sharp with age.
 
As for 16 and 24 bit, there might be some highly unusual situations where one might hear 16 vs 24 bit.  Theoretically one might need somewhere around 18-20 bits for complete transparency for those with good hearing and training of their hearing.   Those are uncommon edge cases.  Little and usually nothing is being lost with a well put together 16 bit medium. 

I dunno, go read that thread about the 16 vs 24 bit myth. I think it would blow what you're saying out of the water.

 

[spruce music]

  I dunno, go read that thread about the 16 vs 24 bit myth. I think it would blow what you're saying out of the water.

I have read and posted in that thread.  Read my post more carefully.  It doesn't contradict the mythical aspects of 24 bit for playback. 

 

[Computerpro3]

I don't believe that the human ear can actually hear the difference of a 16bit flac recording.

 
A lot of this depends on the recording quality and mastering of the final track.  I can a/b 8 bit vs. 16 bit with over 95% confidence of the Neil Young sample or any type of well recorded classical, but I can't do it with the Gangnam style clip in the blind test, or most other hip hop.

16 bit vs. 24 bit....I would not be confident betting on my ability to pass that test.

 

[RRod]

 
Opus seems oriented for high quality at low bit rates.  Still it begins to merge with Vorbis, AAC and MP3 at around 128 kbps and higher according to some formal listening tests.  I believe AAC tests out slightly ahead of the others.  I would be surprised if 128 kbps were inaudible on music.  I haven't tried it. 
 
When I have tried Vorbis or MP3 320 kbps and usually 256 kbps is fine.  Lower and I can detect it.  Of course with age everyone's hearing worsens.  Losing not just high frequencies, but also the ability to hear details into complex sounds because the filtering done by our basilar membrane gets less sharp with age.
 
As for 16 and 24 bit, there might be some highly unusual situations where one might hear 16 vs 24 bit.  Theoretically one might need somewhere around 18-20 bits for complete transparency for those with good hearing and training of their hearing.   Those are uncommon edge cases.  Little and usually nothing is being lost with a well put together 16 bit medium. 

 
Transparency at rates like 128 or 96 is quite content dependent. I've ABX a harpsichord track at 96k Opus, and other samples exist. But even among ABX cases, I find some that don't actually bother me, in the sense that outside of the ABX framework I might not actually hear the difference. In those cases where I have, 128 has done the job. Perhaps some really blipping stuff might require more, but at the end of the day I would rather have an immersive, personalized, virtualized sound experience using 96Opus than whatever lossless, 1024DSD remaster of DSoTM they try to pawn on us next.