Is high-resolution audio audible or not audible and a waste of data?

[markanini]

One could argue that 720p is good enough if you use a 50 inch screen for movies, but the question is why one should put up with this low resolution when there are 4K material for the same cost today ?

Bad argument due to false equivocation, because 16/44.1 can contain information at the limits of normal hearing ability. 720p may be well below the limits of normal vision ability at normal viewing distances.

 

[Tangband]

Bad argument due to false equivocation, because 16/44.1 can contain information at the limits of normal hearing ability. 720p may be well below the limits of normal vision ability at normal viewing distances.

If you read what I wrote, you can see that 16/44,1 is NOT enough to do transparent mastering work using peq, eq, compressor/limiter plugins.

 

[fpitas]

If you read what I wrote, you can see that 16/44,1 is NOT enough to do transparent mastering work using peq, eq, compressor/limiter plugins.

I don't think anyone is suggesting that professionals use 16/44 in their work. That day passed many years ago.

 

[markanini]

If you read what I wrote, you can see that 16/44,1 is NOT enough to do transparent mastering work using peq, eq, compressor/limiter plugins.

1. The discussion is about the content delivery format.
2. Badly designed plugins don't validate hires audio.

I don't think anyone is suggesting that professionals use 16/44 in their work. That day passed many years ago.

It depends, if the target format is 44.1Hz it can be an advantage work in 44.1Hz to skip the final sample rate conversion which introduces some signal distorsion: http://src.infinitewave.ca/help.html

 

[Tangband]

1. The discussion is about the content delivery format.
2. Badly designed plugins don't validate hires audio.

My opinion is that a streaming service for the customer should have 24 bit 48 kHz because there are no valid reason for using less. Its another question that good 16 bit 44,1 kHz DA conversion with good digital filtering at the customer end can be transparent.

 

[SSS]

Hearing ability is the key for each specific listener. In my youth I could hear almost 20 kHz and cymbals sounded sharp with a lot of highs. And I could hear the line frequency 15625 Hz of the vacuum electron tube based television of that time so I instantly knew the TV is on. Now, being old I can hear up 12 kHz depending on the sound pressure volume. And the nice sound of cymbals and drums are lost for me to some extent. But there is a difference between the high frequency limit of my ears and the perception of distortion which I became trained when demonstrating and selling hifi equipment. Thus I can still hear distortion and distorted decay of musical notes. With my STAX electrostatic headphones I tried to recognize the differences between audio formats. But it is hard to find really good produced recordings. My finding was (with my old ears) that I barely can judge whether 24-bit 96 kHz is really better than CD quality. So it was a 50/50 chance. But mp3 can be recognized as a lesser quality sound. But when using the 3xx kbps the difference is for me really small and depends also of the musical content.

 

[Tangband]

There are entire threads here about this. I think the conclusion is that you're wasting your time with anything more than 16 bits for playback. For use in a studio for signal manipulation, a lot more bits of resolution is common and is accepted as superior.

Very true

 

[ahofer]

My opinion is that a streaming service for the customer should have 24 bit 48 kHz because there are no valid reason for using less

The evidence that we can hear the extra 8 bits, apart from some very extreme circumstances, is almost non-existent. That would be the 'valid reason' - it's a waste of space and bandwidth.

But storage is pretty cheap, and it's good to have a margin of safety.

 

[markanini]

Hearing ability is the key for each specific listener. In my youth I could hear almost 20 kHz and cymbals sounded sharp with a lot of highs. And I could hear the line frequency 15625 Hz of the vacuum electron tube based television of that time so I instantly knew the TV is on. Now, being old I can hear up 12 kHz depending on the sound pressure volume. And the nice sound of cymbals and drums are lost for me to some extent. But there is a difference between the high frequency limit of my ears and the perception of distortion which I became trained when demonstrating and selling hifi equipment. Thus I can still hear distortion and distorted decay of musical notes. With my STAX electrostatic headphones I tried to recognize the differences between audio formats. But it is hard to find really good produced recordings. My finding was (with my old ears) that I barely can judge whether 24-bit 96 kHz is really better than CD quality. So it was a 50/50 chance. But mp3 can be recognized as a lesser quality sound. But when using the 3xx kbps the difference is for me really small and depends also of the musical content.

If it's only about hearing ability MP3 128kbps would be good enough: https://listening-tests.hydrogenaud.io/sebastian/mp3-128-1/results.htm
That's why I prefer a standard to based on physical limits such as 44.1kHz PCM.

 

[Blumlein 88]

The evidence that we can hear the extra 8 bits, apart from some very extreme circumstances, is almost non-existent. That would be the 'valid reason' - it's a waste of space and bandwidth.

But storage is pretty cheap, and it's good to have a margin of safety.

Another reason for 24 bits even if verging on overkill (hey what kind of audiophile doesn't like a little overkill), is when you wish to do some RoomEQ for playback. The extra bits might help some when doing processing upon playback.

 

[markanini]

What grinds my gears is color standards for video content. You either have a rec709/sRGB with a limited gamut, or a convoluted proprietary mess that's Dolby Vision HDR. I would be happier with a new standard that simply defined new primaries to contain a wider color gamut, no proprietary protocols and extra bells and whistles that I didn't ask for. I don't desire to be blinded by 1000nits, but I would appreciated some saturated colors mixed in to the presentation.

 

[Blumlein 88]

What grinds my gears is color standards for video content. You either have a rec709/sRGB with a limited gamut, or a convoluted proprietary mess that's Dolby Vision HDR. I would be happier with a new standard that simply defined new primaries to contain a wider color gamut, no proprietary protocols and extra bells and whistles that I didn't ask for. I don't desire to be blinded by 1000nits, but I would appreciated some saturated colors mixed in to the presentation.

I guess DCI P3 is sort of in that direction. Rec 2020 is a bigger step toward a larger color gamut. I also wonder about the 1000 nits however. I've a laptop with 600 nits, and I pretty much always turn it down to about 60% even in a well lit room. That may just be because too many web pages are the video version of loudness wars. I am unclear what the real step forward is on Rec. 2100.

 

[ZolaIII]

@markanini you have HLG OTEF REC 709 conversation to BT. 2020 (& BT. 2100) as part of HDR10+ and that's the pretty much most useful (broadcast) one.

 

[mhardy6647]

Honestly if you're worried about that kind of resolution, your speakers are probably the weak link by orders of magnitude.

Nawwwww... it's his cables.



EDIT: In fact, proper performance can only be achieved if the length of each interconnect is exactly an integer multiple of every wavelength which is to be reproduced. This is rather difficult in Einsteinian space-time. I think Nordost has done it -- but you have to buy their best cables, and it's gonna cost you...

 

[Blumlein 88]

Nawwwww... it's his cables.



EDIT: In fact, proper performance can only be achieved if the length of each interconnect is exactly an integer multiple of every wavelength which is to be reproduced. This is rather difficult in Einsteinian space-time. I think Nordost has done it -- but you have to buy their best cables, and it's gonna cost you...

Oh no, MIT (the cable company not university) accomplished this long ago with their tuned strands that optimizes segments of the frequency spectrum.

 

[mhardy6647]

Oh no, MIT (the cable company not university) accomplished this long ago with their tuned strands that optimizes segments of the frequency spectrum.

I knew it was one of them! Never bet against MIT!

 

[krabapple]

I can't find the main articles on the fly. However you can search the web for "Interaural Time Difference" (or Level). As already reported, you can also find a lot in good psychoacoustic books. A very interesting one that I suggest is "Principle of Cognitive Neuroscience", by D. Purves and others.

[EDIT] About the test signal. As normally happens, to study the effects of any operation on the signal, tests are carried out with "canonical" signals. The step is one of them. Yes, it's not a realistic signal, but it shows well the time spreading effect of the transients, which is a physical fact. I think that on real signals surely such steep transients will occur to a lesser extent.

For discrete phases. It doesn't seem to me that anyone here has talked about discrete phases.

Since you seem comfortable with psychoacoustics concepts, you know that when you use a 'not realistic' signal you have to contend with the possibility that the effect might be masked to the point of irrelevancy when a realistic signal is used.

 

[Pinox67]

I just don't see how you can equate sensitivity to ITD with sensitivity to rise time.

Good question. ITD and ILD are based on an operation analogous to a cross-correlation between the signals that reach the brain from our two ears. Now, any phase shift between the channels of the ITD value, let's say about 5usec which corresponds to 1 sample for a 192KHz sampling, can be "detected" by our brain for both signals in the figure in my first post. But while the top one, at 192KHz, all the information to recognize the transition is available within 5usec, in the bottom one at 48KHz the same information takes 20usec to appear. Thus, the localization mechanism works on different information, it "strains" more to recognize the same event.

 

[krabapple]

Depends on the contents of the song. hihats are trivial at 128

Detection of lossy audio also depends on the encoder. They are not all created equal.

One could argue that 720p is good enough if you use a 50 inch screen for movies, but the question is why one should put up with this low resolution when there are 4K material for the same cost today ?

Because not everyone sits close enough to their TVs that it would matter?

 

[Pinox67]

Since you seem comfortable with psychoacoustics concepts, you know that when you use a 'not realistic' signal you have to contend with the possibility that the effect might be masked to the point of irrelevancy when a realistic signal is used.

You break through an open door with me...
In my studies I have always tried to run tests with "music-like" signals (you can find an example in this link). But to highlight phenomena, it is always necessary to start from simple signals.