Head-Fi.org › Forums › Equipment Forums › Sound Science › Why 24 bit audio and anything over 48k is not only worthless, but bad for music.
New Posts  All Forums:Forum Nav:

Why 24 bit audio and anything over 48k is not only worthless, but bad for music. - Page 5

post #61 of 334

Rewording my post since it was misunderstood and deleted.

 

personally believe that 'high definition audio' and all those who sell it claiming it is audibly different is, like many things in audio, simply a scam. I would loved to be proved wrong with some actual evidence, but at the moment all the evidence supports this view. I apologise if this upsets anyone, but this board is sound science and I know I'm not alone in this belief.

post #62 of 334

I'm wondering if the differences exist in my DAC.  Is it possible that a DAC  designed for 44.1 playback will sound just as good with a higher format sampling, but the reverse is not true?  Maybe a good processor designed for 44.1 will sound exactly like a converter built for higher formats?  Why not more discussion regarding the build and development of DAC processors?  Am I in the wrong forum?  I'm not an engineer, just wondering out loud.

post #63 of 334
Quote:
Originally Posted by fiascogarcia View Post
 

I'm wondering if the differences exist in my DAC.  Is it possible that a DAC  designed for 44.1 playback will sound just as good with a higher format sampling, but the reverse is not true?  Maybe a good processor designed for 44.1 will sound exactly like a converter built for higher formats?  Why not more discussion regarding the build and development of DAC processors?  Am I in the wrong forum?  I'm not an engineer, just wondering out loud.

I'm not a DAC engineer either but I think this is a good route for the discussion to take.

post #64 of 334
Quote:
Originally Posted by kraken2109 View Post
 

Rewording my post since it was misunderstood and deleted.

 

personally believe that 'high definition audio' and all those who sell it claiming it is audibly different is, like many things in audio, simply a scam. I would loved to be proved wrong with some actual evidence, but at the moment all the evidence supports this view. I apologise if this upsets anyone, but this board is sound science and I know I'm not alone in this belief.

 

I personally believe (If I might borrow your formula?) that there is no such thing as "high definition," or "High Resolution" audio.

 

These are not technical terms. There is sample rate, and there is bit depth. The rest is invented by a collusion of marketing departments and tongues-out "audiophiles" that luuuurve those words.

 

More: High-bit-rate is a spin too. It is intended to make people think of their 44/48 as "low." It is not low. Low-bit-rate might have been there in the telephone industry, decades ago. Higher might be accurate, even 44.2 is higher thatn 44.1, but, hey, it won't achieve the double marketing wammy of being saleable and making people discontent with what they have.

 

Does that mean that 16-bit, 44.1khz, is the bee's-knees-ultimate in music reproduction and should be set in stone as perfect for ever and cancel any other research? No, absolutely not. One day, PCM and DSD might seem as old-fashioned as a shellac 78rpm record does today. Who knows what will be discovered and developed in the future!

 

Further more, when I listen to audio scientists (some of whom love music too) they tell that, no, 44.1/16 is not ultimate. Sure it covers the entire audio frequency range, and just about all the entire dynamic range from fff to ppp, but there is more to it than that. There is the implementation of the filtering and reconstruction circuits. 48 is better, and some say 60 would have been ideal. 60, of course, is one standard that no-one gives us! 

 

I'd like to be able to say, my understanding is and mean it... but I'm a maths dunce, formulae make my eyes spin, and this is deeply mathematical stuff, so don't ask me to explain or justify this. The best I could do is point to the contributions of some people on other forums.

 

Anyway, in the loose meaning of the phrase, my understanding is... that 48.1 is just fine but not perfect, 48 is better, 60 would be good (but we are not offered it anyway), but the reasoning is nothing to do with the usual misunderstanding of sample rates, and just climbing the ladder of bigger numbers is senseless, except to the sales departments.

post #65 of 334

 

 

Quote:

Originally Posted by Thad-E-Ginathom View Post
 

 

I personally believe (If I might borrow your formula?) that there is no such thing as "high definition," or "High Resolution" audio.

 

These are not technical terms. There is sample rate, and there is bit depth. The rest is invented by a collusion of marketing departments and tongues-out "audiophiles" that luuuurve those words.

 

More: High-bit-rate is a spin too. It is intended to make people think of their 44/48 as "low." It is not low. Low-bit-rate might have been there in the telephone industry, decades ago. Higher might be accurate, even 44.2 is higher that 44.1, but, hey, it won't achieve the double marketing wammy of being saleable and making people discontent with what they have.

Yes, I agree. I dislike marketing terms but using them was the easiest way of making my point.

 

 

Quote:

Originally Posted by Thad-E-Ginathom View Post

 

 

Does that mean that 16-bit, 44.1khz, is the bee's-knees-ultimate in music reproduction and should be set in stone as perfect for ever and cancel any other research? No, absolutely not. One day, PCM and DSD might seem as old-fashioned as a shellac 78rpm record does today. Who knows what will be discovered and developed in the future!

 

Further more, when I listen to audio scientists (some of whom love music too) they tell that, no, 44.1/16 is not ultimate. Sure it covers the entire audio frequency range, and just about all the entire dynamic range from fff to ppp, but there is more to it than that. There is the implementation of the filtering and reconstruction circuits. 48 is better, and some say 60 would have been ideal. 60, of course, is one standard that no-one gives us! 

 

I'd like to be able to say, my understanding is and mean it... but I'm a maths dunce, formulae make my eyes spin, and this is deeply mathematical stuff, so don't ask me to explain or justify this. The best I could do is point to the contributions of some people on other forums.

 

Anyway, in the loose meaning of the phrase, my understanding is... that 48.1 is just fine but not perfect, 48 is better, 60 would be good (but we are not offered it anyway), but the reasoning is nothing to do with the usual misunderstanding of sample rates, and just climbing the ladder of bigger numbers is senseless, except to the sales departments.

 

 

I'd like some sources for this because I haven't heard any good arguments for 44.1/16 not being perfect.
(This is not me calling you wrong, a liar, or insulting you in any way. This is me attempting to learn more. I felt the need to put this because tone is often hard to read online).
post #66 of 334
Redbook was designed to be perfect. It should be perfect sound. If it isn't, something is wrong somewhere.

There are areas for improvement, but they are outside of the realm of 2 channel audio, and some of them don't involve fidelity, but rather euphonics.
Edited by bigshot - 5/3/14 at 10:32am
post #67 of 334

There are good technical reasons why 16/48 or even 16/60 or so would be better in some ways, but they have nothing to do with the ability of the format to capture the audible spectrum. Rather, the slightly higher sampling rate allows for the use of an anti-aliasing filter that has a much shallower rolloff without cutting off anything in the audible frequency range, which is easier to design. With modern electronics, a filter that is audibly perfect in the passband up to 20kHz but is 100+dB down at the nyquist frequency (22.05kHz for 16/44.1) is doable, and that eliminates most of the reason behind wanting the slightly higher sample rate.

post #68 of 334
Quote:
Originally Posted by kraken2109 View Post

 
I'd like some sources for this because I haven't heard any good arguments for 44.1/16 not being perfect.
(This is not me calling you wrong, a liar, or insulting you in any way. This is me attempting to learn more. I felt the need to put this because tone is often hard to read online).

 

Probably, I have to plead guilty to hearsay on that one --- and perhaps even misinterpretation. Any correction will be a welcome update to my understanding. I suppose I am thinking that if 48, 60, mean easier and ?better? filtering, then, by assumption that that produces a ?better? result, and if there is such a thing as a better result, 44.1 can't have been perfect in the first place. 

 

I have been trying to learn, with the linked article in the opening post being something of a wake up call, and the videos explaining stuff (like the stair-step misconception) that I had never even begun to understand before.

 

My loosely-called understanding is definitely a work in progress, despite a certain personal tendency to talk as if it is set.

 

The "marketing" stuff, by the way, was certainly not aimed at you. You just gave me the conversation jumping-in point, with the nicely-put "I believe..." These high-def/res labels are all over the place now: we'll all start talking like that soon. Until then, it's a regular rant of mine.

post #69 of 334
Quote:
Originally Posted by Thad-E-Ginathom View Post
 

 

Probably, I have to plead guilty to hearsay on that one --- and perhaps even misinterpretation. Any correction will be a welcome update to my understanding. I suppose I am thinking that if 48, 60, mean easier and ?better? filtering, then, by assumption that that produces a ?better? result, and if there is such a thing as a better result, 44.1 can't have been perfect in the first place. 

 

I have been trying to learn, with the linked article in the opening post being something of a wake up call, and the videos explaining stuff (like the stair-step misconception) that I had never even begun to understand before.

 

My loosely-called understanding is definitely a work in progress, despite a certain personal tendency to talk as if it is set.

 

The "marketing" stuff, by the way, was certainly not aimed at you. You just gave me the conversation jumping-in point, with the nicely-put "I believe..." These high-def/res labels are all over the place now: we'll all start talking like that soon. Until then, it's a regular rant of mine.


My understanding of sampling is that 44.1kHz is more than enough. In terms of filtering (e.g. anti-aliasing filtering) I've read that pretty much every modern DAC will oversample to get steeper filters, so there should be no need for a higher sampling rate.

post #70 of 334

Yeah. It's not accurate to say that higher sampling rates are necessary to sound quality because they aren't.

post #71 of 334

There's a post by j_j on Gearslutz, about sample rates and the filtering slopes, in their going-on-for-ever pono/sampling-rates thread. The trouble is, finding it amongst the thousands of others.:o ...


Edited by Thad-E-Ginathom - 5/3/14 at 3:39pm
post #72 of 334

tl;dr version: DBT the benefits of 24 bit 192 kHz over CD quality, or stop fooling yourself.

 

Quote:
Originally Posted by ab initio View Post
 

As for the vinyl bit... yeah, folks like to record their vinyl records at absurd bitdepths and sampling rates because they like to support the harddisk industry.

 

 

:beerchug: lol

post #73 of 334
I think that some of the talk of hi-res being better is due to poor marketing / implementation of the original redbook standard...

The first CD players were based on the Philips TDA1540 DAC which was only 14 bit (still technically good enough, but - we're in it for the numbers! wink.gif), which, it was argued was why CD sounded so bright in the early days (ignoring brick wall filters, people didn't want to know or understand about them!)...

Anyhow, I'd rather have a well implemented 16 bit setup than a poorly implemented 24 bit one...
post #74 of 334
Quote:
Originally Posted by Thad-E-Ginathom View Post

There's a post by j_j on Gearslutz, about sample rates and the filtering slopes, in their going-on-for-ever pono/sampling-rates thread. The trouble is, finding it amongst the thousands of others.redface.gif ...

When you say j_j, do you mean jj as in James Johnston formerly of Bell Labs?

EDIT: Never mind, found him. It is ol' jj.

se
Edited by Steve Eddy - 5/3/14 at 6:30pm
post #75 of 334

Andreas Koch who was involved in the creation of the standards for SACD states 

 

that studies have shown humans can perceive frequencies up to 100k, however at a much lower amplitude

 

(it may have been dynamic range as well)

 

http://www.youtube.com/watch?feature=player_embedded&v=nj7d7Jnx0xc

 

it's about 24 minutes in.   The entire panel  on the video consists of luminary electrical engineers,

 

recording engineers, mastering people, etc.

 

If anyone walked up to one of the people in this RMAF panel that designed or use these high res professional tools

 

 to record or transfer master tapes every day,  and stated  

 

" 24 bit audio and anything over 48k is not only worthless, but bad for music."

 

I can only imagine what their reaction would be.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Why 24 bit audio and anything over 48k is not only worthless, but bad for music.