[castleofargh]

I've got to mention something here.........
I simply don't understand why some people seem so resentful about this entire subject.

I've never driven my car over 90 mph..... yet I still see no reason why it's "bad" that my car "can" go that fast.
In general, in almost every other subject, most people agree that you're better off if your tools can actually deliver, not just adequate performance, but performance that's BETTER than necessary.
It goes by names like "safety margin" and "margin for error" and "headroom" and even "clearance".
And who would really buy a car that can only go 56 mph?

So, why, even if you believe that we can't hear above 20 kHz, is it so awful to allow some safety margin.
If I were recording bats, and found out that their cries extended to 46 kHz, I would buy a microphone whose response extended to 60 kHz; I wouldn't buy one that went up to 46.1 kHz.
So, why, even if humans can only hear to 20 kHz, doesn't it make equal sense to make recordings that extend "well above" 20 kHz..... just in case.... to leave a little safety margin?
Why would anyone specifically choose to use a sample rate that's "just barely good enough"?

There is a reason why the 44.1k sample rate was chosen for CDs.....
The reason is that, with the constraints of the technology at the time CDs were invented, the time/space constraint on CDs was considered to be important.
They couldn't have used the next-higher standard sample rate without reducing the storage time on a standard CD below one hour - which had been established as a target requirement.
Using the 44.1k sample rate, they were able to fit over an hour on a disk, and still deliver frequency response that was a tiny bit above the bare minimum necessary.
The 48k sample rate was already in use on DAT tapes, and was considered to be a sort of standard; they would have used that except that, if they had, they couldn't have fit an hour on a CD.
In fact, most movie audio (on DVDs) is still standardized at 48k.... and not 44.1k.

However, when you're talking about download FILES, that constraint simply doesn't exist.
(in fact, even originally, it was strictly tied to "fitting a complete album on a CD")

to be clear, almost nobody cares that a file is at 44.1 or 48khz, take price increase and storage cost/limits out of the equation and I'd be fine with DXD on my computer. why not? if it has no negative impact why should I care which format it is? you won't find many people going "oh no I have better resolution without any negative impact, what will I do?" ^_^
your argument that I understand, is that there is no point nowadays in risking it all on the lower format. it's simple an intuitive enough. also it's an opinion so it doesn't have to be right or wrong and we're all free to have our own.
but then I read stuff like this:

What's the big deal about paying an extra $5 for a 96k file instead of a 44k file?

and I angrily raise my fist in front of my screen with about the same impact on you that high-res tends to have on my ears. I purchase music for the sound I perceive, not to look at a spectrum in owe. so when I fail to perceive a difference(consistently over several years), the 96k file is worth to me exactly the same, not 5$ more. on the other hand if every 3 albums the 44 or 48k version saves me enough to buy a 4th one. now that's audible evidence right there, and it will pass all blind tests.

 

[bigshot]

What's the big deal about paying an extra $5 for a 96k file instead of a 44k file?

If the mastering is better it's worth it. If the mastering is the same it isn't. If the mastering is worse than the CD, it really isn't. I've found examples of all of these things. Buying a 24/96 track is no guarantee that it sounds better or even as good as the CD. It would be a lot more productive to do listening tests comparing different releases of specific albums to determine which one has the best mastering, than it would to do listening tests to see if we can hear inaudible sound.

This is the core problem with audiophiles. They focus on some tiny detail and ignore the elephant in the corner. They go down the slippery slope of justifying incremental improvements even when it gets to the point of absurdity. When I buy audio equipment, which isn't often any more because my system is tricked out now, I focus on getting something that does the job cleanly, efficiently and inexpensively. Once I get there, I don't worry about specs beyond my ability to hear. Every $5 I save from not having to buy high bit depths that are basically just the same thing stuffed with packing peanuts is $5 I can spend on more great music.

Spend money where it counts. If you have OCD and want to worry about things you can't hear, buying HD tracks won't solve your problem. It will only take you one more step down the rabbit hole.

And yes, people do worry about that top octave of human hearing more than any other. I see it here all the time with people who try to justify the octave beyond our ability to hear. Listen to what 17kHz sounds like and then tell me that is important in your music. Not bloody likely.

96kHz is one octave more than 44.1... seven measly notes that I can't even hear. Why should I pay $5 more for that? I can't figure out why I would possibly pay almost a dollar for each note I can't hear.

 

[Don Hills]

... There is a reason why the 44.1k sample rate was chosen for CDs.....

And it's not the reason that you said.
Kees Immink and others have made it quite clear that disc playing time was not the reason for the sample rate choice. It was dictated by the requirement to be able to store the digital data on videotape.

 

[71 dB]

I think the requirement of "continuous time signals" in Nyquist sampling theory is misunderstood by advocates of high-res / analog audio.

It doesn't mean something like sine waves that started in the distant past and will continue forever. It means continuity in mathematical respect, that a function doesn't "jump" from one value to another in no time. That doesn't happen in real world. Instruments can produce only "continuous time signals". Acoustic waves can be only "continuous time signals". So, the only thing to worry about in digital audio is bandlimiting. If we do that correctly, everything will be fine. Worrying about impulses (mathematical constructions which can't exist in reality) is absurd.

 

[KeithEmo]

I've seen both claims made at different times..... and the technical arguments do seem to favor the fact that 44.1k "fits well" with the data structure on videotape.
However, neither relates to optimizing the audio performance......
And, for that matter, neither relates to the practical limits of audio circuit design.

Before oversampling was invented, the reconstruction filter used on a CD player or DAC had to be flat to 20 kHz.
However, because of the 44.1k sample rate chosen, it also had to have significant attenuation at and above 22 kHz.
These two requirements, taken together, make the required reconstruction filters difficult to design, difficult to manufacture at reasonable cost, and prone to unpleasant side effects.
Selecting a much higher sample rate would have made is simpler to design higher-performance reconstruction filters and cheaper to build them.
It was a bad choice, in terms of design flexibility... even if you don't consider having a little bit of safety margin in terms of frequency response worthwhile.
Even if you really only consider response up to 20 Khz to be important..... a higher sample rate will still make it easier to achieve it with fewer filter design issues and side effects.
(Modern oversampling was basically invented as a solution to this problem.... and would have been unnecessary if a high enough sample rate had been chosen to begin with.)

And it's not the reason that you said.
Kees Immink and others have made it quite clear that disc playing time was not the reason for the sample rate choice. It was dictated by the requirement to be able to store the digital data on videotape.

 

[Don Hills]

I've seen both claims made at different times..... ( ) However, neither relates to optimizing the audio performance...... And, for that matter, neither relates to the practical limits of audio circuit design. ...

Anyone who wants to find more than they ever wanted to know about the history and technical details of CD, DVD and Blu-Ray design and development can visit http://www.turing-machines.com/indeximmink.html
Start with papers 50 and 105. The origin of the "Beethoven's 9th" legend is covered.

Regarding sample rate and bit depth, it's pretty clear that they were pushing the limits of what was available and affordable at the time to develop a consumer oriented playback system. Developing a new low cost recording system to enable a higher sample rate was a bridge too far. So, 44.1 KHz. You can wish all you want, but that's the way it was.

... Modern oversampling was basically invented as a solution to this problem. ...

The first Philips CD player used 2x oversampling.

 

[bigshot]

Lots of stuff there. Thanks!

 

[JaeYoon]

Anyone who wants to find more than they ever wanted to know about the history and technical details of CD, DVD and Blu-Ray design and development can visit http://www.turing-machines.com/indeximmink.html
Start with papers 50 and 105. The origin of the "Beethoven's 9th" legend is covered.

Regarding sample rate and bit depth, it's pretty clear that they were pushing the limits of what was available and affordable at the time to develop a consumer oriented playback system. Developing a new low cost recording system to enable a higher sample rate was a bridge too far. So, 44.1 KHz. You can wish all you want, but that's the way it was.



The first Philips CD player used 2x oversampling.

Someone is on fire! Those papers are very good reads!

 

[71 dB]

It was a bad choice, in terms of design flexibility... even if you don't consider having a little bit of safety margin in terms of frequency response worthwhile.

It was a good compromise in it's day some 40 years ago and provided a huge improvement in audio transparency on consumer market for normal people.

Even if you really only consider response up to 20 Khz to be important..... a higher sample rate will still make it easier to achieve it with fewer filter design issues and side effects.
(Modern oversampling was basically invented as a solution to this problem.... and would have been unnecessary if a high enough sample rate had been chosen to begin with.)

Yes, but they had their reasons to choose 44.1 kHz. Fortunately it is able to provide transparent sonic experiment for human ears despite of filter difficulties. It is easy for us who live the age of 384 kHz DXD recordings to complain about low sample rate, but digital technology was different in the late 70's. Let's be happy about the fact they didn't choose 32 kHz!

The problems of 44.1 kHz are totally overblown out of proportion by hi-res business and other besserwissers who don't understand properly the scale of the issue. For me 16/44.1 is a transparent audio format. If it isn't for your bat ears then bad luck man!

 

[KeithEmo]

I don't disagree with you at all......

I have no complaints at all that 44.1k was chosen as the sample rate for CDs in the 1970's (and horses were a remarkably effective and efficient form of transportation in 1820).
However, as you point out, we now live in an age where there are lots of other alternatives, many of which are better.
The fact that CDs were an excellent compromise in the 1970's doesn't mean that we should stop trying to find something better.

One obvious example is oversampling.
It's impossible to design a practical, cheap, and effective reconstruction filter to operate directly with digital audio sampled at 44.1k.
Therefore, oversampling was invented as a way to sidestep this problem.
However, with modern technology, it would be much simpler to just use a higher sample rate.
Taking a 44.1k input and upsampling it to 192k is still somewhat complicated - and the process itself offers many choices and compromises.
It would be much simpler to distribute audio recorded at a 192k sample rate...
And then convert it using a simpler DAC which didn't need oversampling to deliver good performance.

In short, the reasons that were "compelling" for choosing 44.1k in 1978 are simply no longer true.
44.1k is currently being used as the sample rate of choice simply because "it's what our grandfathers used"... like horses... and gasoline powered cars.
(And that argument is becoming less compelling every year, as less and less music is played from CDs.
After all, unlike plastic discs, there's no particular benefit to standardize digital audio files at a single sample rate at all.)

I find it humorous how so many people seem convinced that "modern music producers" are "all trying to rip us off by foisting yet another audio format on us".
(I seem to recall people saying pretty much the same thing about CDs in the 1970's.)

It was a good compromise in it's day some 40 years ago and provided a huge improvement in audio transparency on consumer market for normal people.


Yes, but they had their reasons to choose 44.1 kHz. Fortunately it is able to provide transparent sonic experiment for human ears despite of filter difficulties. It is easy for us who live the age of 384 kHz DXD recordings to complain about low sample rate, but digital technology was different in the late 70's. Let's be happy about the fact they didn't choose 32 kHz!

The problems of 44.1 kHz are totally overblown out of proportion by hi-res business and other besserwissers who don't understand properly the scale of the issue. For me 16/44.1 is a transparent audio format. If it isn't for your bat ears then bad luck man!

 

[Whazzzup]

im not big on oversampling and its smoothing effects. however AK 380 does sound pretty smooth and warm with it

 

[castleofargh]

I don't disagree with you at all......

I have no complaints at all that 44.1k was chosen as the sample rate for CDs in the 1970's (and horses were a remarkably effective and efficient form of transportation in 1820).
However, as you point out, we now live in an age where there are lots of other alternatives, many of which are better.
The fact that CDs were an excellent compromise in the 1970's doesn't mean that we should stop trying to find something better.

One obvious example is oversampling.
It's impossible to design a practical, cheap, and effective reconstruction filter to operate directly with digital audio sampled at 44.1k.
Therefore, oversampling was invented as a way to sidestep this problem.
However, with modern technology, it would be much simpler to just use a higher sample rate.
Taking a 44.1k input and upsampling it to 192k is still somewhat complicated - and the process itself offers many choices and compromises.
It would be much simpler to distribute audio recorded at a 192k sample rate...
And then convert it using a simpler DAC which didn't need oversampling to deliver good performance.

In short, the reasons that were "compelling" for choosing 44.1k in 1978 are simply no longer true.
44.1k is currently being used as the sample rate of choice simply because "it's what our grandfathers used"... like horses... and gasoline powered cars.
(And that argument is becoming less compelling every year, as less and less music is played from CDs.
After all, unlike plastic discs, there's no particular benefit to standardize digital audio files at a single sample rate at all.)

I find it humorous how so many people seem convinced that "modern music producers" are "all trying to rip us off by foisting yet another audio format on us".
(I seem to recall people saying pretty much the same thing about CDs in the 1970's.)

sure, we all remember the years of arguments to prove an audible difference between CD and k7 tapes, or CD vs vinyls. it was "pretty much the same thing" like apples and oranges are pretty much the same.
what I can concede is that the marketing was already full of crap. high res didn't invent misguiding people with different masters for example.

 

[bigshot]

I have no complaints at all that 44.1k was chosen as the sample rate for CDs in the 1970's (and horses were a remarkably effective and efficient form of transportation in 1820). However, as you point out, we now live in an age where there are lots of other alternatives, many of which are better.

Not "better" just "easier". The end result is the same. It's like compressed audio. Yes it's easier to just encode in PCM. Modern compression codecs are much more complicated, but the end result is the same. At high enough bitrates, the sound is audibly transparent.

All that really matters is what you can hear. If it's audibly transparent and the usability is convenient, it's the exact same thing. There's absolutely no reason to go for a higher sampling rate now. All it will get you is sound to make your dog sit up and take notice.

 

[OddE]

There's absolutely no reason to go for a higher sampling rate now. All it will get you is sound to make your dog sit up and take notice.

-That, I think, is an overly simplistic view.

You will also need a bigger data plan and a ditto hard drive. (Though at today's prices, the latter isn't of much concern even if 32/768 or DSD-4096 is your poison of choice...)

 

[Whazzzup]

I have a cat