Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CD audio is not good enough (Read 44773 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

CD audio is not good enough

Reply #25
Quote
Nyquist called for a *minimum* of 2x the frequency. Many papers have
shown that the bare minimum is not sufficient. Some are even
available on the web, if you choose to go looking.

Can you provide some links please? I may have read them, but I always like something new. If you search on this forum, you'll find this is one of my favourite topics (I nearly did a PostDoc on it), and I'm not always in agreement with KikeG

Quote
But never mind theory. In practice there are many people who can hear
the deficiencies caused by low sampling rates, usually in terms of
loss of clarity in the high end.


This is not my experience. The difference between 44.1k and 96k is like the difference between a good and bad amplifier (for example) - the music just sounds more "real", and the sound stage is slightly better defined. It's got nothing to do with frequency response, or weakness/improvement in any particular range. If anything, I thought the bass was effected most, but that's a quirk of human percetion, rather than there actually being any difference in the reproduction of the bass.

What I'm saying is that what you report is the "expected" improvement. Whenever people do something, and then magically hear the improvement that they expected, there's one word springs to mind...


Quote
...it's nice to remove one source of loss or
distortion from the signal chain, because certainly signals have been
recorded and manipulated that exceed 16-bit. I'd rather overkill than
underkill (if there is such a word).


Of course there's no harm. It's questionable if there's any good, but I agree that theoretically there could be. Again, what experience people actually hear does not match the expectation: the improvement due to increased bitdepth has nothing to do with decreasing the background nosie from 7dB SPL -10dB SPL (typcal measurements if you use an SMPTE calibrated system to replay 16 and 20 bit audio). At least not audibly. It sounds like something completely different - it removes some hash and distortion. Distortion which, theoretically, doesn't exist.


Quote
> The system is very
> robust, and if you commonly encounter audible errors than I suggest
you
> replace damaged CDs or invest in a good quality CD player.

You talk of using a low-end "Techniques" and then say *I* need a
better player. Ha!

> In practice my experience is
> that error correction is not an issue for CD, so I find your claims
to be
> unusual and suspect.

It depends on the CD medium. Metal-stamped CDs are less of a problem
than dye-based CD-Rs. I have examples of these that regularly
demonstrate painful noise that even the most casual listeners can
hear.


If you're expecting to play CD-Rs on your player, but it gives audible noise on them, then something's wrong. Are the CD-Rs burnt with good data? Are their errors in the original .wavs? If not, then it's a compatability problem between the CD-R and your player. It happens. It doesn't mean that this is a general situation with most players and discs.

As for damaged discs: Anyone who wants to play a disc with a 1mm scratch on it, LP or CD, probably isn't too concerned with the sound quality. If they are, they should take better care of their software, or replace it often!

Quote
That is the problem. You want to *see* the evidence. Open your ears
and you can *hear* it plain enough.


Not everyone has the equipment, inclination, or ears to do this. FWIW I heard the difference between 44.1 and 96k using some excellent DCS DACs and excellent amps and speakers. I learnt to hear the difference in a sighted demonstration, and then picked out 44.1k when they were switched without me knowing which was which.

At the other end of the scale, I can pick out 32kHz on my cheapest soundcard at home (and, for my hearing, 32kHz should be enough). The reason is obvious: the soundcard is terrible!

However, at home with my audiophile 2496, I can't hear any advantage of 24 bits over 16, or 96k over 44.1. That's just the way it is!



Quote
I don't see how it can possibly hinder your
enjoyment of music if others choose to listen to a different digital
standard.


Well, if almost everyone in the world listened to 128kbps mp3, that's what the record companies would sell, and everyone here would be very unhappy! As it is, most people prefer overcompressed rubbish, and we are mostly unhappy with this.

DVD-A (and probably SACD) can't sound worse than CD (though some early DVD-A players were reportedly questionable), but you've got to be very sure that they're better before you're going to think about spending huge amounts of money on them. Blind tests to prove this point are notable by their absence - could it be that the "advantages" just aren't that big?

Cheers,
David.

P.S. - Don't point me to audiophile forums where people are saying how great there new expensive SACD player sounds. They might be right, but they're hardly unbiassed. I never believe anyone who tells me how great their new toy is anymore. For example, there are far too many people in the UK saying how great the picture on their new digital 16:9 TV is. The truth is usually an analogue 4:3 image stretched to 16:9 (wow - yeah - that's great - the people all look so, er, fat!); or a digital image with so much blockiness that the folks at doom9.org would tell you to re-encode it! The "new toy" syndrome is very strong. People don't want to think that they've wasted their money, or that they've been conned.

CD audio is not good enough

Reply #26
Quote
It depends on the CD medium. Metal-stamped CDs are less of a problem
than dye-based CD-Rs.

That's right, I didn't take into account CDRs (dye-based), because there are too much that are of poor quality, burned in TAO, or that are simply showing a lot of uncorrectable errors because they are unusable after three years of storage. We were only talking about pressed (metal-stamped) CDs.

CD audio is not good enough

Reply #27
Joseph, you end your last post by saying,

Quote
But others tire of this thread, so I will not comment further. Except
that I must wonder why you cannot accept that others are not happy
with the status quo. I don't see how it can possibly hinder your
enjoyment of music if others choose to listen to a different digital
standard.


I don't think anybody would criticize you for wanting to listen to 96kHz/24bit sound if you simply say that it makes you happy and it gives you a warm sense of security that you aren't missing anything. 

But it was YOU who originated this thread and gave it its inflammatory title, which states plainly that 44.1/16 "is not good enough" and "is bad quality".  Did you expect NOT to be criticized for that?  There is an abundance of theory and practical experience, much of it cited above, that suggests that 44.1/16 is more than adequate for an end-product, consumer CD.  Based on the posts in this thread, it seems to me that a substantial majority of the people here agree with that view, rather than with yours.

I don't think that 96/24 can be established as "superior" in any way in the absence of a thorough (and difficult) double-blind ABX study, and there is no such study.  The work cited by Pio2001 and KikeG above just isn't convincing.

CD audio is not good enough

Reply #28
Quote
That's right, I didn't take into account CDRs (dye-based), because there are too much that are of poor quality, burned in TAO, or that are simply showing a lot of uncorrectable errors because they are unusable after three years of storage. We were only talking about pressed (metal-stamped) CDs.

Assuming one records on an unreliable medium,  is it supposed to sound better in 24/96 with an equal
or greater (if there are more problems with higher density) percentage of missing data?

I don't think so.

CD audio is not good enough

Reply #29
Quote
Based on the posts in this thread, it seems to me that a substantial majority of the people here agree with that view, rather than with yours.

I don't think that shows anything. There are many people on this board that agree with Joseph (to varying degrees) but don't bother to post because they've gone through it all before in previous threads, or don't feel that they can make the required scientifically compelling argument (with links to certified ABX results) to support their "outlandish" claims.

If the thread had been titled "Nothing sounds better than a CD on a $99 player" I suspect that the same people would not bother responding "me too" and "of course" and "duh!".

edit: minor clarification

CD audio is not good enough

Reply #30
Quote
Joseph wrote:

But never mind theory. In practice there are many people who can hear the deficiencies caused by low sampling rates, usually in terms of loss of clarity in the high end.

To clarify some of these points:

1) Nyquist:
Quote
The Nyquist numbers are pure theory and do not take into account
implementation. For example, Nyquist requires a perfect low-pass
filter for the digital-to-analogue conversion. Well, such a thing
does not exist! Real filters are not perfect, but rather introduce
frequency, aliasing, and phase anomolies. Though techniques like
oversampling can help, a sampling rate of 96KHz is the perfect
solution, as it puts all of these distortions above the range of
human hearing.

Nyquist called for a *minimum* of 2x the frequency. Many papers have
shown that the bare minimum is not sufficient. Some are even
available on the web, if you choose to go looking.


I was illustrating that the CD format can perfectly reproduce a 20kHz sinewave and all frequencies below that - the Nyquist theory holds well enough in practice.  I would agree that you need a very good anti-aliasing filter during A/D conversion to prevent aliasing, but you said one is needed for D/A conversion.  This is easily implemented for CD playback with a digital LPF in oversampled systems (4x is probably sufficient in practice) followed by a very simple analog LPF to filter out the ultrasonic alias.  Do you have any evidence that 96k has ABX-able superiority to 44.1k for high end clarity?  I have not seen any of that evidence and would be interested to see it.  In practice the LPF are not really a big problem with digital filters and oversampled A/D and D/A conversion, but I would agree that it would be nice to have the extra margin of 96k, the point being though that I don't agree that this makes "CD audio not good enough" as you titled the thread.

2) Noise shaping:
Quote
Sure, there are many ways to get the most out of CD. I am familiar with POW-R dithering, etc. I'd rather a format that does not require all of this finagling but rather delivers bit-for-bit accurate signals. >edit< Well, there's "quite good" and then there's "indistinguishable from the master". Personally, I'd go for the latter.


The CD format does deliver bit-for-bit accurate results of its 44.1/16 signal.  The point was that with noise shaping to get the most out of the format that 110dB dynamic range is not a limitation and calls the question as to whether "CD audio is not good enough."  Does the medium need to be "indistinguishable from the master" to be good enough for mass distribution?  I don't think that needs to be a requirement.

3) Turntable performance:
Quote
Any decent turntable can play any decently maintained vinyl record with an almost complete lack of background noise. Heck, even my mid-range Linn does a fine job.


Your words "almost complete lack of backgound noise."  Just keeping things honest on the hyperbole.  The Linn is a fantastic piece of equipment by the way.

4) Real world CD players:
Quote
You talk of using a low-end "Techniques" and then say *I* need a better player. Ha!


The point was that even an 18 year old "low-end" Technics player can produce a nearly perfect frequency response from 20Hz to 20kHz - real world example of CD audio being a robust, high quality medium.  The other point was that if your current CD player is not capable of playing back a decent CD without errors then you should replace it becuase it is probably not working properly.

5) Credibility:
Quote
I am not too sure why you doubt my veracity. I have a degrees I could wave around but I'd rather not appeal to authority.


I doubt the claims of frequent high error rates in CD becuase that has not been my experience or observation of others' experiences in the last 20 years.  Yes, please keep the degrees tucked away as they are not relevant to a lively debate and I do hope you don't beleive you are the only one on this forum with academic credentials.

The debate is indeed interesting and I am a fan of "more is better" in signal quality, but I do not agree that CD audio is not good enough.  Especially when the alternative formats (SACD and DVD-A) have so many other real world issues (encryption, watermarking, analog only outputs, high prices, etc.).

Perhaps a better direction for this thread would be to ask "good enough for what?"  Listening to music at a PC?  While jogging?  In the car?  In a typical home?  In an anechoic chamber?  Let's not forget the ability to copy legally and make the music portable - there are sound quality, cost, portability and other real world constraints that make a medium good enough.
Was that a 1 or a 0?

CD audio is not good enough

Reply #31
Quote
Nyquist called for a *minimum* of 2x the frequency.

False. Nyquist said that 2x the max. frequency is enough for perfectly sampling a band-limited signal, from a mathematical point of view. There has been some debate around if you need just 2x or just over 2x, that is. 2.00001x and such would be enough. I agree with the later interpretation.

Still, sampling at a frequency just over 2x would need a perfect filter. It is true that back in real world there are no perfect filters. That is the reason why the sampling frequency is of 44.1 KHz, to have some margin and being able to reproduce up to 20 KHz with good accuracy. In practice, with today's DACs, up to 21 KHz and a little more is possible.

Quote
Many papers have
shown that the bare minimum is not sufficient. Some are even
available on the web, if you choose to go looking.


Well, cd-audio doesn't use bare minimum, as is obvious.

About phase errors: there are no phase errors in today's DACs, because they use oversampling and FIR digital filters that have just linear phase distortion. Linear phase distortion is just a time delay, or in other words, no distortion from an audible point of view.

About aliasing: this is a non-issue in practice, unless you can hear very well on the range between 21.5 KHz and 23 KHz, where there is some remaining aliasing due to the filters used commonly.

Now, just don't say you are not interested anymore in the discussion. You have pointed to some flaws is cd audio, we have rebated them from a reasoned and technical point of view. Please say where you don't agree with these explanations.