or Connect
AVS › AVS Forum › Audio › CD Players & Dedicated Music Transports › Audiophile CD Player? Which One?
New Posts  All Forums:Forum Nav:

Audiophile CD Player? Which One? - Page 2

post #31 of 535
Quote:
Originally Posted by FMW View Post

We already know you haven't really done any bias controlled testing. Why not try it and take as much time as you like to "adapt" to the sound before switching? I don't see any problem with that. It is time consuming but it isn't difficult. I guarantee you will learn from the experience.

I would actually prefer fast switching between two different devices in a test. Why would I prefer it? Because you can very much instantaneously switch between the same passage of music. Then even rewind the material and listen to it again.

If there are differences to be heard in such a case, fast switching would show them in an obvious way. If they were nonexistant or nearly nonexistant, this would be just as obvious.
post #32 of 535
Quote:
Originally Posted by MamaKrayma View Post

Why the heck was there a market for "hi-fi" CD players then? Was everyone fooled for decades by believing that there are superior DACs?

There are a number of reasons. The idea that audio gear in general sounds different was clearly true in the early days of Audio - the 1950s. Most audio gear had audible deficiencies and its just the nature of things that anything that has audible flaws has audible flaws that sound different. Even when this problem became less severe for say, amplifiers it remained true for vinyl playback gear and analog tape recorders. Making two turntables sound the same in an ABX test, sometimes even identical ones, can be a tough chore. Ditto for analog tape machines, and loudspeakers.

Reality is that the first CD players were head and shoulders more accurate than any LP player or analog tape machine of the day. However there was still a confusing factor, which is that their accuracy often stood behind a confusing mix of really well made CDs and some that were wildly substandard. If anything, the accuracy of even the earliest CD players made differences in production standards and techniques for recordings more apparent. How do you tell whether a bad sounding CD sounds bad because of the player it is being played on or the recording itself? It can be very confusing.
Quote:
And I'm actually more of an analog supporter. As such, I'm beginning to think that the articles I've read are propaganda, stating that pure analog always gives purer music.

Despite the conspiracy theories that some analog supporters circulate, pure analog is actually an oxymoron. The fact is that the performance of the best analog music players has always been substandard compared to good digital and good amplifiers.
Quote:
And dedicated CD players have always had cheaper DACs than any low-end receiver today?

Too many global words like always and any to be a true statement. However, a mid-priced or better AVR may perform better than the Redbook CD standard allows. Don't worry, the Redbook CD standard is more than good enough - due to as yet unresolved practical difficulties with recording technology, no recording has sufficient dynamic range to exceed what the Red Book specifies. That's another myth - that your system is the strongest limiting factor to the sound quality of your system. In reality the most severe limitations are often baked into the recording.
Quote:
What am I missing here? Or is this just a lot of junk that poor people say to make themselves feel better about having low-end systems?

Low end systems can be fantastic performers. There's a little $30 digital player called the Sansa Clip. It performs as well as many people's whole system and easily reveals the audible problems in problematical recordings. It sounds (and measures) about as good as a $250 iPod.
post #33 of 535
Quote:
Originally Posted by bo130 View Post

I would actually prefer fast switching between two different devices in a test. Why would I prefer it? Because you can very much instantaneously switch between the same passage of music. Then even rewind the material and listen to it again.

If there are differences to be heard in such a case, fast switching would show them in an obvious way. If they were nonexistant or nearly nonexistant, this would be just as obvious.

I don't disagree with you. I was just trying to make it easier for someone to commit to doing some bias controlled testing.
post #34 of 535
Quote:
Originally Posted by MamaKrayma View Post

Thanks for the insight (same to the rest of you). Excellent to know before I wasted cash on, well, a waste. So are components made by say, Oppo, a rip-off?
And that's why you never will hear the music how is supposed to sound. Because you trust a bunch of nihilists.
post #35 of 535
Quote:
Originally Posted by MamaKrayma View Post

Thanks for the insight (same to the rest of you). Excellent to know before I wasted cash on, well, a waste. So are components made by say, Oppo, a rip-off?

Nobody is suggesting that an Oppo player is a rip off. In fact it is a well made, reliable and flexible product. It is rightfully very popular. We are just trying to explain that you should buy it for the right reason. Better sound and picture is not the reason.
post #36 of 535
Wow... a lot of discussion. I understand there are only a handful of chip makers out there that make DACs but many these chip makers make are designs proprietary to other company's. The OP mentioned that he was looking for audiophile quality in a very low price. Audiophile means one thing to me, but may mean something different to others.

There are different DACs out there...16-bit, 24 bit, 32 bit. The DAC is nothing more than a processor with a firmware of proprietary algorithms converting analog to digital and back. The higher the bit, the higher the floating point precision. A 69 dollar player will use a lower bit DAC than a higher end player. So the higher bit DAC will provide a more accurate presentation. In a system, other things being equal, some people will hear a difference, others won't. When a person is looking for audiophile quality, to me, it is assumed this person can hear a difference.

I am not too familiar with scientific testing involved but I can say that no matter what testing is done, the algorithm used requires a constant representing what the human ear can hear. Apparently the accepted value is 20 to 20000 hz but it is not absolute as other can hear more, others less. I understand my claims of difference is subjective but nonetheless, I trust my ears and I go with it. But, being open minded, perhaps FMW can provide the necessary steps for this bias testing.

There was an interesting video I found some months back. Maybe some of you will find it interesting as well...

http://www.youtube.com/watch?v=1CkyrDIGzOE&list=PLA258FEB2312337B6&index=1

In any event, good luck to the OP on whatever he decides.
Edited by Gary Seven - 8/8/13 at 11:50am
post #37 of 535
Quote:
Originally Posted by Gary Seven View Post

Wow... a lot of discussion. I understand there are only a handful of chip makers out there that make DACs but many these chip makers make are designs proprietary to other company's. The OP mentioned that he was looking for audiophile quality in a very low price. Audiophile means one thing to me, but may mean something different to others.

There are different DACs out there...16-bit, 24 bit, 32 bit. The DAC is nothing more than a processor with a firmware of proprietary algorithms converting analog to digital and back. The higher the bit, the higher the floating point precision. A 69 dollar player will use a lower bit DAC than a higher end player. So the higher bit DAC will provide a more accurate presentation. In a system, other things being equal, some people will hear a difference, others won't. When a person is looking for audiophile quality, to me, it is assumed this person can hear a difference.

I am not too familiar with scientific testing involved but I can say that no matter what testing is done, the algorithm used requires a constant representing what the human ear can hear. Apparently the accepted value is 20 to 20000 hz but it is not absolute as other can hear more, others less. I understand my claims of difference is subjective but nonetheless, I trust my ears and I go with it. But, being open minded, perhaps FMW can provide the necessary steps for this bias testing.

There was an interesting video I found some months back. Maybe some of you will find it interesting as well...

http://www.youtube.com/watch?v=1CkyrDIGzOE&list=PLA258FEB2312337B6&index=1

In any event, good luck to the OP on whatever he decides.

The issue isn't differences in technology. It is differences in sonics. All the different DAC's wind up producing the same sound in a bias controlled listening test.
post #38 of 535
Quote:
So the higher bit DAC will provide a more accurate presentation.
OK.
Quote:
In a system, other things being equal, some people will hear a difference, others won't.
Not OK. There's been a fair bit of scientific work done on how well humans can differentiate sounds, and there are pretty clear thresholds below which nobody can hear differences. The argument being made here—and there's a lot of science behind it—is that even cheap mass-market DACs have distortion levels so low that no one could hear the difference between those DACs and a theoretically perfect DAC.
Quote:
When a person is looking for audiophile quality, to me, it is assumed this person can hear a difference.
It should be assumed that they think they can hear a difference, nothing more. And it's a fairly safe bet they've never really put themselves to the test.
post #39 of 535
Quote:
Originally Posted by FMW View Post

The issue isn't differences in technology. It is differences in sonics. All the different DAC's wind up producing the same sound in a bias controlled listening test.
With crappy reproduction devices following it in the audio chain... of course there is no difference.
post #40 of 535
Quote:
Originally Posted by SoNic67 View Post

Quote:
Originally Posted by FMW View Post

The issue isn't differences in technology. It is differences in sonics. All the different DAC's wind up producing the same sound in a bias controlled listening test.
With crappy reproduction devices following it in the audio chain... of course there is no difference.

It is true that compared to modern DACs, all electroacoustical reproduction devices known at the current state of the art are crappy.

A very good loudspeaker may have artifacts that are as much 40 to 60 dB down. That is one superlatively crappy modern audio DAC indeed! A good modern DAC has artifacts at a minimum of 100 dB down, ranging up to about 130 dB for SOTA products.

The artifacts inherent in the human ear are in the same range as loudspeakers, or worse.
post #41 of 535
That's why any tests that uses speakers driven to their maximum abilities should be dis-considered.
As for human ear being worse, that is just a red herring. Use proper reproduction tools in the chain and then you can qualify the ear...
post #42 of 535
Quote:
Originally Posted by SoNic67 View Post

That's why any tests that uses speakers driven to their maximum abilities should be dis-considered.

The true situation is far more severe than that. I suggest that you spend some time on DIY speaker forums. The people there have measurements that address this universal problem with speakers having order of magnitudes more distortion than even mediocre DACs.
Quote:
As for human ear being worse, that is just a red herring. Use proper reproduction tools in the chain and then you can qualify the ear...

You apparently have not been keeping up with research in this area since the 1980s. Interestingly enough many of the artifacts inherent in human hearing have been determined from that physical structure of the ear and first principles of physics.

Psychoacoustics: Facts and Models (Springer Series in Information Sciences)
Hugo Fastl (Author), Eberhard Zwicker

Is one of the standard references in this area.
post #43 of 535
Quote:
Originally Posted by SoNic67 View Post

With crappy reproduction devices following it in the audio chain... of course there is no difference.

Did you have some criticism of the equipment we used to do our bias controlled tests without even know the equipment was? Just a blind shot in the dark? Do the tests. Knock off this nonsense. You have no idea what you are talking about.
post #44 of 535
Quote:
Originally Posted by Gary Seven View Post

Wow... a lot of discussion. I understand there are only a handful of chip makers out there that make DACs but many these chip makers make are designs proprietary to other company's. The OP mentioned that he was looking for audiophile quality in a very low price. Audiophile means one thing to me, but may mean something different to others.

There are different DACs out there...16-bit, 24 bit, 32 bit. The DAC is nothing more than a processor with a firmware of proprietary algorithms converting analog to digital and back. The higher the bit, the higher the floating point precision. A 69 dollar player will use a lower bit DAC than a higher end player. So the higher bit DAC will provide a more accurate presentation. In a system, other things being equal, some people will hear a difference, others won't. When a person is looking for audiophile quality, to me, it is assumed this person can hear a difference.

I am not too familiar with scientific testing involved but I can say that no matter what testing is done, the algorithm used requires a constant representing what the human ear can hear. Apparently the accepted value is 20 to 20000 hz but it is not absolute as other can hear more, others less. I understand my claims of difference is subjective but nonetheless, I trust my ears and I go with it. But, being open minded, perhaps FMW can provide the necessary steps for this bias testing.

There was an interesting video I found some months back. Maybe some of you will find it interesting as well...

http://www.youtube.com/watch?v=1CkyrDIGzOE&list=PLA258FEB2312337B6&index=1

In any event, good luck to the OP on whatever he decides.

The part I italicized is making my head spin far more than the DSP classes I took in grad school ever did. eek.gif
post #45 of 535
Quote:
Originally Posted by Gary Seven View Post


The DAC is nothing more than a processor with a firmware of proprietary algorithms converting analog to digital and back.

Nope. DACs are purpose-built chips. The mixture of analog and digital processing requires more complex processing. There are a number of SOC (System On A Chip) products that have both a full-fledged computer function along with a number of DACs on the same chip, but these are far more complex chips than the ordinary DAC chip you find in a digital player.
Quote:
The higher the bit, the higher the floating point precision.

Digital audio is generally based on fixed point arithmetic. Floating point formats exist, but they are not widely used.
Quote:
A 69 dollar player will use a lower bit DAC than a higher end player. So the higher bit DAC will provide a more accurate presentation.

Not necessarily true. Repeats the audiophile myth that their is a linear relationship between cost and sound quality. In many cases the extra cost of higher priced products are due to the manufacturer and distribution/dealership chain simply taking higher profits. Also, there are high end products that make a big show out of expensive features such as milling the chassis of the device out of a piece of billet metal which does not necessarily improve sound quality. Finally detailed examination of some high end music players shows that they may use the same transport mechanism as a $30 boom box, may contain subassemblies from lower-cost products, or may be a lower cost product that is sold under a different brand name rebranded.

Quote:
In a system, other things being equal, some people will hear a difference, others won't.

Repeats the audiophile myth that there are extreme differences among the sensitivity of hearing among people with normal hearing, and that there is no limit to the sensitivity of human hearing.
Quote:
When a person is looking for audiophile quality, to me, it is assumed this person can hear a difference.

That assumption is just that an assumption, and it is easy to show that it often false. Many differences are perceived differences and relate to the circumstances, not the audible performance of the equipment.

Quote:
I am not too familiar with scientific testing involved but I can say that no matter what testing is done, the algorithm used requires a constant representing what the human ear can hear. Apparently the accepted value is 20 to 20000 hz but it is not absolute as other can hear more, others less.

There are variations in the range and sensitiviy of human hearing but just like other areas of human performance, there are limits. For example, a person who can run a mile in 1 minute remains elusive.
Quote:
I understand my claims of difference is subjective but nonetheless, I trust my ears and I go with it.

That is hard to quibble with, but baseless disputation of well known scientific findings is easy to quibble with.
Quote:
But, being open minded, perhaps FMW can provide the necessary steps for this bias testing.

It's up to you to educate yourself. ITU Recommendation BS 1116-1 is in a downloadable PDF, and contains the details of one generally recognized procedure for doing bias-controllled listening tests. If you have any specific questions based on it, several of us would be happy to explain them to you.
post #46 of 535
Quote:
Originally Posted by mcnarus View Post

Absolutely not. They play numerous audio formats. They handle video extremely well. If you need a machine with those capabilities, they are the right tool for the job. And there's nothing wrong with buying something because it's extremely well-made, or novel in some way, or just old-school.

For that matter, there's nothing wrong with buying something because you think it sounds better, even though it really doesn't. The only problem comes when you try to give other people advice based on your own faulty comparisons. Then you're going to get push-back, at least in forums like this one where all viewpoints are allowed.


I agree with McNarus - If the physical appearance is important (say your system is not in in a closet) then spending somewhat extra for well build and well designed (e.g front panel interface) components has value. If you want a reliable system (e.g. mechanical transports, quality PCB's inside) that will also cost somewhat more. Everything else is debatable.

A long time (in the early '90's) when I was a grad student, we did some collaborative work at Bell Labs which also happened to be nearby. As you can imagine, these guys had some expertise in audio reproduction (how many calls can we put on a copper line and still have you recognize grandma's voice?) and access to some pretty good equipment.There was an audiophile crowd there with a friendly debate on analog vs. digital sources (i.e. LP's vs CD's) and the favorite speakers of the time. Everyone recognized the superior fidelity of digital, but also that many preferred what they were used to - LP's, but that was part of the debate. Anyway, I sat in one of their weekend tests looking at CD players of the time, using a 'reference' system in a sound chamber. Of the ones tested (~4 or 5), people could reliable differentiate only one player - a mass market Sony. Why? well upon disassembly, a soldering defect (whisker) was found, causing the fifth bit up from the LSB to be held high. So it was good that trained ears could pick this up (I missed it, failed the ABX test), but it was illuminating that opinions were divided on whether it sounded better or worse.
post #47 of 535
Digital crap mixing? Don't you realize that the source is digital as well, which means any manipulation of the signal in digital domain has very little if any side effects, unlike when it's done in analog. The so called signal purity and length have no meaning in the digital world.Analog jockeys need to wise up to that fact.
Edit: this post was in response to post#14.
Edited by thehun - 8/11/13 at 8:23pm
post #48 of 535
Quote:
Originally Posted by SoNic67 View Post

I did my own tests, measurements. I know what to "look for". I did my own ABX to see if is meaningful. I don't need nobody control to know what I heard.
Your "controlled" tests will show nothing useful. The are the same like saying that holding a hand in ice and the other in boiling water, in average you feel good.
Statistics without 'population standard deviation' understanding means squat. If only 1 person in 100 hears the difference, the difference exists, indifferent if the rest 99 can't hear it.

when I saw this thread I immediately thought of fireworks and here we go!

anyway, i'm not here to say whether or not the said differences exist, as i'm no high end audiophile, rather a budget bang for buck enthusiast.

but I know a bit of statistics. let me explain (if you would be kind enough to read)

if you have two people with a painful problem - give someone a sugar pill and the other nothing, the person with sugar pill feels less pain. guaranteed. search placebo effect
if you have two sick people and one person you research their condition and keep visiting them and doing meaningless tests they feel better because of the extra care put into their package compared to the other sick person who got the standard care hospital care who feels less well. (not the best analogy for this situation but it's true - get some highly trained researchers looking after you and their collective experience and knowledge put into your care will be superior than If they weren't part of your care package)

if you ask a researcher to interview the mood of a patient and you told the researcher that the patient took the study pill s/he would rate the patient's mood better, than if you told the researcher the patient took a sugar pill. hence the researcher is not supposed to know which patient took a study pill or a sugar pill. this is called interviewer bias.

if you told an audiophile system X costs $10,000 and system Y costs $1,000 dollars s/he will say system X sounds better. this is called bias. unless you are a robot you will rate it that way.

when these guys suggest a "controlled test" the listener must not have any preconceived ideas about the system they listen to. the mind is more powerful than you think.
i'm sure you already know they want the room to be the same, the SPL to be the same, the everything else in the chain to be the same, etc but nowhere near as important as a preconceived idea of what you are listening to.

i'd love to do one of those tests myself to see for myself.


PS
holding a hand in boiling water and a hand in ice and averaging them is different statistics and must not be applied here. you are saying give two SEPARATE people system X and system Y and then both give one score and you average out the score. in this statistic you are talking about ONE hand in TWO waters. not TWO hands in TWO waters.

PPS
population standard deviation in this test means, if you took a random sample of sound enthusiasts make EACH person you picked to do the SAME ABX test. write the result down for EACH person's test. plug that result into a correct statistical test (let's say a binomial test). that statistical test should tell you - if there is no difference in system X vs Y then 50% of each will pick X and 50% pick Y. if there is a difference then the binomial test will tell you that. in this test there is no population standard deviation because it is either 50-50 or it is not. the population standard deviation of binomial tests in this case exists because your sample is an attempt to see if the entire population would give you the same result as your test. which you will never know unless you test everybody, which you won't be able to. hence the population standard deviation of binomial tests is then derived from a few binomial tests to find out the probability of your test being accurate reflection of the population of sound enthusiasts
Edited by joker97 - 8/12/13 at 3:17am
post #49 of 535
The only difference is that I am talking about systems that cost UNDER $1000.
For example one that is $30 and one that is $300 - you always can tell the difference. Don't go where the diminishing returns law works in your advantage and generalize that EVERY device, regardless of price will sound identical/good.
post #50 of 535
Thread Starter 
Quote:
Originally Posted by thehun View Post

Digital crap mixing? Don't you realize that the source is digital as well, which means any manipulation of the signal in digital domain has very little if any side effects, unlike when it's done in analog. The so called signal purity and length have no meaning in the digital world.Analog jockeys need to wise up to that fact.
Edit: this post was in response to post#14.
Which is why this thread has changed my perception of such. It doesn't help that the first person who told me about MP3s being lossy was also the first person to indoctrinate me to analog "purity." He was a good source for some points, biased source for others.
post #51 of 535
Quote:
Originally Posted by SoNic67 View Post

The only difference is that I am talking about systems that cost UNDER $1000.
For example one that is $30 and one that is $300 - you always can tell the difference.

As long as good digital players sell for less than $30 that claim will be false.
Quote:
Don't go where the diminishing returns law works in your advantage and generalize that EVERY device, regardless of price will sound identical/good.

I can remember when all portable digital players cost more than $200, but time and technology march on.
post #52 of 535
Quote:
Originally Posted by joker97 View Post

if there is no difference in system X vs Y then 50% of each will pick X and 50% pick Y. if there is a difference then the binomial test will tell you that. in this test there is no population standard deviation because it is either 50-50 or it is not.

It isn't quite that simple. Our tests had 10 iterations X 10 listeners so we had a numeric base of 100. 49/51 for 48/52, was considered just as random as 50/50. In those instances concluding that the listeners were guessing was valid. You can flip a coin 100 times and not get 50/50.
post #53 of 535
Quote:
Originally Posted by FMW View Post

Quote:
Originally Posted by joker97 View Post

if there is no difference in system X vs Y then 50% of each will pick X and 50% pick Y. if there is a difference then the binomial test will tell you that. in this test there is no population standard deviation because it is either 50-50 or it is not.

It isn't quite that simple. Our tests had 10 iterations X 10 listeners so we had a numeric base of 100. 49/51 for 48/52, was considered just as random as 50/50. In those instances concluding that the listeners were guessing was valid. You can flip a coin 100 times and not get 50/50.

I think he sort of covered that with his mention of "...the binomial test..." meaning comparison to a binomial distribution, which feeds exactly into your point.
post #54 of 535
Quote:
Originally Posted by FMW View Post

It isn't quite that simple. Our tests had 10 iterations X 10 listeners so we had a numeric base of 100. 49/51 for 48/52, was considered just as random as 50/50. In those instances concluding that the listeners were guessing was valid. You can flip a coin 100 times and not get 50/50.

either you are doing a wrong method of validating a hypothesis or you are using the wrong statistical test on a well designed study

ie if the way you set up the test may need to change or if you are doing a perfectly good study then the binomial test does not apply and you need to use a different test.

if you flip a coin 100 times and you don't get 50-50 does not make the binomial distribution incorrect - that's just the way statistics works - a lot of numbers and if you do the numbers right you come up with a conclusion to accept or reject a null hypothesis
post #55 of 535
Quote:
Originally Posted by joker97 View Post

Quote:
Originally Posted by FMW View Post

It isn't quite that simple. Our tests had 10 iterations X 10 listeners so we had a numeric base of 100. 49/51 for 48/52, was considered just as random as 50/50. In those instances concluding that the listeners were guessing was valid. You can flip a coin 100 times and not get 50/50.

either you are doing a wrong method of validating a hypothesis or you are using the wrong statistical test on a well designed study

ie if the way you set up the test may need to change or if you are doing a perfectly good study then the binomial test does not apply and you need to use a different test.

if you flip a coin 100 times and you don't get 50-50 does not make the binomial distribution incorrect - that's just the way statistics works - a lot of numbers and if you do the numbers right you come up with a conclusion to accept or reject a null hypothesis

Let's be clear about this. If you flip a coin 100 times over and over again, sooner or later you are going to encounter a run of 100 heads or 100 tails. More likely will be sets of 100 trials of with 55 heads or 51 tails or whatever. The binomial distribution has the exact answers about what to expect.
post #56 of 535
ah that's what I was supposed to say biggrin.gif thanks ... I haven't done statistics since the old uni days LOL
post #57 of 535
As always, listen for yourself; if you hear a difference that improves the sound quality in a way that makes a difference for you, purchase that player.
post #58 of 535
Quote:
Originally Posted by kucharsk View Post

As always, listen for yourself; if you hear a difference that improves the sound quality in a way that makes a difference for you, purchase that player.

This is a way to spend money and get nothing in return. If treating your audio dealer as a charity is your goal, then this is good advice!

Reason why? If you casually listen for yourself, you will probably hear a difference and it might even be what you want. This is called expectation bias, and its wired into our brains. Regrettably, it is not wired into audio equipment. This is one reason so many audiophiles are on a purchase, purchase, purchase treadmill.
post #59 of 535
Quote:
Originally Posted by arnyk View Post

This is one reason so many audiophiles are on a purchase, purchase, purchase treadmill.

Amen. I went through decades of that. Been there, done that. What waste of money. Since seeing the light, I get much more enjoyment out of my music listening because I'm not listening for gremlins. My cars are now worth more than my audio system so I've saved a bundle. You can make meaningful improvements to your sonics but CD players aren't the way to do it. Speakers and room acoustics are the way to do it.
post #60 of 535
It's funny how these debates are dominated by tone-deaf statisticians. The energy they put forth trying to beat their opinions and values into others is awesome smile.gif
Me, I'm happy for ANYONE who has arrived at a point where they can "hear the music", smile, sit back, and relax. Did yours keep you and the wife up too late again, listening and tapping? Mine did, damn proud of it too.
As I remember, that is the basic point, true?
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Audio › CD Players & Dedicated Music Transports › Audiophile CD Player? Which One?