Understanding specs for CD players..

[Taco Vine]

Hi, I have an older CD player and 2 dvd players. Which provides the best audio sound?

A) DVD
- Frequency range of 48khz sampling 4 hz to 22 khz and 96 khz sampling 4 hz to 44 hz.
- Signal to noise ratio of "more than 90 dB"
- harmonic distortion of "less than 65 dB"
- dynamic range: "more than 80dB"

B) DVD
- Frequency range of 48khz sampling 4 hz to 22 khz and 96 khz sampling 4 hz to 44 hz
- "total harmonic distortion of .004%"
- Signal to noise ratio: 110dB
- dynamic rage: 100dB

C) CD player
- Freq range of 2 hz - 20 khz +- 0.5 dB
- Dynamic range of 98dB
- Harmonic distortion "less than .003% (1khz)
- Signal to noise: 110dB

thanks!

[REDone]

I posed a similar question on another forum

Answer given ~ "What do your ears tell you?"

It doesn't help much if you are trying to understand or equate the specs with the sound before buying.

But if you already have these .. What do your ears tell you?

my guess in your case is the cdp is smoother sounding

[ericson38]

You can have great CD specs (like the ones you show), but these are valid even for a player that does not oversample. You can see this at the Yamaha manual website for CD players-

http://www.yamaha.co.jp/manual/english/result.php?div_code=av

So look into the # of bits used to encode the amplitude, and the oversampling rate, as well.

[Taco Vine]

You can have great CD specs (like the ones you show), but these are valid even for a player that does not oversample. You can see this at the Yamaha manual website for CD players-

http://www.yamaha.co.jp/manual/english/result.php?div_code=av

So look into the # of bits used to encode the amplitude, and the oversampling rate, as well.

?Not sure I understand.. I'm these specs means something since the mfg list them in the manual..

[ericson38]

What you should look at for CD specs, besides the parameters you listed, are oversampling rates and the number of bits used to encode the amplitude of the signal. For the Yamaha CDX920 for example, the # of bits is 18, versus 16 (nominal) and the bits are sampled by the player at 8 times the data rate of 44.1 Kb.

http://www.usersmanualguide.com/yamaha_audio/cd_players/cdx-920

Other CD players can have low noise and distortion, but it is not the whole story, since the CD player is more than an amplifier, its first chore is digital to analog conversion, then filtering and small amplification.

[Taco Vine]

What you should look at for CD specs, besides the parameters you listed...

I'm not sure how to read these specs.. based on what I provided, which has the "best" specs? thanks..

[slow_jazz]

The human ear is the best spec.....

[JustBuyNew]

Isn't the proof in the pudding?

[gusten]

One problem with spec is that they are very complicated to read and understand. Another is that spec presented are some that are rather easy to achieve and safe for the manufacturer to present.
All in all one can normally say, for consumer products, there is to much missing in presented data for a serious judgement regarding what to choose.
/gusten

[Taco Vine]

The human ear is the best spec.....

My guess is that most people would not be able to hear the difference between any two CD players.

[Byrdsmaniac]

Taco Vine:
As I understand it, the Red book CD standard contains the specs originally defined by Sony and Philips to make the CD medium capable of making transparent copies of more than 99% of studio master tapes of music. It also allows properly encoded discs to be playable on any player that has the Compact Disc logo on it.

Among the specs are:
a frequency range up to 22.05 kHz*
44,100 samples per second
16 bits per sample
96 dB signal to noise ratio ( 16 bits)
.002% distortion is approximately 16 bit performance (which is -96 dBs. .001% is equal to -100 dBs )


*(I don't recall a +/- range, but many inexpensive players have been capable of measured frequency response deviations of a small fraction of a decibel which is completely inaudible.)

This means that the specified performance of player A in your example does not actually meet the Red book standard for dynamic range, signal to noise ratio, or distortion. Dynamic range and signal to noise ratio are calculated at 6 dBs per bit, so a dynamic range of 80 dBs is only 13.33 bit performance and a 90 dB signal to noise ratio is equivalent to just 15 bit performance. The distortion spec is particularly poor since -65 db is just under 11 bit performance and is getting close to the -60 dB mark which is equivalent to .1% which may be audible at some frequencies. I imagine this is a very cheap DVD player.

The specs for player B are better. Distortion is far below audibility, S/N ratio is better than 17 bit performance and dynamic range is above the 96 dBs required for true 16 bit performance. Player C is equally good and the numbers are especially good for a CD player. I say that because the DVD standard requires better performing DACs (digital to analog converters.) than the Red book CD standard does.
The reason that these specs matter is that they tell you something about the quality of the DACs in your players. In digital audio, overall performance of a system can't be better than the measured performance of the DACs. It can not. If your digital to analog conversion isn't up to at least Red book level, it won't matter how good your speakers or amplifiers are becuase the ultimate quality of sound the system reproduces will be compromised. The system can't output information that has been filtered out by low resolution DACs.

However, the differences in your players may or may not be readily audible because of lack of resolution in the rest of the system, a noisy environment, or as you correctly surmised: most people just can't hear the difference most of the time.

[ericson38]

Question on your analysis (thanks)-

The Yamaha CDX 550 (sampled X 1 and 16 bit quantization) specs are:

2 Hz-20kHz
S/N-106dB
Dynamic range-100dB
% Distortion- 0.0028% (distortion and noise)

channel seperation (not specified)

The CDX 920 (sampled X 8, 18 bit quantization)-

2 Hz- 20 kHz
S/N-120dB
Dynamic Range-more than 100 dB
% distortion - 0.003%

channel seperation - >96 dB.

Unless I have something wrong here, the #s advertized for the CDX 550 are better than one can have with 16 bit/44.1 KHz sampling you predict is possible.

The 550 is slightly cleaner in distortion at 1KHz than the 920 (where the level is also specified), but the distortion accross the whole band (10-20 KHz) is going to be lower with the 920, since with oversampling the ailiasing errors can be kept out of the band edges on the high side, where presence is developed.

[REDone]

My guess is that most people would not be able to hear the difference between any two CD players.

Incorrect in practice

Tried a rotel RCD02 24 bit ~ terrible sound not even close to reality & measured against an old Mission DAD 7000 14+2 bit

Just got a Yamaha CDX 880 20 bit very close to vinyl & would push that Rotel model into the Victorian era of sound reproduction ie 19th C

[Byrdsmaniac]

ericson38, you said:
<< Unless I have something wrong here, the #s advertised for the CDX 550 are better than one can have with 16 bit/44.1 KHz sampling you predict is possible. >>

Yes, you are correct. I was just reporting the theoretical limit for 16 bit equivalent performance. 16 bit performance is the limit for the compact discs themselves. Playback devices can always be made better than the media that they are designed to play. What your quoted specs mean is that Yamaha used DACs and other components that are actually specified better than the CD standard calls for. That's not unusual for Yamaha. It's also a good thing because over specified components can lead to better actual measured performance. Your player shouldn't limit the output from your CDs in any audible way.

This does not mean that a 16 bit recording will yield more bits on playback, it only means you are more certain of getting a full 16 bits worth of performance under any circumstances. In actual practice, audio devices often don't live up to their written specs. In addition DACs themselves are specified as typical or worst case. The difference is usually at least a bit. I have observed that Yamaha's players usually measure as good as or better than their written specs. I think you will be hard pressed to find amplification that is cleaner than the output from this player. Only very good speakers are likely to do full justice to the signal quality put out by this unit.

The trivially slight difference between your units in distortion is most likely completely inaudible, but the CDX 920 has a significantly better specified S/N ratio. The CDX 920's 120 dB S/N is equivalent to 20 bits, where the 550's S/N is equivalent to 16.66 bits. I doubt that anything over true 16 bit performance would make an audible difference, but again, these are just specs and not actual performance measurements. I personally feel that lower noise is more important than reducing distortion much below .09%. In any case, it would appear that measuring distortion at only 1 kHz is just a ploy to make the unit stand out on paper. It would seem much more relevant to know the maximum distortion over the entire audible frequency range.


REDone: Unless there is something wrong with your Rotel, I would venture to guess guess that you did not match levels, or try to eliminate any possible influence of bias in your comparison. Level differences are often perceived as quality differences. Even unconscious bias colors everyone's perceptions. It really makes a tremendous difference when you are comparing components that you don't know which you are listening to. Even beer comparisons can come out differently than expected once the tasters don't know what they are drinking. Just my opinion...

[ericson38]

Thanks for your reply.

I'm actually in the market to replace our portable Sony Car Discman (1986 model) that is driving a Sansui G8000. That's one reason I have been watching this forum.

Charlie

[REDone]

REDone: Unless there is something wrong with your Rotel, I would venture to guess guess that you did not match levels, or try to eliminate any possible influence of bias in your comparison. Level differences are often perceived as quality differences. Even unconscious bias colors everyone's perceptions. It really makes a tremendous difference when you are comparing components that you don't know which you are listening to. Even beer comparisons can come out differently than expected once the tasters don't know what they are drinking. Just my opinion...

The Rotel was a second hand model from a local dealer & I didn't know its history or whether it had been tampered with, but as far as matching I really wanted to like it & tried hard but it played about an octave higher.

Piano was awful .. Pavarotti had a hernia ..

I can only speak as I find & this Rotel £350 $700 CD player was trounced by a Philips £30 $60 DVD player

Anyway I used the example to illustrate that CD Players can sound different

[wsjoe]

I think we tend to rely a little too much on specs. To me audio is a very subjective matter because of our deteorating hearing, different speakers and room acoustics, it's impossible that two systems, let alone components, will sound the same.

The best specs in cd player technology will not always sound the best.

my two cents...

[Byrdsmaniac]

<<it's impossible that two systems, let alone components, will sound the same.>>

Philosophically speaking, I suppose there may be something to what you are saying. However, on the ordinary day to day level of verifiable facts I'd have to say this is clearly untrue. If you can't reliably tell which of two components is playing, which is a very common occurrence, how else would one describe the situation except to say that they "sound the same?"

I know it goes against the prejudices of some participants on this excellent forum, but when it comes to digital equipment measured performance correlates very strongly with perceived performance. Now, please refrain from dumping on me the usual anecdotes about how you heard such and such to counter what I've just said. Such tales are cheap because they are so common and so unreliable. They only serve to reenforce popular beliefs that some may find it uncomfortable to question while adding little of substance to any discussion. Saying "but I heard it" in this context is little more convincing than the latest guy on TV saying that he saw big-foot or Elvis. The question is: can he prove that the king has returned or that creatures walk among us?

No one entirely is immune to personal bias when making subjective evaluations, so it only makes sense to do comparisons that filter out the possible effects of that bias. That way any claims that result will be more likely to taken seriously. Don't forget that there are countless documented instances where people making subjective preference choices had strong preferences even when the items in question were being compared only to themselves. Human perceptions are fallible, that's why measurements and controlled experiments exist.

Nevertheless, I'd love to have an Accuphase E-308 integrated amp, but since I don't have unlimited funds I make do with a Yamaha which costs one eighth as much. It doesn't bother me a bit because the money I saved buys a lot of concert tickets, recordings, and meals and I can not hear any difference anyway. If I had a ton of dough, I might still like the Accuphase though. I know I will likely perceive it as sounding better than my Yamaha as long as I actually know which one is playing; which if you think about it is actually kind of cool. The human mind is a very weird thing.

[sleddogman]

Hi, I have an older CD player and 2 dvd players. Which provides the best audio sound?

A) DVD
- Frequency range of 48khz sampling 4 hz to 22 khz and 96 khz sampling 4 hz to 44 hz.
- Signal to noise ratio of "more than 90 dB"
- harmonic distortion of "less than 65 dB"
- dynamic range: "more than 80dB"

B) DVD
- Frequency range of 48khz sampling 4 hz to 22 khz and 96 khz sampling 4 hz to 44 hz
- "total harmonic distortion of .004%"
- Signal to noise ratio: 110dB
- dynamic rage: 100dB

C) CD player
- Freq range of 2 hz - 20 khz +- 0.5 dB
- Dynamic range of 98dB
- Harmonic distortion "less than .003% (1khz)
- Signal to noise: 110dB

thanks!
As it pertains to your three choices, I'd say it depends on what type media you put in it.

PCM/2 and AC3/5.1 sound tracks on a DVD movie disk will be processed within the DVD audio specs you gave, but I wouldn't guarantee you could expect much better than standard Redbook specs if you put an audio CD in either DVD deck. Any mention in the manual of those DVD decks being SACD capable?

[Kiwick]

What about my 1983 Philips CD100?

Frequency range 20-20,000 Hz +/- 0.3Db
Dynamic range >90Db
S/N ratio >90Db
Channel separation 90Db @ 1000Hz, 86Db across the entire range
Total harmonic distortion at max out level 0.005%
D/A conversion 16 bit/channel with digital filtering

[Byrdsmaniac]

Sorry, I lost track of this thread. Divide your dynamic range figure by six to get the equivalent number of bits. So 90/6= 15. Same for S/N ratio. That's 15 bit performance. Quite decent for the time period.

The .005% distortion figure is also equivalent to 15 bit performance which is totally inaudible even at high frequencies using test tones designed to maximized distortion audibility. Distortion and channel separation are both excellent.

An easy way to remember the dB to percent distortion conversion is this: -100 dB distortion is equal to .001%. Just reverse the digits on the other side of the decimal point. Every change of 20 dBs changes the percentage by a factor of ten so:

120 dB = .0001%
100 dB = .001%
80 dB = .01%
60 dB = .1%
40 dB = 1%
and so on. Again, to convert dB to equivalent number of bits just divide by 6.

[Twenty20Man]

ericson38, you said:
<< Unless I have something wrong here, the #s advertised for the CDX 550 are better than one can have with 16 bit/44.1 KHz sampling you predict is possible. >>

Yes, you are correct. I was just reporting the theoretical limit for 16 bit equivalent performance. 16 bit performance is the limit for the compact discs themselves. Playback devices can always be made better than the media that they are designed to play. What your quoted specs mean is that Yamaha used DACs and other components that are actually specified better than the CD standard calls for. That's not unusual for Yamaha. It's also a good thing because over specified components can lead to better actual measured performance. Your player shouldn't limit the output from your CDs in any audible way.

This does not mean that a 16 bit recording will yield more bits on playback, it only means you are more certain of getting a full 16 bits worth of performance under any circumstances. In actual practice, audio devices often don't live up to their written specs. In addition DACs themselves are specified as typical or worst case. The difference is usually at least a bit. I have observed that Yamaha's players usually measure as good as or better than their written specs. I think you will be hard pressed to find amplification that is cleaner than the output from this player. Only very good speakers are likely to do full justice to the signal quality put out by this unit.

The trivially slight difference between your units in distortion is most likely completely inaudible, but the CDX 920 has a significantly better specified S/N ratio. The CDX 920's 120 dB S/N is equivalent to 20 bits, where the 550's S/N is equivalent to 16.66 bits. I doubt that anything over true 16 bit performance would make an audible difference, but again, these are just specs and not actual performance measurements. I personally feel that lower noise is more important than reducing distortion much below .09%. In any case, it would appear that measuring distortion at only 1 kHz is just a ploy to make the unit stand out on paper. It would seem much more relevant to know the maximum distortion over the entire audible frequency range.


REDone: Unless there is something wrong with your Rotel, I would venture to guess guess that you did not match levels, or try to eliminate any possible influence of bias in your comparison. Level differences are often perceived as quality differences. Even unconscious bias colors everyone's perceptions. It really makes a tremendous difference when you are comparing components that you don't know which you are listening to. Even beer comparisons can come out differently than expected once the tasters don't know what they are drinking. Just my opinion...

would you happen to be able to tell me the numbers for my Yamaha CDC715 i cant find anything on it even on yamahas website, ty

[Byrdsmaniac]

I looked at the Yamaha Japan site. I didn't find any info. Maybe someone here has a manual or something.