Do high end CD players actually make a difference? (1 Viewer)

[Chu Gai]

So it was Bob Harley I saw digging late last night in the cemetary doing that ground breaking research
Lee stated that the better transports reduce jitter to 50 picoseconds or less. Now I'll take that to mean the cumulative effect should result in a jitter of that value. While this is a low number, and Mr. Atkinson has been said to be able to detect jitter of 100 picoseconds or less, this is in rather stark contrast to the information presented in this paper
http://www.nanophon.com/audio/1394_sampling_jitter.pdf
which states that

...practical research that found the lowest level of jitter level at which the jitter made a noticeable difference to be about 10ns rms. This was with a high level test tone at 17kHz. With music none of their subjects found jitter below 20ns rms to be audible. I certainly urge people to read the paper as the pdf is not a very long download.

For more reading consider the following publication:
Benjamin, Eric and Gannon, Benjamin ' Theoretical and Audible Effects of Jitter on Digital Audio Quality,' 105th AES Convention, 1998, Print 4826.

 

[WillieM]

Steve,

Sounds like you have things pretty well in hand. I think it's great that you have the time and resources to build your HT with trial and swap. I'm sure most of us wish we could also.

The only other thing you were commenting on was your speaker sound. The things I would persue in this area are: Real Time Analyzer/ room acoustics/acoustical treatments/ Equalizer.

I guess the problem with room acoustics and treatment is that it's kind of a "voodoo science". A lot of info is available on the Web.

But since you do seem to have some time and resources available, I really think your room acoustics and speakers is where you will now find the most dramatic inprovements possible.

Have fun...good luck, and...(dump the CD player- the 9200 is not "perfect" but it's plenty times better than most folks have for DVD's OR CD's).

 

[BruceD]

I guess the problem with room acoustics and treatment is that it's kind of a "voodoo science".

voodoo? Not at all!!!! Just read "Master Handbook of Acoustics" by F. Alton Everest for a good scientific analysis of room acoustics.

Also, visit the ETFacoustic website for actual demo room graph analysis and what acoustic changes actually look like based on scientific analysis. Anybody can use these inexpensive tools in their own listening room.

 

[Lee Scoggins]

Robert,
The GenX master clock I used in a recent session has claimed around 25-50 picoseconds I believe. I will try to get an official statement. They use Miller Labs jitter analyzers and others that measure jitter in very fine amounts since they provide pro equipment.
I am about to create a low jitter and high jitter CD track for purposes of illuminating these discussions. Perhaps I can send you a disc at some point. The differences are very audible in the studio and on my home system as well as Bob Katz mentions on his web site.
As far as AES reprints, I have know interest in paying these guys for research that quickly goes out of date and in any event the AES considers Bob Katz a leading jitter expert so obviously his peers give him credit there.
In addition the spectrum of jitter is also important which is why Miller Audio and others show a frequency-based plot.
The cleaner plots always belong to the best sounding high end CD players.
Go figure!
You should be happy we found something where metrics do matter.

 

[Lee Scoggins]

Mr. Atkinson has been said to be able to detect jitter of 100 picoseconds or less

Yes, but my money is on John since I know his work in recording and have discussed mics with him.
There is a disagreement between what audiophile or purist engineers hear in terms of jitter and those measured by scientists in some of the older papers.
Recent advances since 1999 in high rez audio are changing the threshold, however, to be lower for audibility.
I am working on getting some more recent studies by pro equipment mfrs for the thread.
Stay tuned, I need to take the dogs out.

 

[Lee Scoggins]

Just read "Master Handbook of Acoustics" by F. Alton Everest for a good scientific analysis of room acoustics.

Chu and I both agree on this: room acoustics are critically important.

 

[Lee Scoggins]

Original estimates of acceptable jitter in A/D and D/A converters were around 100 to 200 picoseconds (pS). However, research into oversampling converters revealed that jitter below 10 pS is highly desirable.

Obviously my friend Bob Katz has the research in hand based on this statement where 10ps is important. I will get his sources and report back tomorrow morning or later.

I should also be able to reach some mfrs tomorrow.

 

[RobertR]

It appears that Mr. Atkinson claims truly prodigious hearing. If we use the following math:

The period of a 4 kHz tone is 250 microseconds, so, expressed as a

percentage, we are talking 10 nanoseconds divided by 250,000

nanoseconds, or 1/25,000. This is 1/250th of a percent, or 0.004

percent. Expressed as a dB ratio, this is 88 dB.

We can construct a table as such:

20173.9794001

500146.0205999

1000140

2000133.9794001

4000127.9588002

8000121.9382003

15000116.4781748

20000113.9794001

At 20Hz, Atkinson would be claiming to hear a signal 174 dB below the fundamental. At 8 kHz, a still astounding 122 dB below the fundamental. At 20 kHz, 114 dB. Truly remarkable.

 

[Saurav]

Yes, except that this is a difference in the time domain, not the amplitude/volume/power domain. You can express any ratio in dB because that's just a mathematical construct, but saying "122dB below the fundamental" is incorrect, because that means something that's 122dB below another in volume. What you really have are two tones that are 122dB apart in frequency, if you can call it that. So, the normal levels of SNR and audibility do not apply here, as those levels are for amplitude/volume dB ratios.
I hope you can follow that, because I didn't do the best job of explaining it

 

[RobertR]

But what matters is the audible manifestation. According to this:

....the overall rms is 108 dB relative to the 1 kHz sine.

and shows the effect graphically.

 

[Craig_Kg]

The audible manifestation of jitter IS describable in amplitude terms.

But that is not the audible effect.

 

[Ian Montgomerie]

Anyway.. thanks for an informative post, that was very interesting. Do you work with digital audio components for a living, by any chance?

LSI Logic Ziva5 (DVD/CD player and A/V receiver chip).

 

[Ian Montgomerie]

Ian, I beg to differ as some cable do exhibit lower jitter characteristics and the receiver-CD interaction of the receiver can benefit from high end players depending on configuration and whether the receiver reclocks all incoming time signals. It depends on the receiver and CD players on a case by case basis. If you eliminate jitter effects, the high end player will also allow better reproduction via use fo better chips, etc.

Exactly how do you determine that some cables exhibit "lower jitter characteristics"? Recall that anything coming off that cable is going through a PLL. And the typical digital cable is a few meters long, with signals travelling at nearly the speed of light in a vacuum. There is extraordinarily little room for the cable to introduce variation between the detected byte arrival times, especially in optical cables which would not be susceptible to electrical interference from whatever industrial strength high frequency emitter happens to be sitting right next to the cable.

And... "whether the receiver reclocks all incoming time signals"? With a normal coax/optical digital cable, ANYTHING that plays audio must generate an internal clock, and use some type of PLL to adjust it to the input. The cable does not carry any useful clock signal for running a DAC.

And, "if you eliminate jitter effects", exactly how will a high end player allow "better reproduction"? The perfect reproduction of what is on the CD is to send out every bit of it exactly and without error. This is something the average consumer CD player does easily. I guess you could declare that running some DSP effect to alter the data is "better reproduction" if it sounds better to you, but in that case why wouldn't you have the receiver do that, rather than the CD player?

 

[mike_decock]

I am about to create a low jitter and high jitter CD track for purposes of illuminating these discussions.

Lee, if you have the means at your disposal to do this, what would be a great experiment is to have a disc with the same song (or snippet) exhibiting various rates of jitter, in random order. Send it out to some people and have them guess which tracks sound bad. Once you have the feedback, you can reveal the jitter values on each track.

Stereophile did something similar on one of their CDs where one track used 16 bit mastering and the other 20 bit mastering (or something like that). Listeners had to guess which was which.

-Mike...

 

[Craig_Kg]

And, "if you eliminate jitter effects", exactly how will a high end player allow "better reproduction"? The perfect reproduction of what is on the CD is to send out every bit of it exactly and without error. This is something the average consumer CD player does easily. I guess you could declare that running some DSP effect to alter the data is "better reproduction" if it sounds better to you, but in that case why wouldn't you have the receiver do that, rather than the CD player?

This sounds easy but it isn't. I used to feel exactly the same until I heard the difference between a very good CD player and a fabulous DAC. Producing the correct outlput levels/word is easy. The timing is the hard part hence all the talk about jitter. Then there is also the quality of the analog stages past the DAC to consider.

 

[RobertR]

Here's a page with jitter samples:
http://www.pcabx.com/technical/jitter_power/index.htm

 

[Saurav]

I hardly think anyone is worried about an increase in noise from -120 to -110 db, especially considering that this is the case with a "huge" value of 2000 picoseconds.

It's a 10dB increase in just the chip's noise floor. No full component has a genuine 120dB noise floor, the analog amplifying/buffering electronics following the digital chip would raise the overall noise floor of the component. If the ADC (or DAC) chip's noise floor gets raised, that additional noise can get amplified by the following electronics (depends on the frequency distribution of the noise and other stuff, I think), which would raise the overall noise floor from 100 to 90, or 90 to 80, or whatever. 2ns is a lot of jitter though.

 

[Dave Moritz]

Hello Steve Paul,

Was just wondering why you would use the dac's in the Toshiba DVD instead of the dac's in the Krell? In a average inexspensive receiver I would use the dac inside the cd and or the dvd player. I have had a few cd players and have owned two different dvd players for my HT system. I currently purchased an open box Sony DVP-S9000ES DVD Player. That I had bought for $799 that replaced my Pioneer DV-525 DVD player. I do not have the level of preamp/receiver that you have. But with my Pioneer I used the dacs in my Yamaha 995 receiver. With the Sony 9000ES I use the dac's in the Sony and not the Yamaha. IMHO I beleive that the higher end CD & DVD players are better due to better transports, build quality, circuitry and d/a converters. I have heard alot of high end players and if one can afford it. IMHO I feel that they are worth every penny as long as you have great speakers and the rest of the components to go with it. I would definently use a seperate CD player for better cd reproduction. I have my 9000ES for DVD/SACD and my Denon 1290 for playing CD's. IMHO you have a great preamp and amp combo and I would look at your DVD player as the culprate for playing CD's. As I am not aware of Toshiba making a audiophile quality unit? Let us know how you solve your delema? Oh what projector are you using? I am looking to get a projector to replace my Toshiba 31" tv.

 

[Saurav]

Beautifully put, Saurav. You don't actually hear jitter as a sound. You notice its effect by the reduction image and soundstage.

Well, I wouldn't know. I've never heard jitter because I've only ever had one CD player that only has an analog output I like to learn about digital audio technology, but don't really enjoy listening to it. At least, not the level of technology I can buy

 

[Manuel Delaflor]

What do you listen Saurav?