or Connect
AVS › AVS Forum › Audio › 2 Channel Audio › USB VS HDMI for 2ch audio to receiver
New Posts  All Forums:Forum Nav:

USB VS HDMI for 2ch audio to receiver - Page 5

post #121 of 584
Quote:
Originally Posted by arnyk View Post
Yet another rookie mistake caued no doubt by a rush to judgement. The plot above is no way representative of the jitter of a real-world AVR. They are never that clean.
True. This was probably done using the AP's own transmitter which is much cleaner than a mass market consumer device. See more below.

Quote:
Your characterization of this being representative of 7 nSec jitter is way off. It represents 100s of nSec of jitter.
I may be way off but thankfully, I didn't create that graph. Julian Dunn did in his excellent tutorial on how to measure digital devices using Audio Precision analyzer. A copy of said graph is in this article which is where I hot-linked: http://www.audioasylum.com/forums/pc...s/3/30104.html



"Figure 1. Periodic jitter of 7ns (Jpp) at 3kHz for an actual 10kHz pure input tone. Sidebands of 7 & 13kHz occur and its distance from the fundamental is given by the jitter frequency (3kHz). Source: Audio Precision, “Measurement Techniques for Digital Audio” by Julian Dunn."

Same thing is in my copy of Julian's book. I have the whole chapter memorized on Jitter as you can probably tell.

Besides, I also showed you the math. We can do it together again since it seems hard for you to figure it out on your own:

db = 20 Log(JitterAmplitude*Frequency*2*Pi/4) = -79.18db. So pretty close to actual measurements.

Plugging in 100ns, the formula yields -56db. Going up to "100s ns" as in 500ns, we get ridiculously poor figure of -42db. Note however that Julian’s formula starts to lose accuracy as J becomes larger so I am not sure these later values are 100% reliable. Likely the number is worse as the modulation index becomes larger. See more below.

Quote:
Slow down Armirm. Check your work! Better yet, get someone who knows what they are doing check your work!
Well, clearly I can't come to you for that! As you have seen from all the typos and missing words and poor grammar, I think faster than I can type! Fortunately I don’t rely on my failing memory for hardcore data like what I quoted.

BTW, if you really want to learn this stuff rather than thinking you know it, our resident DAC expert on WBF has written a series of excellent articles: http://www.whatsbestforum.com/showth...-s-Tech-Series. There, in part 2 of Jitter, you find this simulation he ran:



His 10 KHz line shows even worse figures. I suspect his simulation model is more sophisticated than Julian's as the man designs DACs for crazy RF frequencies. Look also what happens when you get to 1000ns, the number you used in your previous post. See the SNR? Not pretty, eh?

As to some AVRs having good jitter performance, that is true. I showed you that on the first page even. Now you are a smarter shopper, able to decipher data that one in a million audiophiles understand.

Do remember though: all tests are faulty and Paul's is no exception . HDMI's jitter is a function of cable and the source you use to connect to it (see my note above regarding AP’s output). So while Paul's data is good to look at as a first-order estimate, don't take it as absolute. Devices may do worse or better than what he has although probably not hugely so..

While talking about graphs, look at audio asylum guy's measurement of dcs Scarlatti DAC:



Don’t look at the harmonic distortions at the higher frequencies. Pay attention to the two equal sized peaks around the center frequency of 7Khz. Those are jitter since they are symmetrical. They have a frequency of 2 Khz. Likely some clock that runs around inside the box.

What is their level? -122db! If you do the math in reverse, you get peak to peak jitter of just 72ps, using his sound card setup. Money may not buy happiness but it sure buys pretty graphs and measurements!

What did you say that Yamaha jitter sidebands were? -88? Well now you know with more money and effort, you can do better, much better.

So where do we go from here Arny? On every technical statement you have made about digital audio, I have had to correct you. Whether it was HDMI, USB, Ethernet, Jitter, you name it, you came into this thread with incorrect technical knowledge. And I didn't do it with just saying it as you keep doing. But showing external references for it all. Wish you had done the same so that I could learn something in return. As it is, I feel like I did all the work, while you watched .
post #122 of 584
Quote:
Originally Posted by arnyk View Post
The hook in Amirm's anecdote is that contrary to BS 1116, untrained listeners were used.
I have to remember to dismiss any ABX blind report you show me from now on that is devoid of trained listeners . Bookmarked!

Quote:
It is no secret that untrained listeners can perform poorly. His study showed that untrainel listeners were not sensitive listeners.
Nope. I could make the same change I suggested and have the same group vote the opposite way. Or almost the same way.

BTW, ~10% of the general public didn't fall for the trick. Shows you that some people do have good hearing abiltiy even if they are not formally trained. Might they be audiophiles? One would want to hope some are .

Quote:
Amirm then tried to use the results of his non-standard test to criticize the whole idea of using DBTs.
Nope. I am not criticizing DBTs. Again, I used to use them as a powerful tool in development of our products. What I am trying to do is show you the weaknesses of such tests even if textbook methodology is used. All is not what it seems.

This is a great exercise. It will show you the things you need to examine as you read a report. Previous poster came up with a good one. Certainly dumbing down the reproduction chain could dilute detectability. Since you say you are an expert in the field of audio ABX, I expect you to do even better. Indeed I am surprised you didn't.

Note: if you do guess right, my next post will take you where you don't want to go. Don't say I didn't warn you. .
post #123 of 584
Quote:
Originally Posted by bcruiser View Post
The rest quoted below is not doing anything to change the verdict that your post, "It means that to get equal audio performance in an AVR, you may have to spend a lot more money than a dedicated audio DAC. And even then, you may not quite get there.", was just a speculation. What I find disturbing is that you did not say it's your speculation, thus making it easy for misinformation to spread. I hope it wasn't intentional. If it was, that would be even more disturbing.
Disturbing? How? I didn't tell you it takes more money to create a more audible improvement. I said "audio performance." You took that and inserted double-blind test results. Not me.

If I am designing a high-performance AVR and DAC, my first goal is to have them measure the best that they possibly can. That is "performance." In that sense, I can tell you with confidence as a hardware engineer, that extracting the same measured performance costs more money in the AVR envelop than DAC. That is not speculation. It is informed experience.

Take the same DAC IC and drop it in an AVR vs a simple 2-channel DAC and your performance will be lower. All the DSPs, video circuits and power amp are going to stomp on your DAC's performance like there is no tomorrow. Likely I would have to create an isolated power supply for each subsystem so right there, you have extra cost. Next, since jitter and DAC non-linearities can be induced through EMI/RFI, I would need to invest in shielding and physical separation. And instead of just guessing at that, I would need to hire a handful of experts that really understand what is going on there to get it right.

So no, it is not speculation that achieving demonstrable performance parity costs more. Whether you or a bunch of people can hear that improvement is another matter and not one that either one of us can prove as this thread shows. You extended the question in this area and I answered it appropriately. Please don't then mix your argument with mine and then talk about me spreading mis-information. Mis-information would be thinking you can take a delicate DAC circuit and stuff it any place and by magic, it will do what it did on a test lab at the IC design house (similar to what you would have in a stereo DAC).

Knowing that the very next post would be someone calling me wrong and ugly , let me post the answer to that one here and save us some traffic.

Here is a sample TI DAC used in a $6000 processor (Dataset). The performance in TI's lab looks with the chip "naked" (just audio) looks like this:



This is the data published by the manufacturer in the AVS thread we were discussing the very product:



Manufacturer tried to be clever by squashing the vertical scale. But if you squint, you can see that their spurious distortion products are at -90 db relative to TI's -110. So a 20 db drop when the chip is in an AV processor.

Speculation? I think not .
post #124 of 584
BTW, a quick poll. All the posts I am seeing are from folks on the side of the argument. Are there people on the other side who find this information useful? It is taking a lot of time and energy to post these technical notes. I don't want to keep doing it if the only people reading it are the ones rolling their eyes .
post #125 of 584
Quote:
Originally Posted by arnyk View Post
...t. It was a demonstration.
I guess that impresses some as having some meaning.
post #126 of 584
Quote:
Originally Posted by arnyk View Post
...

Those who wish to see some *real world* data for a Yamaha AVR should check out [http://www.milleraudioresearch.com/d...pcm_hdmi).html

....
Unfortunately that seems to require an authentication.
post #127 of 584
Quote:
Originally Posted by amirm View Post
BTW, a quick poll. All the posts I am seeing are from folks on the side of the argument. Are there people on the other side who find this information useful? It is taking a lot of time and energy to post these technical notes. I don't want to keep doing it if the only people reading it are the ones rolling their eyes .
Keep it up. One learns from the exchanges here no matter which side they are on.
post #128 of 584
Quote:
Originally Posted by CharlesJ View Post

Unfortunately that seems to require an authentication.

It's free and fast.

Go to http://www.milleraudioresearch.com/avtech/index.html
Press the register button and provide them with your email and name.
post #129 of 584
Quote:
Originally Posted by amirm View Post

Disturbing? How? I didn't tell you it takes more money to create a more audible improvement. I said "audio performance." You took that and inserted double-blind test results. Not me.

You were talking about the audible sound!
Quote:
Originally Posted by amirm View Post

It means that to get equal audio performance in an AVR, you may have to spend a lot more money than a dedicated audio DAC. And even then, you may not quite get there.

Here is a way to experiment with this idea. Does your AVR have a button to turn off the video circuits and front panel display? If so, hook up a CD/DVD player using S/PDIF coax cable. Play something quiet with lots of ambiance. Now turn up the volume good and loud (or else use headphones). Play it with the video and front panel circuits on (on both the source and AVR) and then turn them all off. Do you hear a difference? If you do, then the above factor is in play. And you now have a free tweak for your system you didn't have to pay for .

Quote:
Originally Posted by amirm View Post

You see that those additional spurs created on either side that was not in the source? Well, depending on where they land, they can have the effect of distorting what was there. They also step on low level detail that is of lower magnitude than them.

to the extent that some of what you consider soundstage comes from reverbs in the original recording and the spurs above step on them, you will also lose a sense of space although I would not call it soundstage widening per-se.


Plus, you are repeating your reply to arnyk and I'll refer you back to this.
Quote:
Originally Posted by arnyk View Post

No, the question was:

"Just curious if there are any differences in SQ between using the USB out on my Macbook and PC, VS using the HDMI out?"

Your answer was:

"Depending on the USB implementation, yes."

The question was not about measured performance, it was about sound quality. However, if we consider only the question for which actual reliable evidence has been presented, the proper answer would be:

"The answer depends on the implementation of the conversion from USB to analog audio signal, and the implementation of the conversion from HDMI to analog audio signal."

Your surreptitious changing of the question that the OP asked is yet another distraction.


...




amirm, you are moving the goalposts around to fit your agenda. This kind of act does not create meaningful discussion. Adios.
post #130 of 584
While we don’t think them as such, I think these debates are like mock trials where there is a prosecutor and defending attorney trying to prove their respective points of view. As with a real trial, it is useful to have a summary statement at the end as to make sure each party’s position is not lost in the all the back and forth. So I am going to do that now and unless there are substantial new points to be made, call it done.

This debate started with Arny calling me “presumptuous” as I declared USB as the way to go in interconnecting audio systems to your computer. Question is, did he succeed in demonstrating that? For that answer we need to think through the architecture of the systems we use. When playing music, we know that we can distill the content into files. We can copy those files on our computers with reckless abandon and even send them across the world over the unreliable and non-real-time Internet and still, unless something goes wrong, get the original back intact. Digital audio in that sense is perfect!

Sadly, when our current interfaces were designed, they were not architected the way they should have been. We take digital data that is nicely marked with its timing and data in a computer file and turn it into a real-time stream across a cable. The source is the “master” telling the destination when to play each and every sample. On paper, the timing is perfect vertical pulses that instantly go from zero to one, and with zero variation. Sampling theory says if we did that, indeed we can have perfect reproduction. That is, if we reproduce the samples at the same time they were captured (digitized), we can reconstruct the signal perfectly.

Alas, real world doesn’t work that way. The above definition of our timing signal is that of a square wave. The wiki on square wave says it nicely: “An ideal square wave requires that the signal changes from the high to the low state cleanly and instantaneously. This is impossible to achieve in real-world systems, as it would require infinite bandwidth.” Let me repeat: you need infinite bandwidth. No cable or interface has infinite bandwidth. So we know what gets to the other side has less than perfect edges. As soon as we modify those waveforms, we also start to mess with the timing that can be extracted at the receiver. Yup, horrors of horrors. Your digital cables can have a sound!

We get confused looking at the low-speed audio signals thinking not much accuracy is needed to represent their timing. Arny made that mistake of thinking 1 microsecond of timing should still be great. That is one millionth of a second. By a person not schooled in the science of digital audio and signal processing, that does look like the right metric relative to CD’s 44,100 samples per second with each sample taking 22 microseconds.

The science unfortunately is not that forgiving. It doesn’t just look at the sampling rate of the audio but more importantly, how much resolution and how low of a noise floor you want to have. We love our digital systems because they are so quiet. A 16-bit system has a range of quietest to loudest signal of 96 decibels. That is the lowest level digital system we use for high-fidelity music. 30+ years after introduction of CD, it sure would be nice to be able to achieve what its specs say on paper, in a real system. Don't you think?

Given the above, and making some significant simplifications, we can compute how much timing change it takes for one of those bits to get corrupted. With each bit representing 6 db of distortion products, the math can be computed as I showed from Julian’s formula. That math says that timing accuracy better be 500 picoseconds or else your system has less resolution than an ideal 16-bit system. A picosecond is one million microseconds! Wow oh wow!!! What have we gotten ourselves into? We need accuracy that is 2000 times smaller than Arny’s one microsecond number.

While not a topic here, but a point I made, the receiver cannot throw out timing variations. Why? Because it really doesn’t know how fast the source is going to send it data. Indeed, sampling rates like 44,100 are “nominal” values. The receiver cannot use them as the source of timing. It is entirely legal and indeed happens all the time that the transmitter will run slower or faster than that rate. The standard allows +-5% variation. So the receiver is tasked with the tough job of throwing out some variations but not others.

Smart designers over the years have figured out good ways to deal with that in S/PDIF domain. In HDMI, they are kind of stuck with off-the-shelf silicon which is first designed for video, and secondary for audio. Clock recovery being a “mixed signal type of problem” involving both analog and digital, means that there are far more engineers who get it wrong than right. I know. I have had the unhappy fortune of having some of those engineers work for me, nearly destroying hardware products we built for major television networks which could not properly extract said signals. Every engineer is taught about “PLL” design in school but the reality is very different in real world than in a textbook.

While we have focused on interface jitter in this topic, that is not the only area of problems. DAC performance can be impacted in a number of other ways, given the delicate signals it is trying to reproduce. Take the voltage of an AA battery and divide by 65536 for a 16 bit system, which is ~35 millionth of a volt. Heaven help you if you try to reproduce 24 bits because then you divide by 16 million!

Digital audio reproduction is therefore highly complex and difficult. With mass market consumers being so price conscious, a typical engineer working for on a mass market design, is not going to try to be heroic. He has to hit severe price points so he is going to go for what you can see: the list of logos on the box, wattage numbers and such. That is the priority. Not getting that last bit of 16 bit audio sample accurately. He is doing his job, making sure he can put food on the table and keep his company in business.

Enter high-end companies. With cost shackles removed, they can go as far as they dare to go. Now, just as you can’t become a better artist with a more expensive brush, there is no guarantee that because you don’t have cost constraints, you are producing great products. I recently reviewed a $16,000 DAC+amp combo and found its sound anything but refined. But companies like Harman (Mark Levinson, Revel, JBL, Crown, Lexicon) that base their design not just on some gray hair engineer’s idea of good sound, but couple it with careful listening tests and measurements do put the dollars to good use.

This nicely segues into the next topic: blind listening tests to prove audibility of such artifacts. Unfortunately, timing problems in digital interfaces do not lend themselves well to controlled experiments. For one thing, there is infinite variety in jitter. It can be random, periodical, data dependent, or discontinuous. And all combined in different levels. Where do you start? Well, folks started with random – the worst kind to go after. Why? Because random jitter just adds noise to the system and in that sense, it is least audible. See my debate with Ethan on more. I post the spectrum of the dcs DAC where you see it had jitter at 2KHz. That was not random at all.

Worse yet in my opinion is selection of material. I don’t know why people think “audiophile” music is the right content. We are not trying to enjoy music in these tests. We are trying to instrument a system with our ears. This brings me to the answer to the study that found 90% of the people could not tell the difference between CD and 64 Kbps version. At this rate, 95% of the original file was thrown way yet folks thought nothing had happened to it. Music codecs are good but not that good! Answer was the selection of music. The test agency thought that they should pick what audiophiles might listen to and naturally went for classical music and such. Well, classical music is harmonic and perceptual compression systems do wonderfully there. Where they get in trouble is when you have sharp transitions such as guitar strings, voices, etc. Even there, you need some quiet around it so that you can hear the so called quantization noise. Any one of my own stash of “codec buster” tracks would have blown the door open letting people hear the difference far, far easier. But that was not picked and from then on, the test was doomed despite perfect methodology otherwise.

So it is not enough to say this test was blind this, and ABX that, and run off with their conclusions. You need to first prove, as I just did with the science of compression that such content was going to be revealing of the problem we are chasing. MPEG put together its suite of audio tests we use to evaluate audio compression. None are audiophile music by any stretch. But they are very revealing as they must be. Where is the similar set of test files for jitter? Or frankly, for all the ills of digital audio? They don’t exist. People use random selection of music and then wonder why their outcome is close to random. Well duh!

I have been thrown at these tests before. Lack of good content makes the job very hard. You are under the stress of answering an AB question and you shouldn’t have to squint to read the tiny differences. Magnify them for me. Don’t tie my hand behind my back and expect me to perform miracle. My ear is not an instrument and there is a limit to my patience. Don’t push me to vote randomly and dilute the overall results that way. If you give the right track and I still couldn’t tell, then I will live with the results.

Given the paucity of data in this space then, what should we do? One answer is to put one’s head in the sand and say it is all good. Well, don’t you want to be sure? Isn’t that why you spend so much time here? One way to get there, at least partially, is to look at measurements. Within bounds, they are pretty reliable metrics of the quality that went into a design. I like to shoot for 16 bits of performance. If the system does that truthfully, I feel good. 20 and 24? The former is heroic, the latter impossible.

To be sure, Arny’s flag is super tempting to follow. Wouldn’t it be good if all equipment is cheap and I can just go by linear specs like power and number of logos as I mentioned? Sure. I won’t deny that as I used to do the same, making fun of all my audiophile friends. But then I took the first step going beyond textbook theory and into the real world of building such products and testing people left and right on all of this. Learned the value of auditory training in this space and seeing that occasional person walk off the street and beating me at that game! To be sure, most audiophiles are quite bad in hearing such artifacts but not all.

Hopefully I have demonstrated in this thread how deep this rabbit hole really is. And that the opposing view is a tough place to stand in the absolute. The math, the numbers, and the graphs are ruthless and powerful in the way they convey their message.

You do not have to take my position by the way. All I want you to do is be more informed. Understand the complexity of topic and don’t let one liners thrown out by the “it all sounds the same” as what guides you. Use this thread as a primer to learn more. Digital audio is not intuitive to any of us as this thread hopefully shows. You want to follow science, do it right.

On a personal note, I have grown to like Arny in this thread. Don’t ask why but I find him likable. Maybe because he is an old analog hack like I am. Or maybe because he makes it possible for me to answer his challenges with smile and excitement. Wish I didn’t have to take him on as I did. Alas, you can’t lead an army for a cause if you don’t understand the cause itself. And the cause is complex here. It doesn’t lend itself to one-liners that he came into this thread with and damning the future of where we need to go which is the target device being in charge of reproduction over a data bus like USB/network.

Can I blame him for not knowing all that is needed to know? No. As shown, the topic requires wide-ranging knowledge across an incredible array of topics from math to audio and computers. I have been fortunate that my employers have paid me to learn this stuff over the last 30 years, augmented by interacting with fine folks in this forum and elsewhere.

Don’t take the above as I know it all. I don’t. There are layers of complexity here and you can only hope to peel back some of them.

So how was this for a Sunday sermon?
post #131 of 584
Quote:
Originally Posted by bcruiser View Post

amirm, you are moving the goalposts around to fit your agenda. This kind of act does not create meaningful discussion. Adios.

I love it when masters of Calvinball accuse me of the same:

Calvin: "Dad, are you vicariously living through me in the hope that my accomplishments will validate your mediocre life and in some way compensate for all of the opportunities you botched?"
Dad: "If I were, you can bet I'd be re-evaluating my strategy."
Calvin to his mom, later: "Mom, Dad keeps insulting me."

. Remember, it is just a forum discussion. It should be fun first, everything else second.....
post #132 of 584
Quote:
Originally Posted by amirm View Post

That math says that timing accuracy better be 500 picoseconds or else your system has less resolution than an ideal 16-bit system.

So for playing back redbook CD, all we need is to have jitter be 500 picoseconds or less for full resolution. Seems like a pretty low bar.
post #133 of 584
Quote:
Originally Posted by amirm View Post

So how was this for a Sunday sermon?

When can we expect the aircraft with leaflets to be flying overhead?
post #134 of 584
Quote:
Originally Posted by CharlesJ View Post

Quote:
Originally Posted by amirm View Post

If "good" amps all sound the same, did they not use good amps in their ABX testing? If so, how come?

They used 5 trials???? That should impress? Wasn't that an en-mass test of dubious quality at best?

The blind amplifier took place at the 1988 AES Convention in Los Angeles and in order to allow the maximum number of listeners to take part, each individual was restricted to 5 trials. As was mentioned, Michael Fremer got 5 identifications correct and I got 4 out of 5. Neither score reaches the 95% confidence limit and both Michael and I requested to take the test a second time, to see if we could repeat the score and thus increase the statistical probability that we were not just "lucky coins," but that wasn't possible.

A cynical person might suspect that the limit of 5 trials per person was decided upon to avoid a perfect 5/5 score being used as evidence for there being audible differences between the amplifiers chosen for the test. :-)

Regarding Arnyk's claim that this 1998 test was not a test but a "dem," the report on it in the Journal of the AES did describe it as a listening _test_.

And Amirm, thank you for the commentary in this thread.

John Atkinson
Editor, Stereophile
post #135 of 584
Quote:
Originally Posted by amirm View Post

While we have focused on interface jitter in this topic, that is not the only area of problems. DAC performance can be impacted in a number of other ways, given the delicate signals it is trying to reproduce. Take the voltage of an AA battery and divide by 65536 for a 16 bit system, which is ~35 millionth of a volt. Heaven help you if you try to reproduce 24 bits because then you divide by 16 million!

Digital audio reproduction is therefore highly complex and difficult. With mass market consumers being so price conscious, a typical engineer working for on a mass market design, is not going to try to be heroic. He has to hit severe price points so he is going to go for what you can see: the list of logos on the box, wattage numbers and such. That is the priority. Not getting that last bit of 16 bit audio sample accurately. He is doing his job, making sure he can put food on the table and keep his company in business.

Enter high-end companies. With cost shackles removed, they can go as far as they dare to go.

Since you mentioned DAC linearity issues, it's fair to point out that these issues are internal to the DAC chips, and as such, high-end audio companies have had negligible contributions in this area. These innovations have been developed almost solely by semiconductor companies, using such techniques as dynamic element matching (see this Ph.D dissertation PDF for example). This kind of research is way beyond what the vast majority of high-end audio companies are capable of. The myth of the genius tinkerer is rarely true in practice. The most prevalent type of research I've noticed in the high-end audio business is related to Optimal Wallet Extraction Theory.

Quote:
Originally Posted by amirm View Post

This nicely segues into the next topic: blind listening tests to prove audibility of such artifacts.

Quote:
Originally Posted by amirm View Post

I have been thrown at these tests before. Lack of good content makes the job very hard. You are under the stress of answering an AB question and you shouldn’t have to squint to read the tiny differences. Magnify them for me. Don’t tie my hand behind my back and expect me to perform miracle. My ear is not an instrument and there is a limit to my patience. Don’t push me to vote randomly and dilute the overall results that way. If you give the right track and I still couldn’t tell, then I will live with the results.

What's been going on for years though, is that audiophiles make claims about the audibility of things that are said to not be subtle, and if you don't hear what they hear, you are considered deaf and/or your system sucks ("not resolving enough" is the standard phrase here). Yet when asked to verify the claim under controlled conditions and subsequently failing to do so, the excuse factory goes into high-rate production: the stress of forced choice suddenly prevents them from hearing this heretofore obvious effect, somebody implemented such a test poorly in 1988, so all such tests must be invalid, yada, yada. It's the same old same old, and quite tiresome.

Quote:
Originally Posted by amirm View Post

Hopefully I have demonstrated in this thread how deep this rabbit hole really is. And that the opposing view is a tough place to stand in the absolute. The math, the numbers, and the graphs are ruthless and powerful in the way they convey their message.

Of what importance are the math, the numbers, the graphs and the measurements to the subjectivist? They are only important to the extent they can be used to justify immutable audiophile dogma. Is distortion important? If a capacitor can be shown to have 1 ppm of distortion, then yes, because that can be used to support the "capacitor sound" dogma. But if a SET power amp has 10 percent distortion, then distortion doesn't matter.

Also, good jitter measurements are not confined to the realm of the high end, so it's not necessary to pay the big bucks to get there. This $350 CD player has jitter below the residual of the Miller jitter analyzer. Getting performance that good for $350 is just outstanding engineering.
post #136 of 584
Quote:
Originally Posted by stereoeditor; View Post


Regarding Arnyk's claim that this 1998 test was not a test but a "dem," the report on it in the Journal of the AES did describe it as a listening _test_.

That was probably part of a casual conference report, and did not appear in a a peer-reviewed paper.

I thihk a little common sense needs to be applied here. 4-5 trials is not enough trials to obtain reliable statistics. Since we've got a JAES peer-reviewed article that says that reliable statistical analysis is an important component part of ABX testing, any event that doesn't include enough trials for reliable statistical analysis can't be a proper ABX test.

Here is a more likely explanation of what happened, because this sort of thing has happened over and over again in the past. Yes this is speculation, but when you see a certain kind of train wreck several dozen times, you kinda sorta almost expect it, or are not surprized at all when it happens again..

Prideful golden ears have been telling themselves for decades that the scientific types can't get reliable results in DBTs because the scientific types have cloth ears. Obviously, not having cloth ears, the golden ears fully expect to get statistically significant results without a lot of trouble. In fact Atkinson and Fremer had to work a lot harder than they expected to get what they got in their first few trials. They realized that discression is the better part of valor and tried to save face while making a hasty retreat,

Again the above is speculation, but it is highly informed speculation.
post #137 of 584
Quote:
Originally Posted by amirm View Post




"Figure 1. Periodic jitter of 7ns (Jpp) at 3kHz for an actual 10kHz pure input tone. Sidebands of 7 & 13kHz occur and its distance from the fundamental is given by the jitter frequency (3kHz). Source: Audio Precision, Measurement Techniques for Digital Audio by Julian Dunn."


db = 20 Log(JitterAmplitude*Frequency*2*Pi/4) = -79.18db. So pretty close to actual measurements.

Plugging in 100ns, the formula yields -56db. Going up to "100s ns" as in 500ns, we get ridiculously poor figure of -42db. Note however that Julian's formula starts to lose accuracy as J becomes larger so I am not sure these later values are 100% reliable.

I hate to see you twist in the wind Amirm, so I'll explain your little jitter mystery for you.

If you'd look at the test results I provided a link to over at Miller Research and actually understand the results they show there, there would be no mystery and my comments would make perfect sense.

The problem with the Figure 1 jitter plot is that it shows only one pair of sidebands due to jitter. If you look at the HDMI jitter test results for AV receivers over at Miller Research, you'd see that they show many, many pairs of sidebands due to jitter. This is nature's way of telling you that the modulating waveform for their jitter is not a pure since wave (which is what Dunn's figure 1 shows) but is rather something far more complex, perhaps a square or triangle wave with lots of harmonics. Each harmonic generates a pair of sidebands.

Because there are so many pairs of sidebands, their height need not be very high in order for the energy in them to sum up to a large number of ps.

It's a simple as that. Amirm, you've made yet another rookie mistake because you are apparently unaware of how to properly interpret the real-world jitter plots over at Miller Research. You have been pursuing the overly simplified idea that they can be explained using the simple calculations from the Dunn paper, when you need to add another level of complexity and *sum* the energy in the sidebands when there are more than one pair of them.

Psychoacoustically, there is a big difference between the audibility of a lone pair of sidebands and the kind of forest of them that we often see. Since they are just about all inside each other's critical band they tend to mask each other. All you actually hear the loudest pair in each critical band.

Therefore looking at simple sums of jitter makes no sense when the jitter is not due to pure sine waves but is rather due to complex waves such a square waves or triangle waves.
post #138 of 584
Quote:
Originally Posted by rock_bottom View Post

Since you mentioned DAC linearity issues, it's fair to point out that these issues are internal to the DAC chips, and as such, high-end audio companies have had negligible contributions in this area.

Unless the DAC is synthesizing its own power and clock, and is shielded capacitively, magnetically and otherwise from the rest of the world, this is not true. I showed you clear example of performance dropping 20 db when said DAC gone from workbench to inside of a high-end, $6000 processor: http://www.avsforum.com/avs-vb/showp...&postcount=123

I really hope we don't have to rehash topics already covered.

Quote:


These innovations have been developed almost solely by semiconductor companies, using such techniques as dynamic element matching

There are noted exceptions such as dcs and its ring dac: http://www.dcsltd.co.uk/page/oursecret:

"...we do not use commercial off-the-shelf DAC chips."

In addition, many companies add external upsampling or filtering. These are some of the secret sauce for Berkeley Alpha DAC: http://www.berkeleyaudiodesign.com/products.html

This on top of all the other careful design techniques mentioned about your first point.

Quote:


The myth of the genius tinkerer is rarely true in practice.

Well, don't take this personally but if you are not aware of their work, it is not an indication that it doesn't exist. Yes, IC DACs are used. But no, that doesn't mean they are all the same.

Quote:


What's been going on for years though, is that audiophiles make claims about the audibility of things that are claimed to not be subtle, and if you don't hear what they hear, you are considered deaf and/or your system sucks ("not resolving enough" is the standard phrase here)...

That's neither, nor there. I am not the one with those claims. And further, I have never understood the angst that this brings to the other party. How do you know it doesn't come across to them differently than it would to you?

I have been going to Japan for 20 years now. Over the years, I have gotten to a level where I can taste and appreciate very subtle things about Japanese food. I can instantly tell when frozen fish is used for sushi and can't stand it (other than tuna). I can tell when the fish used is in season vs not. I can tell the difference between tofu made that day and special water used to make it. I am not a snob about this but I have developed a level of taste for these ingredients that is quite foreign, literally, to the uninitiated. It took me years, from hating all of this to liking it and then appreciating them so much when the quality is great as it can be some times.

Is your position that if the general public can't tell the difference between canned tuna and fresh caught fish that is still jumping on your plate, I should not describe the difference as astonishing? Now we need limits to expression of differences?

Why? This is a hobby and no more important than me eating and appreciating sushi. No one is selling you their excitement. Why be bothered so much about it? Why, oh why? It is not logical. They are doing their thing with their money. If you don't want to be called deaf, then don't engage in the conversation! Now, if they say something that is not technically correct, sure, let's correct them and do it calmly and unemotionally.

Quote:


Of what importance are the math, the numbers, the graphs and the measurements to the subjectivist?

Before we go there, how about the other camp? It is important to show that objectivists don't understand these things so the firm ground they think they are standing on, is quicksand. You want to know why I get engaged in these conversations? This is it. In the name of science, they throw out information that is simply wrong.

I show the same thing to subjectivists. I see them appreciating it. They see that they not being called names and they listen. Look at the comments from subjectivists in the WBF tech library. No venom being spat people like me who are posting scientific data. Now look at this thread and the attitude from people who are supposed to be in favor of information. I don't know that there is another active thread that has as much engineering and science information in it in all of AVS but what was the reaction? A bunch of name calling. Clean up your own house for heaven's sake before complaining about the other side! .

Quote:


They are only important to the extent they can be used to justify immutable audiophile dogma.

We are all dogmatic. Let's not kid ourselves. Everyone sticks their fingers in their ears and says the other guy is wrong. Look at your flag carrier: Arny. It took 14 tries and I still don't know if he knows what jitter really is and what it looks like given his last back and forth with me. And yet he keeps saying I don't know what I am talking about. How much clearer does it get? If your flag bearers look like this, what hope there is for good dialog?

WBF was created by the way because so many people got tired of not being to have a dialog without getting beat up. Almost all the active posters used to be here and they have all left. I am one of a handful who still frequents this forum and frankly, regrets it most of the time. You all make it so unpleasant to have a conversation. Why do you act this way? Were you smarter without this thread? I would think not.

Quote:


But if a SET power amp has 10 percent distortion, then distortion doesn't matter.

Another great topic I wish I had time and inclination to participate in but frankly have no desire. For now, maybe you will respect the work of a senior scientist who is 100% in your camp, owns a $500 "chip amp" AVR. Yet has some interesting views on how we may be measuring distortion incorrectly. Read this by Dr. Geddes and then create a new thread on WBF and we can discuss it: http://hephaestusaudio.com/media/200...ion_aes_ii.pdf

I am in dire need of someone putting in a fork me as I am all done.
post #139 of 584
I don't know why we are even looking at a graph of 7ns (7000ps) of jitter. Any decently designed CD player or DAC is going to have less than 200ps of jitter, and the best have less than 20ps. I'd like to see the graph re-run at 200ps, and then let's have a conversation. I'm sure there would be a lot less hand-wringing.
post #140 of 584
Quote:
Originally Posted by arnyk View Post

Quote:
Originally Posted by stereoeditor View Post

Regarding Arnyk's claim that this 1998 test was not a test but a "dem," the report on it in the Journal of the AES did describe it as a listening _test_

That was probably part of a casual conference report, and did not appear in a a peer-reviewed paper.

It was not a peer-reviewed paper but as it was published in the JAES, was hardly "casual," I would have thought.

Quote:


I [think] a little common sense needs to be applied here. 4-5 trials is not enough trials to obtain reliable statistics.

That was my point, at least as it concerned any single listener. The organizers of the test, whom I believe included your associate David Clark, wanted to examine the results of _all_ the listeners as a whole and therefore restricted the number of trials per session to 5, in order to maximize the number of participants. As as Michael Fremer described in the quote provided by Amirm, those results were null.

Quote:


Here is a more likely explanation of what happened. . . Atkinson and Fremer had to work a lot harder than they expected to get what they got in their first few trials. They realized that [discretion] is the better part of valor and tried to save face while making a hasty retreat.

That is incorrect. _All_ the listeners, including Michael and myself, were subjected to 5 trials. Michael scored 5/5 identifications; I scored 4/5 correct. It did not prove possible for us to retake another set of 5 trials to increase the statistical power for each of us considered on our own. So it goes.

John Atkinson
Editor, Stereophile
post #141 of 584
Quote:
Originally Posted by amirm View Post

Sadly, when our current interfaces were designed, they were not architected the way they should have been. We take digital data that is nicely marked with its timing and data in a computer file and turn it into a real-time stream across a cable. The source is the “master” telling the destination when to play each and every sample. On paper, the timing is perfect vertical pulses that instantly go from zero to one, and with zero variation. Sampling theory says if we did that, indeed we can have perfect reproduction. That is, if we reproduce the samples at the same time they were captured (digitized), we can reconstruct the signal perfectly.

Amirm shows us his inexperience with real-world digtial once again. If you get a high-bandwidth scope and probe around some SP/DIF cables. Pretty good chance that you won't see anything like near perfect square waves because as a rule SP/DIF outputs pass through low pass filters that are added to meet FCC Part 15 rules for emission of EMI. Sometimes what you see looks a lot like a sine wave.

An even more impessive situation is the wave that comes off of the laser pickup of an optical disc player. The jitter is so great that you can often see it quite easily if you pick the right kind of data coming off of the CD.

So, Amirm's statement that " timing is perfect vertical pulses that instantly go from zero to one, and with zero variation" describes a theoretical world that never happens and in fact what happens in the real world is almost the exact opposite. Yet, we've had optical disc players with low jitter for decades. It took longer to get the kinks and bends out of signals on SPDIF cables, but that has been pretty much a solved problem for about 10 years.

One pf the ironies of high end audio is that high end CD players that used SP/DIF to send signals to external DACs were far more likely to have jitter problems than CD players with the DACs inside the same box as the transport. This was particularly true from the middle 80s to nearly the end of the 90s.

There's nothing wrong with putting DACs inside a CD player box. Both external DACs and CD players have digital busses with sharp square waves on them. Remember these are inside the box and are not as exposed to regulatory scrutiny.
post #142 of 584
Quote:
Originally Posted by stereoeditor View Post

_All_ the listeners, including Michael and myself, were subjected to 5 trials. Michael scored 5/5 identifications; I scored 4/5 correct. It did not prove possible for us to retake another set of 5 trials to increase the statistical power for each of us considered on our own. So it goes.

I wonder what kind of statistics training Physics majors get in the UK, An evaluation with only 5 trials should scream "not a valid test" to anybody who has even just taken first year statistics.

Fremer seems to lack formal training and so his confusion does not surprise me. OTOH, who do we know who as a degree in Physics? ;-)
post #143 of 584
Quote:
Originally Posted by arnyk View Post

Amirm shows us his inexperience with real-world digtial once again. If you get a high-bandwidth scope and probe around some SP/DIF cables. Pretty good chance that you won't see anything like near perfect square waves because as a rule SP/DIF outputs pass through low pass filters that are added to meet FCC Part 15 rules for emission of EMI. Sometimes what you see looks a lot like a sine wave.

You are out of your own arguments are now manufacturing some for me it seems . You are actually right for a change. Look at a sample of scope outputs this guy captured from S/PDIF output of different CD players:



Second, I would never say it looks like squarewave. Indeed, the fact that it does not, causes the zero-crossing point to change and with it, cause jitter. This is called "cable induced jitter." From AES recommendation doc on interface requirements:



How much more of this back and forth do we need to go through?
post #144 of 584
Quote:
Originally Posted by amirm View Post


Another great topic I wish I had time and inclination to participate in but frankly have no desire. For now, maybe you will respect the work of a senior scientist who is 100% in your camp, owns a $500 "chip amp" AVR. Yet has some interesting views on how we may be measuring distortion incorrectly. Read this by Dr. Geddes and then create a new thread on WBF and we can discuss it: http://hephaestusaudio.com/media/200...ion_aes_ii.pdf

This is an unecessary distraction. It needs to be dropped quickly FWIW Dr. Earl Geddes has devoted most of his his life to loudspeakers. The paper makes sense in the context of loudspeakers which very often have audible distortion. It is irrelevant to amplifiers and DACs because so many of them are sonically transparent.

Quote:


I am in dire need of someone putting in a fork me as I am all done.

Yes you are Amirm, and not in a good way! This last attempt at drawing a red herring through the discsuion is what seems to be the failing gasp of someone who has been proven to be wrong at so many points in this discussion.
post #145 of 584
Quote:
Originally Posted by arnyk View Post

I wonder what kind of statistics training Physics majors get in the UK, An evaluation with only 5 trials should scream "not a valid test" to anybody who has even just taken first year statistics.

Fremer seems to lack formal training and so his confusion does not surprise me. OTOH, who do we know who as a degree in Physics? ;-)

Question: can you get 5/5 right?
post #146 of 584
Quote:
Originally Posted by amirm View Post

You are out of your own arguments are now manufacturing some for me it seems . You are actually right for a change. Look at a sample of scope outputs this guy captured from S/PDIF output of different CD players:



I see no evidence that these waves came from actual production equipment that is being sold on the open market and meets FCC rules.

I suspect that you are confusing lab results with what happens in the field. If you are, its not the first time you've been caught making this kind of mistake.

Besides, they aren't anyything like square waves and are very sinusoidal so you are just helping me prove my point.
post #147 of 584
The examples of cable-induced, pattern-dependent jitter show what happens when you use analog cables to transmit digital signals. I'm not sure why that's relevant to the discussion.
post #148 of 584
Quote:
Originally Posted by amirm View Post

Question: can you get 5/5 right?

Sure. Even 16/16. It all depnds on what is being evaluated.

Hint: restrict yourself to things that are known to be audible, and you will be a happy ABXer.

And this is why golden ears are so frustrated with ABX tests. Because they have no clue about qunatification and psychoacoustics, they don't realize that they are trying to do something that is comparable to running a 2 minute mile while still in diapers.
post #149 of 584
Quote:
Originally Posted by audiophilesavant View Post

I don't know why we are even looking at a graph of 7ns (7000ps) of jitter. Any decently designed CD player or DAC is going to have less than 200ps of jitter, and the best have less than 20ps. I'd like to see the graph re-run at 200ps, and then let's have a conversation. I'm sure there would be a lot less hand-wringing.

Read the first page:

1. OP wants to know which computer interface to use. Fidelity of digital output on a PC is all over the map and no where close to 20ps. I don't think anyone cares this day and age about stand-alone CD players. At least many of us don't.

2. OP asked about HDMI. I showed data from Miller Research that HDMI does far worse.

The graphs were used because that is what is what I can find to post. There was no magic to 7ns number other than Arny claiming it was for hundreds of nanoseconds.

I have already said that once you achieve 500ps, you are golden for 16-bit audio. So not sure why you are asking for a graph. Problem has been is that for a number of people in the thread, the whole subject seems to have been new so we used a set of examples to discuss it.
post #150 of 584
Quote:
Originally Posted by arnyk View Post

Sure. Even 16/16. It all depnds on what is being evaluated.

I will ask again. If you were in a blind amp test as they were, can you get 5/5 right?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: 2 Channel Audio
AVS › AVS Forum › Audio › 2 Channel Audio › USB VS HDMI for 2ch audio to receiver