USB VS HDMI for 2ch audio to receiver - Page 15

What's especially funny about JA's quote there is that audiophiles are notorious for being perpetually UNsatisfied with their systems, always nagged by the idea of something better out there, always looking for the next bit of gear or the next tweak that will bring them (asymptotically) towards nirvana. For some, the perpetual quest appears to be the hobby.

But then again, no one has claimed audio cables should behave the same into all loads....



We can always detect (measure) 'something'. The question for audio, is whether that something is likely to be audible, and under what conditions -- and how to detect THAT. Measurements , plus past research on hearing, can give us an indicator as to whether some measured difference is LIKELY to be audible. Conversely, positive listening test results indicate we SHOULD find something measurably and nontrivially different. If we don't, or if we only find very tiny differences, it's time to start experimenting.

(FWIW, 0.1dB difference at 20kHz would fall into the 'trivial' category, IMO)

That's a very Faustian way of looking at things. Goethe would've understood.

I didn't say economies of scale DETERMINES the amount of audio quality for any given device....I mean it has facilitated high-quality audio in mass market devices. Now , to me, something that 'sounds very good' qualifies as 'high quality' in audio terms -- even if its jitter measurements look like a porcupine. To you, that can't be 'high quality', even though we haven't even got a handle on whether that porcupine graph is likely to make an audible difference.




Beats me! You are the one claiming audibility, not I. So, tell me again you know there was an audible difference. Conditions of the blind test, please, and stats of the results.




How many complaining that is it is 'too green' is not the question. The question is how many can tell there's a difference. But maybe I made too many assumptions. I assumed first you and others have actually reliably detected that difference in controlled tests. Please describe them if so. And can you say what magnitude and spectrum of audio jitter a 5% green shift is analogous to?

I assume you are referring to Keith Howard's article in Stereophile. John Marks and I have also discussed the issue in Stereophile.


Who says objectivists don't have a sense of humor!?! You throw out an unsubstantiated and unsupported allegation that Stereophile's reviewers have incorrectly praised Red Book-quality recordings as hi-rez; I ask you for an example; and you want _me_ to do the legwork to substantiate your argument. Ridiculous.

When you wrote "I myself have personally documented DVD-A and SACD releases that have less dynamic range than a well-mastered CD..." I erroneously assumed you would have remembered which recordings these were.


Forgive me being literal, Mr. Sullivan, but on all four of the forums on which this speech of mine was being discussed, it was _you_ or Arny Krueger who raised the subject, not me. As you did here. It is hardly appropriate, therefore, for you subsequently to refer to _me_ as "not being shy" about this matter. A less kind person than I happen to be might think you were descending to the level of an Internet troll.

John Atkinson
Editor, Stereophile

So that's what you should do then. It's easy to big-mouth trash-talk "bad science" and heck, there's plenty out there, for sure. However, unless you prove them wrong on their own turf with a *flawless* paper, you're just as bad, if not worse as "them". You'd just be playing the FUD game instead of measure up as a man, do the hard work of proving them wrong or right on their own turf. No?

There is that and then there is this:
Something I have repeated often.

To a music lover (like myself), the stereo is a means of hearing, listening to and enjoying the music.

Whereas to an Audiophile, the music is a means of hearing, listening to and enjoying his stereo.

As you pointed out, the enjoying part might be debatable.

I may have some audiophilic tendencies myself, as a speaker builder, always nagged by the idea of some better transducer out there, believing that we may indeed be making real (soundwave) progress since the days of Western Electric....though audiophile might debate me on that as well.

cheers,

AJ

I keep having flashbacks of John Risch's "dielectric hearing" listening tests, Dr "Darque Knight" Smith et al Audiophile Fuse "listening tests", etc.....all done under strict scientific controls, yielding positive results....and then immediately stamped "Top Secret" and locked in a vault, deep underground somewhere.
So tantalizingly close...and yet oh so far away...the Holy Grail.

cheers,

AJ

There was a bad link in his original post on the subject ("lookee here") that was later fixed in a repost.

Keith Howard's article it was. Very enjoyable.


No, I said I'd do the data mining -- the work -- but I do need the data file to mine because, contrary to what you might imagine, I don't keep a running file on the silliness your team spouts. Btw, I can't see there being much 'legwork' involved, haven't you digitally archived and categorized everything at Stereophile yet? If it's all online, that will do.



Oh, THAT! That I have done -- I posted a link to examples of mine earlier on this thread (the first attempt was glitched). Here's the URL itself:
http://www.avsforum.com/avs-vb/showthread.php



It's funny that your first response was not to address my point about your lack-of-retesting story, but to attack me and my motives for even bringing it up. And later suggest that the tale is unrelated to this thread , even though I raised it in the context of ...*retesting*, which Amir and I were talking about (specifically, retesting Fremer and you after your 5/5/ and 4/5 results).

Not only was it directly pertinent to *retesting*, it's a great illustration of the subjectivist mindset. It works in multiple contexts. So own it, JA!

Don't you know anything about human auditory threshold? Is that why you continue to wonder what if...?


If... maybe... in other words you don't know yet? And what good is claim? Wouldn't proof be necessary in this case?

Well, our sign looks like a grail .

but...but..that post is 5 years old! WHY CAN'T YOU MOVE ON?

Seems like we've gone back in time to the 1988 AES convention quite a few times as well. Lots of mileage out of that one .

I guess not. Please explain it to me.

Which as I showed with cable test, is precisely what I did .


Excuse me? This is what I said about their work:

There is no "trash-talk," "bad science, etc. That said, I happen to do be someone who has done what they do professionally. They are the Boston Audio Society. That was a curiosity for them. It was not for me when I did my testing at Microsoft. It was serious business. So when I comment on someone's work in this area, and the behest of you all constantly asking me to comment, it is not trash-talking. It is the voice of someone who has created such tests and has an informed opinion to share.

I made a point of what *I* would do in such a test. I showed that I practiced it so it was not just pretend talk. If the message is too hard to accept, then please speak to merits of it. Why would you not test the threshold of improvement further?


Then you are saying let's not have a discussion about which I am fine with. Please decide what you want. 'cause it is clear as mud .

Not at all. Let me have the court reporter read back to you what you said


You clearly said then that the mass market companies have tools that when it comes to audio fidelity, high-end companies lack. I am challenging you to list those tools. There is nothing in common between the two messages above.


Oh no you don't . You said that I should ignore a measurement because you know it is too good. It is your assertion that it is too good. So you are the one who has to demonstrate that.

Remember, on my side I have the math and physics of the thing saying it is not good. So any counter position is the one that requires proof. You want me to ignore a performance metric, show me why.

Now, we both know that you can't answer that question. Given that, you are advocating bad advice. You are telling me to stick my head in the sand without any justification.


Read the article. It wasn't my test. Their trained listeners didn't just say the system sounded back, but said it sounded broken! This is a system with a measured THD of 0.03%. Direct quote from the article:

“When the guys in charge listened to the prototype I saw dubious faces and was asked a variety of questions such as "Is the source coming from the PC corrupted?" In the end I was told to measure the audio performance. When I announced the results in a subsequent meeting I was told the distortion was an order of magnitude too high; the THD+N was 0.03%.”

You think they needed a blind test to tell the system was corrupting data? Do you need a blind listening test if your audio system acted in that manner one day?


I will repeat again: I will be getting better than 90% if not 100% of the people to say there is no difference to them between the original and one shifted 5%. Remember, the stats are in my favor. Tens of millions of TVs are sold which are way, way more screwed up than 5%. Do you see headlines anywhere in mass newspapers and magazines that people are buying non-performing devices? Nope.


I am not going to let you change the topic . I gave you a perfect analogy and one that I know you associate with as much as I do. I said I am proposing a *measurement* standard and quality bar for digital audio. You are advocating against it. I want to know why you won't do that for video, but will for audio. They are both measurements.

If I came here and said I have measured my projector with my $20,000 Minolta meter and my colors were off by 5%, would you ask me to ignore that and go run a blind test to see if I can tell the difference that way and if not, ignore the problem? If so, I like to see a few posts in this forum where people have done that.

I split off a thread about the WBF 'debunking Meyer and Moran' claims by Bruce of Puget Sound, referenced by Amir earlier in this thread, to a new thread here:

http://www.avsforum.com/avs-vb/showt...php?p=20333673

(bolding mine) No, that's not what I said, Amir. You were touting boutique capabilities and I provided a counterpoint that mass marketers have some firepower towards providing high-quality audio at their disposal too. Meaning, stuff that would have been 'high end' at one time can trickle down to the mass market thanks to economies of scale. Period. Give it up!



Huh? Ignore a measurement because I know it is too good? Haven't a clue what you're on about here. Apparently we're not communicating.




I'm asking you to provide the evidence if you heard something different on your system, and it was due to jitter. Haven't you claimed to do so? With the Revels?



As I've written, measurements + past research on human hearing can indicate when an audible difference is likely, without a blind test.


Well for one thing, show me the methods and data. Details...this is what people keep asking of you. For another thing, is that success rate anything close to what success you'd get with jitter audibility tests? If 5% green shift on TVs is significantly more obvious than the audio jitter levels we're talking about, your analogy breaks down. Just trying to keep things apples to apples here.

Heck, why not just use 320 kbps mp3? That is for sure vanishingly unlikely to be audible to most listeners, on most material, in most situations, but no one knowledgable says it COULDN'T EVER be audible. So some listeners, even those who have ABX tested themselves and have never been able to tell the diff, might still never use it, 'to be safe' (personally I'm fine with 198kbps VBR, LAME, for my lossy codec needs) . That seems a better analogy, at least it's still audio.

I would posit, though, that we haven't even established that jitter levels reported in from HDMI are audible as often as 320 kbps mp3! So again, we should worry about hearing audio jitter in our home HDMI connections? Seriously?




And of course, buying or not buying is not the same as seeing no difference. Tolerating difference doesn't mean it isn't seen. Again, though, I really need to know if you have data showing that a 5% green shift is comparable to jitter levels reported for 'failed' audio gear, in terms of perceptibility. Perhaps we can start with what your own success rate at both is, in controlled tests?



pretty funny for you to write that, given your divergences on this thread. It's not like I'm the one who brought up video, for example. Sheesh.



'Perfect' analogy? Well, we can differ on that. Video analogies to audio are a minefield, in my experience. I'm all for quality standards, btw. Let's aim for the stars. But let's not imply that if you use HDMI for 2-channel audio, you're losing out, time to upgrade! I'm not for claims of artefact audibility that aren't backed up well.

Give what up? I am just trying to get you to explain your own statement. Again, this is what you said:

"But there are economies of scale that a mass market mfr can employ that a boutique can't."

I ask again. What economies of scale mass market manufacturer can use that boutique company cannot when it comes to producing best audio fidelity?

I have explained a number of *disadvantages* mass market companies have such as cost, size, channels, etc. Why is it hard for you to back your counterpoint specifically as I just did?

You are welcome to back off and concede that high-end companies are the ones with the advantage and critical tools that mass market companies do not have. Then I will give up. Until then, if we can't settle such a simple debate, there is no hope for us otherwise.

OK, I will repeat again what started this discussion. I said that we must strive to have a minimum standard for digital audio reproduction. I said when it comes to jitter, let's pick a low bar of 500ps peak to peak since we can show mathematically that is the requirement for correct reproduction of 16 bits of resolution at 20 KHz (in some situations).

So there it is. Is it too good in your opinion? If so, please demonstrate with some evidence as to why. I gave you an example of Dirac Comb jitter for you to use.

What companies do you have in mind as being representative of each camp?

Amir, is it possible to take an existing musical piece and synthesize that jitter for audible comparison? Maybe Arny has the tools since I'm pretty sure he's done something like that using Cool Edit.

I see that you are not being serious.


I find this one very interesting.

I am dead serious. You saw no emoticon in my post.

You all keep lecturing me on audibility and psychoacoustics. Let's see some meat on the plate. Tell me what you know.

Silence means you know absolutely, positively, nothing!

Let me warn you, that was a trap and seems to have only caught your toe in it. Let see if the whole foot will next get stuck in it.

Tell me why you are finding it interesting. Again, don't say I didn't warn you.

It seems krabapple has already conceded that high-end companies are the ones with the advantage from which improvements are taken. He did after all say-- "stuff that would have been 'high end' at one time can trickle down to the mass market thanks to economies of scale. Period. Give it up!"

This presumes the 'mass marketer' becomes aware of improvements the high-end product offers and implements them to suit. Otherwise a paradox it is.


>>>

Huh?
Joe, can you name me a single audible "improvement" made by a "high-end' company?
The moving coil loudspeaker? Phonograph record? Magnetic Tape? Vacuum tube? Transistor? Electrostatic transducers? IC's? Digital conversion? Optical disc? Controlled directivity? JNDs? Perceptual thresholds? What?
Please Joe, do tell us.

cheers,

AJ

I'm not so sure krabapple is using the term hi-end like you are, Joe. I could be wrong but I interpret it as taking the flagship products from say a Sony or even Pro-audio and the computer industry and seeing the technology and performance filter into the less expensive models.

Perhaps Joe is referring to "improved" hi-end products like the not-so-jittery Pioneer SC-LX81 Amir is so fond of?