USB or SPDIF to DAC Questions - Page 2

Roll your eyes if you will, but it will cost you in your pocketbook. So if wasting money on the latest-greatest DAC is what you want to do, be my guest but use your money.

If one uses the "scientific double-blind wizardry" that has been around for over 30 years, one quickly finds that threads like this one that has become littered with all sorts of double-talk about the inadequacy of SP/DIF and the latest-greatest asynchronous whatsis is all about people's illusions that they are hearing differences that are actually inaudible.

It's like this: You can improve the sound quality of your system for minimal cost but encounter some amount of hassle and even possibly negotiations with the dreaded WAF, or you can blow your money on unnecessary DACs and jittter busters, keep the wife happier with a succession of relatively small but vastly overpriced signal processors and golden digital players, and make zero actual progress on improving the sound quality of your audio system. It is really as easy as that!

Is this your choice for record players? I've heard great things about it. It takes a very basic approach with no excessive engineering or silly, overpriced circuitry. I wish DACs would take the same approach. There is too much evolution of technology. We need to bring it way back to when things were simple!

Name calling wasn't enough so now you resort to trolling. That's sad considering how much learning you could have gotten by digesting what's been replied to you.

I've learned plenty from my own experience and research into the subject of DAC design and more so USB transport implementation. I hope to learn much more because I certainly am not an authority on the subject. Unfortunately, the "experience" of others that I've found posting on this forum weighs closer to stubborn ideologies based on technology of the past, and an unwillingness to entertain new technology and design at the very least. This approach seems to label anything that may be new to some of AVSForum posters as "inferior" or unnecessary because it doesn't incorporate something familiar. For example, USB Class 2 audio. If Sony/Philips didn't put their label on it, it must not be good, right? I don't believe that to be accurate.

First time poster here. Registered just to be able to say this.

The presentation of information by Arnyk is typically condescending and aggressive. I think that its important to critically evaluate proposed technologies in a way that can get around some of our human limitations on detection, but make sure that you communicate that to people in a way that doesn't make them not want to share their experiences with you.

Forums like this are a good place for people to talk about their experiences with different technologies. Typically even the most wealthy and diligent of individuals can only do limited testing of capability equivalence between units. Most of our homes are not laboratories, we should however try to provide our feedback in a way that isn't biased based on what we've decided to purchase with our own money. I believe that Hydrogenaudio does a good job of communicating an A/B testing standard, is there an article or thread for AVS that talks about how you should test and then communicate back results?

I do feel that Arnyk has made a considerable investment in this community in terms of his sheer amount of participation. So I don't critique him to be destructive, I just feel that you guys could do a better job of welcoming the presentation of new data. Put all your testing/bias/decision stuff in a single article and link new users to it... don't flame new people because of their lack of experience with SDT or testing.

I also think the Fischer price photo was pretty funny, not necessarily inflammatory.

Actually, back in the day when vinyl was all we had, the below is exactly what I used while you were possibly using the above:



What you can't see is the V15 III, IV or V depending on when the picture was taken.
Edited by arnyk - 6/19/12 at 9:02am

Apparently, you didn't' do any proper listening tests because if you did, you'd most likely sing a violently different tune.

You could do something weird like listen to people who have done their homework. ;-)

That is true, only in your mind.

You clearly disrespect people who have actually done their homework in this regard. I don't know why I bother to point this out. You are clearly enjoying your blindered view of reality.

No, it is all about what would happen if you had to distinguish a Behringer UCA 202 from your choice of more modern products in a proper listening test.

What you don't seem to understand is that when the price of an adequate solution gets too low for high margin suppliers to make any money on it, some of the more unscrupulous of them simply resort to "The Big Lie" school of advertising and other forms of promotion. That's how you sell $4000 Martin-Logan speakers when most listeners prefer the $400 Infinity Primus in a blind preference test. They just do what worked the last time they sold the Emperor's new Clothes. ;-)

What I find disappointing is that both you and arny are the ones off base.

Some useful info was mentioned by several others on asynchronous USB.

Has it gotten that one has to be so politically correct on AVS in order to not ruffle any feathers?

I liked it better when arny actually posted meaningful refs, and didn't act like an old broken record.

Really?

My point has been that you don't need to spend a bunch of money on fancy DACs anymore, That often equipment that is relatively simple and inexpensive performs in a way that is audibly indistinguishable from far more expensive and complex solutions.
I've personally done dozens of DBTs involving these issues with a variety of systems and audiophile listeners, and the results support this kind of thinking.

Got any reliable evidence that shows that Asynchronous DACs sound different at all, let alone better in real world situations?

I don't think so because you are the one who seems to have the ruffled feathers.

Let's talk about the "Every DAC sounds different in my non-level-matched, non-time-synched, sighted evaluations" broken record....

I'd like to hear something different from that, something that shows a little sophistication when it comes to doing listening tests.

Lots of people around here seem to think that it is still reasonable to do listening tests like they were done in the 1950s, back in the days when pretty much everything did sound different because most of it was so crappy.You know, the days of tubes, vinyl, and analog tape.

Still too expensive for my blood.

But the hot Chicago weather did get me to start this new "2 Channel Audio" thread: Yes you CAN hear jitter!

There's no doubt that some jitter is audible. But only some of it. How do you know that you are hearing jitter?

Wow!

Hi Arny, I don't think Asynchronous USB is about doing it better, it's about doing it differently. The goal would be to make it cheaper to do it well. So it's ironic that Async USB is more expensive right now.

The problem with the old isochronous USB protocol was the the DAC timing was a slave to the host's USB timing. Besides the issue that the host's timing might not be the most accurate, the main problem was that USB's 1-millisecond "tick" is too course to lock a good 1.4 mHz (minimum, 12 mHz typical) DAC clock to. It is especially difficult if the USB tick has jitter.

The Asynchronous USB solution was simple: Allow the DAC to define the clock, and have it request the audio-data from the host as needed. It's a simple change to the software protocol (actually, it is adding a fifth protocol to USB's existing four protocols).

As for the cost, the current high price is simply exploitation. The added software protocol allows for a reduction in hardware, so async-USB should and will be cheaper.


As an aside: a complaint against S/PDIF is also that it contains the timing, and a DAC needs to slave to the clock of the source. But at least S/PDIF's embedded clock is at the same bit-rate as the data, and not over a thousand-times courser, as it is with iso-USB.

My review of the technology says that Asynch USB is:

(1) More complex
(2) Different

Either is a sufficient excuse for higher prices, at least in the short term.

I see a lot of if's. Where the rubber hits the road relates to the presence or non-presence of something audible. This forum has a well-known poster who has hyped aysnch USB as some kind of major sonic advantage. Many people myself included have asked him about listening tests to support his claims. Not so much. Its not like this guy doesn't know better. He's an EE and an ex Microsoft vice president who at one point retained the services of one of the best names in audio listening test technology. To me, his lack of productivity in this matter is just another example of the actual lack of faith that he has iin aysnch USB technology.

It appears to be solution looking for a real-world problem.


Where's the beef? The above discussion is not about any results in terms of acutal audible sound quality, but rather it is a proposed means to an end that is not known to be an audible problem.
Edited by arnyk - 6/20/12 at 12:32am

As I suspected, just another troll.

Both true. However, the added complexity is in the software/firmware, where it belongs. In exchange for the additional handshaking required, the hardware is greatly simplified.
The engineer needs to consider each of those "if"s, and address them correctly to get a well-performing DAC. But if (another one) the time-base coming from the host USB is not solid, then there is little that the DAC can do to get a low-jitter clock and also stay synchronized with the data-stream. So a normally excellent isochronous DAC can be compromised by a poor source.

Can be compromised, but usually not. So normally an Asynchronous-USB DAC would offer no audible advantage over an Isochronous-USB DAC. There can be those situations where the Async DAC will perform better than an Isoch DAC, but it is not that one is sonically better, it's just that the other is broken. Asynch adds a level "source-immunity".
The real-world problem is clock-recovery, an engineering issue. Async-USB replaces the complex PLL requirement within the DAC with a simple oscillator. The end-user won't see a benefit until the price comes down to where it should be, which is cheaper than current DACs.

As I said, the current high prices are simply exploiting the hype. If I had invested in Async this early, I probably would do the same.

You and I seem to have different ideas about what constitutes a problem, which is in its way, fine.

I think that a real world problem in audio is something that affects sound quality. Whether or not some engineer is happy with how something relatively esoteric like clock recovery works does not strike me as being real world when it has no audible consequences.

As far as ragging on SPDIF works goes, this is very old news. The world is full of people who are using SPDIF, and perceive no problems. I have personally done many straight-wire bypass tests of computer systems that were tied together with SPDIF and Toslink, and there is no audible difference between them and ideal data transmission.

I have a number of modestly-priced DACs that do a near-perfect job of turning questionable SPDIF and Toslink bit streams into very low jitter audio signals. This is part of the promise of digital, and IME it works out in the real world.

My challenge remains - I'd like to see someone do a good DBT of a Behringer UCA 202 USB converter, with positive results. It's about as non-Asynch as they come. Until someone shows that there is an audible problem, what have they fixed?

I have become aware that there is a clique of people around here who say that I'm against progress and favor technical mediocrity. Of course those are their self-serving perceptions. I don't have a problem with change as long as the costs and benefits are in line.

One of the goals of product development is to be able to claim that your product is different and better. As long as people base these decisions on sighted evaluations, they are exposed to a very strong possibility that they are hearing what they want to hear, regardless of whether there is any difference that can actually be heard.

I look at the millions of dollars that were wasted, and management careers that were hurt or ended by the rush to so-called high definition audio distribution formats and wonder when people will learn?

yeah yeah, 0's are 0's and 1's are 1's, we know that song 

 

you wanna hear jitter induced THD? Listen to CS8416, I posted a PDF earlier about its sluggish performance.

Who said that? Oh, you said that!

So that's how it works - you make stuff up and then try to create the impression that other people said it?

Do you really think that the levels of THD shown there are audible? At worst its 80 dB down, which is well below what can be heard under even the most ideal conditions (ca. -60 dB).

Please come back with the results of DBTs showing an audible difference. ;-)

For your review: a paper pointing out the well known problems with the use of THD as a means for characterizing the audibility of nonlinear distortion:

http://www.gedlee.com/downloads/The%20Perception%20of%20Distortion.pdf

Note that it shows real world examples of situations where nonlinear distortion that is far higher than you seem to be crating alram over, even in the -10 to -20 dB range, was not audible.
Edited by arnyk - 6/20/12 at 7:36pm

BTW, if there is an audible problem due to this, where are the positive results from a DBT relating to it?

hehe, did AVS get hydrogenized overnight or something? I like my forums w/ ppl who try stuff IRL w/ their own ears, the day you'll be able measure SQ won't be anytime soon.

 

Yes, jitter is audible and yes, the XMOS solution sounds better than anything else. Now listen to it if you dare, and stop wasting your time trolling on internet forums for a change 

Where is your listening test evidence of this claim?

How can something sound better than products that already are indistinguishable from the original signal source?

Another troll jumps in.