USB VS HDMI for 2ch audio to receiver

Depending on the USB implementation, yes. Usually that means asynchrnous USB. Which DAC do you have?

Does the internal DAC in Onkyo act like a USB audio device? Quick read seemed to imply it was for playing content from storage media.

Let me expand on the topic:

If you are playing video from the PC, you need to go through HDMI. USB path is not secure which means fidelity may be reduced and at any rate, you will sign yourself up for more aggravation trying to get two devices to work.

So the only reason to look at anything else is for pure music experience.

Both HDMI and USB are digital transports so that data samples go over unchanged. However, as with other digital interconnects, the timing is also conveyed and it is that timing that we worry about. Both HDMI and USB are actually pretty bad in this regard as compared to say, digital coax.

USB can be improved substantially in this regard if it is the asynchronous variety which puts the target device in charge of timing, rather than the PC. Alas, this is not in your Onkyo even if it can act like a USB audio device. That feature is available in a handful of DACs today.

What you can get is an asynchronous USB to S/PDIF converter and feed that signal to your Onkyo. My recommendation is Audiophilleo which performs this function: http://www.audiophilleo.com/

Alas, it is $500 and unless you are a true audiophile, you may not care .

Net, net, if this is all too much, the short answer is go with HDMI.

Understood. Again check the manual to make sure the USB on the Onkyo can be used for that purpose. Playing files from USB is very different animal than hooking up a wire to it and have it play from the PC. It might do that but it is not a given.

Bit-perfect it is. However, they will not both sound the same!

You need two things to reproduce digital audio:

1. Audio samples. These are the bits stored on your computer, disc, etc. All digital interfaces, above failure point, transfer these bits correctly. Bit-perfect refers to sending these bits without further processing (resampling, etc.).

2. Audio timing. This is not transmitted as digital data. Instead, the receiver looks at the incoming waveform and attempts to extract when each sample arrives. This part of the system is "analog" in that it can vary from receiver to transmitter implementation. It can even be impacted by the cable. The airport express does an OK job here, but not nearly as well as the device I mentioned.

That said, the difference in #2 is not audible to most people. So unless you have a good ear and dollars to go with it, it may not be worth the investment .

If it is only for music, you do not need USB or HDMI. Your receiver can pull music files from computer via network, if you run any software that supports DLNA protocol. Then receiver will play them like they are on local USB stick. You can share all your music library this way. I am not that familiar with Macs, thus can not recommend any particular software to use as DLNA server, so you have to do your own search.

I think the Airport Express will accomplish the same thing as letting the receiver use DLNA (I say "think will" because they may not be fully feature inter-changeable), and it will all be perfectly fine/the same/etc in terms of quality - go for it.

A few seconds on Google netted this discussion on DLNA for OS X: http://www.mactalk.com.au/19/35839-d...rvers-osx.html, might be at least a starting point (if you wanted to get into that, but the Airport Express should let you do most of the same things, namely streaming your music).

The difference is in where you control playback: receiver or computer. For home use receiver side control is usually preferred. Main problem will be with exporting iTunes library via DLNA, since Apple in its weird wisdom decided not to support DLNA natively.

Saying that they will always sound different seems highly presumptuous. Technical differences don't always correlate with audible differences. Modern gear can be so close to being perfect that two devices with excellent but vastly different technical performance will still sound the same.

The two points above are true, but their audible consequences vary to say the least.

The second point, timing, is AKA jitter which is a common audiophile Bogey Man. If jitter were so important why do some of us still tolerate and even cherish analog media which is chock full of jitter? Audiophiles should know that while digital processing can involve picoseconds of jitter, analog media has orders of magnitude more jitter and at frequencies where it is more audible.

We also see a common audiophile myth, being that there is some magic set of ears someplace that can reliably hear the consequences of some possible technical difference. In fact the sensitivity of the human ears is well known and easy to prove to be limited, even highly limited.

The scientific facts tell a completely different story than the post I'm replying to. Once audio performance reaches a certain level, there can still be differences but nobody anywhere will ever hear the differences.

If we state the problem differently, and ask which form of processing is the most likely to provide the best perceptible sound quality, then it still makes sense to talk about alternatives.

However, presuming that just because processing is done differently there *has* to be an audible difference that can be heard by some magical set of ears is simply not supported by the available facts.

Nobody can run a 2 minute mile and there are technical differences are so minute that they that have no audible consequences to any human that can possibly exist.

Maybe. There are so many variables downstream from the USB port and HDMI output that it is impossible to say much with certainty.

In the larger scheme of things the most important variable is the downstream device and to a lesser degree the path to it.

One rather obvious issue is the lip synch between the sound and the picture. Depending on the equipment and paths I use with my system, this can vary tremendously. All other things being equal, using paths and equipment that keep the video and audio signals together on the same media and equipment as much as possible will give you the best chance of good results.

When it comes to issues like jitter on the audio signal, you can pretty much count on a HDMI-driven A/V processor doing the best possible job of buffering, reclocking and generally giving effective care to the audio signal.

Dedicated paths such as HDMI seem to be preferable to paths that are routinely shared like USB or computer networking. This does not mean that those approaches always sound inferior. It can all be very good.

Thanks for the compliment .


You mean they don't always correlate for everyone. Surely you have not tested all people and all equipment. But yes, not all technical differences are audible and audible to everyone.

That fact that something is "modern" is no indication of its transparency or excellence. Your second statement of "vast technical differences" is a random generalization which means nothing. I can show you many examples of vast technical differences that are quite audible.

Because even poor digital reproduction still has full frequency response, tons of dynamic range, dead quiet noise floor, high convenience, etc. So if that is what you judge the sound by, and I often do myself in non-critical listening, digital is wonderful. You have to decide then what your expectations are.

As humans, we naturally have the ability to hear analog deficiencies. If I took everything away below 100 Hz, you will likely hear it. Typical digital distortion however can be much harder for unskilled and non-critical listeners to hear. But that doesn't mean they don't exist for all people and in all situations.

Take 128 Kbps MP3. I am sure the first time you heard it, you were amazed at how close it was to the CD. But listen to it for a while, and compare it to the original in careful A/B tests, and you start to hear where the shortcuts are taken and then an interesting thing happens: you will also spot the same distortion even when its level is much, much lower, say in 320kbps AAC. It is like learning a magician's trick. Once you do, then you become better at seeing what is hidden.

Remember, digital artifacts are not in the loud portion of the music where analog had the most trouble. It is in the reproduction of the subtle. In that sense, it is almost reverse of analog.


Putting aside that your statement is incorrect on two fronts , so? OP didn't ask anything about analog. It is like me asking which one of the two cars is better and you telling me, "well both have more power than a horse and a wagon!" I was asked if the two digital interfaces are different and I explained the objective (science) and subjective (my experience). I accept that you don't hear a difference and that is cool. If you don't need my advice, you don't have to take it.


What do you know about that science? What do you think I know about the same? Do a bit of research on person's experience and past posts before you take them on . I am not a random Joe repeating audiophile myths. I literally create them!


Where would I read about that?


What facts? Do you have a double-blind test of USB or HDMI out of mac laptop into OP's AVR?


How many blind tests of DACs and digital interconnects have you performed?

You worry about spreading myths and go on and commit the worse sin of spreading the above non-audiophile myth. There is no "reclocking" going on with HDMI. Yes, the DAC will have a clock. That clock however, is phased locked to the HDMI video pixel clock. It must. Otherwise, you lose Audio/Video sync. The PLL will perform some amount of filtering but unless it is superbly designed, it will pass through fair amount of jitter in addition to noise and other potential nasties on its ground wire and such. In addition, the high-speed HDMI signals whaling in the receiver, make it much harder to keep the DAC and its associated circuits clean. This is why dedicated DACs tend to have a leg up over integrated products like AVRs.

If HDMI had such great performance, why would we have the following jitter measurements for it by Hi-fi News (Paul Miller), vs S/PDIF which itself is a pain to get right?

"Denon AVR-3803A
---------------
SPDIF: 560psec
HDMI: 3700psec

Onkyo TX-NR906
---------------
SPDIF: 470psec
HDMI: 3860psec

Pioneer SC-LX81
---------------
SPDIF: 37psec
HDMI: 50psec

Yamaha RX-V3900
---------------
SPDIF: 183psec
HDMI: 7660psec"

Clearly if perfect reclocking was occuring, the HDMI jitter would not be 10X more than we need to reproduce 16 bits at 20 KHz.


There is nothing "dedicated" about HDMI. Its main purpose is to send video. Audio was an add-on and its performance suffers at the absolute because of it.

On the other hand, both USB and network connection can be treated as pure "data pumps," completely isolating the target DAC timing from the source. This path then has the potential to produce the best fidelity. That said, as I mentioned earlier, cheap USB implementations can actually be bad due to USB framing sourced jitter and lack of electrical isolation from noisy PCs. And networked interfaces can bring with them other sources of noise. So good implementations are important here for them to match and beat HDMI.

I won't speak for him, but I'm chuckling because you apparently don't know who arnyk is. I think you should have done a bit of research on who he is before you lectured him and questioned his background in the field.


Oops. Again, I suspect arnyk could point you to peer-reviewed articles about audio and perceptual limits as well as or better than just about anyone at AVS.


Ha ha. Do you have any idea how many DBTs concerning audibility arnyk has been involved with? I suspect if anyone in the field has done one similar to what you describe, arnyk knows about it or was involved in it.

Cheers. I don't have a dog in this fight, but was amused at your tone with someone who is one of the giants in audio DBT and myth debunking.

Oh, I did my homework. Trust me . I actually took the time to respond because people told me about him and his background.

That would be great. I love to learn if he has something to teach me. My question there was genuine even though you took it as sarcastic. I apologize if it came across to arnyk just the same.

Well, for starters, he doesn't know about any of the DBTs I have conducted so on that front, we are dead even.

Really, the point of threads like this is to learn. Declarations that we know all there is to know about a topic like this isn't data. It is an expressed opinion that is of only value to the person saying it. For members to learn something, there needs to be some meat behind the statements. If those statements hold water, then we have all learned something. If they don't, then the poster has learned something . See my example of posting jitter measurements from Hi-Fi news. That is data. Him saying HDMI does away with jitter due to reclocking is not unless he can show how that is.


Oh, let's be clear, all of this is about entertainment and amusement. Folks throw people at the two ends of this argument and then go out and sell tickets while the two gladiators spill blood. In the words of Matt Damon one of my favorite movies, Good Will Hunting, "I don't know much, but I know that!"

Seriously, I have been through these debate it seems every day of the week and twice on Sunday. The person I discuss these topics with is usually quite intelligent, very often with an engineering degree, and often but sadly not always, with some amount of DBT experience. That I take them on means one of two things: I am stupid or b) I do know as much about their misconceptions as they know about mine.

Assuming the latter, the fact that they have been involved in DBTs and such doesn't sway me one way or the other to engage in the discussion. I believe in DBTs and routinely conduct my own in addition to having done them professionally. I strongly believe in engineering and scientific principals. So if your and his assumption is that I am an anti-DBT guy, that is a very false assumption. Indeed, in the company of audiophiles, I often come across as in arnyk's camp than where I really like to be (a pragmatic in the middle).

BTW, I am usually not on a crusade to set the world right as arnyk seems to be in these discussions. My motivation is usually to allow normal discussion to go on in these threads and not have people pounded into the ground for daring to ask what is best when it comes to audio. As I noted, I was prompted to post here for that very reason. It is very hard for audiophiles to answer such challenges because they lack the engineering experience to counter with big words like the rest of us can.

What does that mean for the consumers?

What a good question!

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 .

Of the noise floor or of those infamous "wider soundstage", "more dynamic lows", "engaging mids", "smoother highs" and etc.?

Shouldn't' the comparison be between dedicated DAC and AVR?

I don't know what engaging mids is so no, that guy is not famous .

I realize you are being sarcastic but in an interesting twist, you are actually on the right track with some of those terms. Like you, I throw out most of the words high-end reviewers use to describe what they hear but occasionally, I do read something that correlates with what the machine could be doing.

This is an example of periodic 3Khz jitter, modulating a single 10 KHz tone:



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.

If jitter is random, then it works to simply raise the noise floor. Here is an example of it, albeit in a situation where it is still not audible:



This form is benign since it is not modulating the original signal in any way.

Net, net, as I mentioned other than the one guy, the rest deserve to walk on the red carpet .


Not necessarily. If you are able to turn off those circuits and hearing their effect, then you approximate the situation with the DAC where you don't have any video circuits to worry about.

I think you should think about the data you quoted. One of the products has 7660 psec of jitter and another has 50, both via HDMI. Is the difference due to some inherent problem with HDMI, or these differences in how decoding was implemented in each case?

The answer seems pretty obvious - there are differences in implementation.

I think that if you want to make a case, you should concentrate on presenting data that supports your point. ;-) You didn't.

Another issue is whether or not the SPDIF inputs in these devices also buffer the digital data. I suspect that in virtually every case the SPDIF inputs also buffer the data.

Why is there so much variation? I have no clue at this time. If I were to try to understand that I would need a ton more information.

What Miller seems to have failed to do is provide reliable listening test-based evidence that any of these measurements represented anything but what they appear to be when taken at face value, which is that they are measurements, amd not necessarily reliable indicators of sound quality.

The data above is obviously incompete, BTW. Jitter only means something if given as a function of the frequency of the signal. I see nothing about that. Jitter can change as a function o signal frequency, and often does.

The above seems to be either incomplete or misleading. Given that an AVR *does* a lot more than a dedicated audio DAC, why shouldn't it cost more?

The ironic fact is that often the AVR costs less than the DAC all by itself.

The fact of the matter is that modern digital media involves complex coding and prototocls that are generally not economical to process without complex decoder chips. These chips must be made and sold in huge quantities to be at all economical to produce. Therfore they must have reasonable prices. Because of the development costs there are not a large number of options.

This leads to the curious situation where supposedly high end digital players and decoders may be based on the same signal path components as relatively inexpensive AVRs. In some cases the high end products are mid-fi components with tirival cosmetic changes or no meaningful changes at all.


Why should it matter? Is our correspondent suggesting that in 2011 the SOTA in mixed signal design is so poor that we can't put a DAC in the same box with a front panel display? What about all those high end DACs with non-switchable front panel displays? All junk?


This hardly looks like a proper listening test to me. If our correspondent is such an advocate of DBTs, why is he advocating such a crude sighted evaluation?

I did a little research. Madrona Digital is a dealer/installer in the Belleville Wa area. I wonder if I could get JJ to drop in and give him a reality check? ;-)

You missed the key point: given the same company and presumably the same designer, they could not achieve the same performance for S/PDIF vs HDMI.

Furthermore, in 3 out of 4 examples, the jitter over HDMI keeps the system from reproducing 16 bits at the highest frequencies of 44.1 Khz sampling -- the lowest rate we run our digital systems.

OP's question was the differential in digital audio interfaces. I am showing you direct measurements that demonstrate that there is indeed an objective difference and that the notion of "all digital interfaces are the same since we buffer and reclock" is clearly not true or else we would see perfect measurements for HDMI. But we did not. Lest you think they don't know or can afford a few bytes of buffer data. Clearly this is a much more complicated affair that doesn't lend itself to trivial solutions.

As to one unit performing well, yes, excellent design does exist or else I would give up the hobby . I did want to show both sides of the coin and not hand pick data to just serve my case as I easily could have done. As I said, I believe in everything you believe in, just not as absolutely. So the fact that I put forth data to also show that with care, HDMI can be done well, should not be surprising given my position.


Indeed. So the assertion that all "modern" systems perform superbly is not true. Only one out of four managed to get there in that sampling.


They do. But buffering solves no problem there any more than it does for HDMI. If I connect my video using component or HDMI, and audio over S/PDIF, I expect the audio to stay in sync even if I watch TV for a week straight. That requirements means that S/PDIF must also stay in lock with the actual rate of samples on its input, and hence track its clock. Without good implementation, that can also cause problems.


Right. And since we can't expect him to go and conducting a 100 person blind survey with A/B tests that may be very difficult to build, you are stuck with very limited data. It is easy to demand such a test, it is entirely another matter for it to be a practical thing to get! Indeed, this is the biggest problem with claims of relying on DBT. Where is the "peer reviewed DBT" of the four products I listed if I want to pick from one of them? It does not exist and will never exist.

Given that you are not going to ever receive such data, what do you do? Blindly buy equipment and be happy? I think not if this is a serious hobby for you and you have a good ear and want the best sounding system to brag to your friends if for nothing else . The method then is to start with learning the science and engineering as I have tried to briefly explain here (much more exists here: http://www.whatsbestforum.com/showth...-s-Tech-Series). Then the measurements. Pair that up with your own listening tests and you are golden. OK, maybe also rely on the opinion of the people who have done the same.

The above combination is still guess-work but you are way better off walking in the dark and making random choices. At least when someone asks, "how did you know that was the best AVR?" You at least have something to hang your hat on.

Jitter's effect on audio that we are playing is frequency dependent. Its destructive effect at measured level gets worst with frequency (although its audibility does not track that way). Jitter itself has no dependency on anything else. If the the 60 Hz ripple from power supply modulates the DAC clock at that frequency, it is going to do that regardless of what that clock is. The source does not change in that manner.

Yes, to the extent jitter is caused by activity in your system and you change what it is doing (per my example of turning off video), then the amount could vary and indeed get worse or better. I doubt however that any of those high HDMI numbers will all of a sudden vanish. Also recall what I said we needed just to achieve the required minimum spec for 16 bits, 24 Khz.

What you have presented to me are theories which I thank you for. What I am curious about is how these theories hold up under actual use by end users. IOW, do dedicated DAC vs AVR or vs CD player make audible difference in ABX? Where are the cases? This is audio area of the forum so I suppose the focus should be in the sound aspect of of these gears.

You are both correct and not . It is true that high-end companies are stuck with some of the same mass-market parts as the high-volume, cost-focused products. With their low volumes for example, they are in no position to to go and build their own HDMI transceiver.

That said, what happens regarding jitter and DAC performance in general, relies a lot on good hygiene. And that hygiene in circuit design is usually outside of these major parts. A superbly clean linear power supply is much more expensive and heavier to produce than a cheap switchmode power supply. But clean power becomes important when even when reproducing 16 bit data, every bit of that sample represents 0.00002 volts (an AA battery is 1.5 volts). And for 24 bits, it is an incredible 0.00000008 volts (hence the reason we cannot achieve it). A lot can get in the way of such small values and good designs can sharply reduce the level of distortion.

So while it would be great if high-end designers could cook up their own optimized HDMI solutions, we have what we have now. My solution to it is go where the designer can optimize things. If I have a music only system, right now my choice of interconnect is USB with the right receiver. Here is one device I use has this great article on their jitter measurement: http://www.audiophilleo.com/definitions.aspx?jitter. As you see he is achieving 8 picoseconds of jitter (albeit, in his modified test jig). For $500, can get ultra clean interconnect to your DAC. The device is plug-and-play and goes in-line between your PC and any S/PDIF input. I tried it on the $1,200 Nova and even there, the performance improvement was quite nice! It took the device from mid-fi to near high-end. It was quite enjoyable to listen to it then, relative to its own USB input.


You are taking my homework exercise for the readers as a PhD thesis . I always think of cheap and simple exercises people can run at home to learn more about their equipment and the limits of their hearing. That was the purpose of the exercise. That said, both my Mark Levinson and Berkeley DACs let you turn off their front displays. I would be OK if they did not because I know they pay far more attention to front-panel design than mass market designs. To wit, I do not notice a difference in switching their displays on and off but do in mass market sources such as DVD players. Their switch then is to keep their glowing lights from disturbing you.

What's crude about it? Go ahead and run it blind as I have done. Close your eye, and push the switch on and off many times until you lose track of whether it is on or off. Then do it slowly and decide which switch setting is producing better quality if at all. Then open your eye. Write down which way you preferred it (assuming you did). Then do the test 2 more times. Do all the results agree? Then you have pretty conclusive data for yourself that there may be some smoke there. Not enough to get published in AES but good enough to base a post on it in a forum.

If you want to be rigorous, then subject friends and family to it. Do this, and you will be years ahead of just about everyone out there. You want to learn, this is how you learn, not reading opinion and generalization of people like me . Nothing replaces your ear, testing your gear. I could test 10,000 people with 10,000 other pieces of gear and may still not be representative of what you have.

Yes, Madrona Digital is my toy shop . I get to play with even more gear than I can buy for myself. It is amazing how that accelerates learning.

As for JJ, I should ask him to come over. You probably don't know that I hired him at Microsoft from AT&T Bell Labs to come and revamp the awful audio pipeline in Windows XP. Within the limits of what we could get away with our PC OEMs who never wanted to spend a dime extra, I think he did very well.

Yes, I know JJ thinks all audiophiles are crazy and delusional. But he worked in my group and put up with me there. I am sure he will be tolerant of me in the showroom.

If you know JJ very well, you would also know that he takes certain enjoyment out of beating the crap out of audiophiles for the sake of it. They do make it easy for him if they don't do their homework and learn the science and engineering. While I am not as extreme, I enjoy doing the same at times to the other camp. Maybe I can get him to see the other side a bit by listening to our superb PC-based music system. OK, maybe not.

"End users?" End users can do their own tests and arrive at their own conclusions. They are not going to find their favorite AVR compared to their favorite DAC. And extrapolating from other data can be quite dangerous if you are not skilled enough to know the limitations of the tests performed.

Here are some hopefully useful bits:

1. Decide if you are average Joe or real audiophile. If 320K MP3 sounds identical to you across all content, then you are the former. Don't be sad. You will save a boatload of money in buying audio gear . That is what I say about people who don't like Uni (sea Urchin). It is one of the most expensive types of sushi so not liking it, can actually be good! Me, I love it.

2. I gave examples of ways to test theories at home. Some is hard to do, others easy. For example, I find HDMI testing pain in the neck because input switching can take forever and often includes muting, glitching, etc. Others though, like switching between two analog source or S/PDIF is not problematic this way. Ultimately this is your money and your ears. So your data may matter far more than me saying Foo DAC sounded better than Foobar AVR, neither of which is something you want to buy.

3. Content is kind. It really is. Digital artifacts are dynamic in nature and your ears make for lousy instruments to use for detectability tests. You have to work at making the equation simpler to solve. And don't fall for the typical pitfall of thinking audiophile music = good test content. That can be the case but many times it is not.

Here is an example: we use Suzanne Vega's song "Tom's Diner" for codec testing (e.g. MP3 vs another codec). She signs there with no instruments and is anything but an audiophile track. Why do we use it? Because so called "music" codecs are not good at compressing speech (different science is good for that which we use in cell phones and such). Artifacts there are quite easy to hear. Now take the same voice and stick it in the middle of music and you may not be able to hear the same artifacts as easily. Why make your job harder than it should be?

One of my big gripes about jitter and research is that there has been no work done in quantifying what content is most revealing with it as we have done for audio codec where our standard 8 to 10 tracks are the bible we go by. It is not hard to work backward knowing what we know about jitter, and identify the most revealing content. Go and read any AES paper on jitter audibility and you see the sloppy approach here with tracks used with no criteria.

With compressed music, I can show you full transparency at 128K against the CD in a blind test, yet fail it at 256 with a different content.

4. Give up on some of your ideals. ABX is a pain in the neck to set up for hardware in your home. For testing computer files it is trivial. For in-home testing, I suggest using AB tests. You are not out to find the cure for cancer. No animal will be hurt in the process if you use AB instead of ABX . All you want is to learn more about your equipment, your biases, and your equipment. Searching for perfection means not ever getting started and that is not good.

5. Train your ears. As I noted before, you don't need training to clearly identify analog artifacts. If there is high noise floor, you will hear it. But how do you know the sound of jitter? Now, you could tell me ignorance is bliss. But I am not convinced. Trained listeners are not superhuman. But rather they are more reliable in blind testing because they have a calibrated reference and know quickly how to find an artifacts.

I can't easily teach you how to hear jitter (I wrote a bit earlier). I can teach you however how to hear compression artifacts. Now, the two fields are different but turns out, once you learn the difference between enjoying music and using your ear as an instrument in codec testing, you will become better at hearing other digital artifacts. It certainly worked for me.

I don't have to write a full tutorial here but the one line way to do this is to listen to the artifact when it is most severe and then go down from there. If all compressed audio sounds the same as the CD to you, compress some audio at 64 kbps. Listen carefully. Not turn up the bit rate. You will surprise yourself in now hearing the same artifact even at higher bit rates where you could not before.

Anyway, probably way more than you wanted to know .

and some of which have been DBT'd, are those using your headphones or speakers?

My blind testing has been done all with headphones. That said, I finally have access to speaker systems that are even more transparent than my headphone and find that differences easier to hear on them. But have not taken the time to do DBTs using them. I prefer watching paint dry more than that . DBT takes a fun hobby and turns it into "work."

I think of such endeavors that are for fun as doing a calibration of your lab equipment in the morning. If the results are good, you run hundreds of samples through it and don't keep rechecking every other test. Same with blind testing I did. Once I got reasonably objective data that I could hear differences blind, and correlate it with what I heard sighted and design of the system, I have stopped repeating them over and over again. Occasionally though, I resort to them like when I was testing the filter settings in Berkeley DAC.

Yes, I am mindful of bias. And as such, do make sure I don't assert my conclusions as absolute.

I take it that you have no experience with research and development. If you did, you'd know that far higher fevelopment costs are generally incurred for high volume equipment. One consequence of this is that high priestly pretenses aside, high volume equipment (which necessarily has low cost) gets a far higher development budget, gets more man hours and it likely to be assigned more skilled developers than equipment that is produced in very low volumes.


It is probable that there were one or more devices with switchmode power supplies in the production chain. There is zero evidence that it is impossible to built SOTA clean audio gear with switchmode power supplies. Numerous examples exist.


PCs are now ubiquitous in both production and reproduction signal chains. Of course they all have swichmode power supplies. Some of the quietest audio gear around takes the form of computer digital audio interfaces that work quite happily inside PC chassis, switchmode power supplies and all.


The best performing audio interfaces around take the form of PCI and PCI-E cards. They are widely used to produce recordings. If they do something bad to audio, its probably already baked into the media you buy.


You are making PhD-level claims.


Everybody who is well-informed about audio knows about the total failure of sighted evaluations when the audible differences are small or sublte. I can predict that people who try your alleged experiment will nearly 100% get the outcome you predicted, which will drop to close to 50% if reasonble bias controls are used.

Your cheap and simple exercise is a sales pitch in disguise. By giving you the benefit of the doubt and presuming that you simply don't know any better, I'll spare the other readers the ugliness of questions about your character.


No the switch it there to provide perceived value and possibly mask design failings due to the small size of the production runs.


Most front panel on/off switches have discernable on and off positions. Thus the proposed experience is likely to not be the least bit blind in actuality.

Who can say that some part of one's semi-conscious isn't privy to the actual setting of the switch, even if it is mecanically symmetrical?


No, its a highly flawed evaluation and one that has a predictable outcome - peole are falsly convinced of your mystical claims.



If family and friends are involved but visible, then you have a single blind test, which is simply a flawed double blind test.

BTW it was utterly predictable that your alledged DVTs would thrun out to be cheap imitations of the real thing.


Especially true if you're a hifi dealer and you can say the right words like "blind test" while not doing any that can stand even superficial scrutiny.



That's one reason why we rely on abstractions like thresholds of audibility and bench measurements.


If you current state of mind and posting is your proof of that, well lets say I remain unconvinced. ;-)