Have a few questions regarding DAC's - Page 4

I think that it is completely reasonable to want to ensure that digital signals are converted to analog as accurately as possible. Unfortunately, a lot of misunderstandings about this process persist, even though A->D and D->A conversion are among the most economical and perfected operations in audio.

Of course this was not always the case. Back in the early days of digital audio, and continuing as late as say 2002, it was not that hard to encounter examples of D-> A done badly which is to say with audible defects. Common situations where this problem remained included portable music players and PCs.

Even in the mid-1990s, most CD players for home audio as a rule had good (which is to say free of audible defects) convertors. The first generation CD players had converters added undesirable but subtle changes to the music they played. The second generation players were generally free of this problem, but were still quiet expensive.

Since the early 1990s, there have been major two areas of change in good DACs. One is that they have become profoundly less expensive. The other is that what I would call overkill DACs have become common and are also becoming very inexpensive. By overkill I mean DACs that perform at or beyond the level of Redbook CD which is to say 20-20 KHz response within a small fraction of a dB and 90 or better dB dynamic range.

In fact DACs with > 110 dB dynamic range and capable of flat response to 100 KHz are available for pretty reasonable prices. They show up with corresponding ADCs in equipment that sells for less than $200. The extra performance has no audible benefits except perhaps for audio production where audio is often converted back and forth between analog and digtial a number of times for a single piece of music.

How inexpensive have DACs become? For example one of the less expensive music players around is the Sansa Clip. It provides Ipod-level performance for down in the $20-30 range. It is tiny and it runs for hours on an internal rechargable battery. It also performs as well as a good home CD player matched to a very good integrated amplifier or separate preamp and amp, or a receiver. Its core circuitry is a chip that is what used to be a supercomputer with all components except mass storage, and 2 DACs and 2 ADCs that sells in production quantities for less than $10. The DACs are almost an afterthought, but they still perform very well on the test bench and in listening tests. The Clip+ includes a digital DSP-based FM stereo tuner that performs near the theoretical limits possible for FM reception in the same package., so it is literally a portable digital player and FM receiver in one postage-stamp sized package.

I've seen this recommendation in several places from Sonos users, although I can't remember the precise links off the top of my head. I seem to recall that a Stereophile reviewer has suggested using an outboard DAC with Sonos as part of reviews of relatively inexpensive DACs.

SDL

P.S. Here's a link to one related discussion on a Sonos forum: http://forums.sonos.com/showthread.php?t=24607

I took a look at the SP technical Measurements of the ZP80-100 http://www.stereophile.com/content/s...m-measurements and I reached the following conclusions:

The ZP100 built in power amp does not measure up to even the general standard of good mid-fi. Like a lot of class-D amplifiers its source impedance is all over the map within the audible range and its frequency response can vary to a possibly audible degree into some speakers. Even so, compared to the other audible defects of those same speakers, its not a big problem.

The line level output, particularly at a real-world -1 dB output level was really just fine. I don't know how an external DAC of any high grade or lofty pretensions could possibly provide a reliably audible benefit.




The comentary there looks pretty thin speculative to me. Saying that adding some megabuck DAC would improve sound quality is like a self-appointed eggspurt cliche these days.

Measurements aren't everything but in this case they compose all of the reliable evidence that seems to be available.

There are a a number of possible explanations for that which have nothing whatever to do with the innate sound quality of those components. I don't want to get into a big long thing about the necessity of blind tests and bias controls here, so suffice it to say that you should be wary of drawing firm conclusions from such experiences.

Thanks for taking the time to look through the Sonos measurements. I have 2 of the ZP90 units (now called the Connect) and one of the Play 5 powered speakers (for use in the kitchen for less crtical listening). I don't have the ZP120 amp, which is the updated version of the ZP100 that was tested by Stereophile. My interest in a higher end outboard DAC was stimulated by John Atkinson's Stereophile reviews of the Sonos gear, as he used a very expensive Levinson DAC with the Sonos with reportedly great results. He thought the internal Sonos DAC was pretty good but not in the same class as his high-end DAC.

SDL

What John Atkinson thinks he hears should be of no concern to you. If he can show you measurable differences sufficient to explain audible differences, then you should listen to him. Other than that, he's just talking out his you-know-what like every other fool reviewer.

Thanks for not moving the discussion back to focus mainly on blind testing and related issues. I already know those arguments. In fact, I'm a behavioral scientist who has published double-blind treatment studies in respected medical journals, so I have more than a casual understanding of placebo effects and participant bias.

My experience with preferring some amplification choices over others in my home system has been based on how well I can create a realistic musical image in my living room that draws me into the music whenever I listen to it. I don't have the equipment to do blind, level-adjusted A/B comparisons between components, but I do have the ability to change components in and out of my system and to "live" with equipment long enough to decide if I like it. My tastes have not always been based on what I expected to hear or on the cost of the components in the system, but I have learned what sound I like. I do not have "golden ears" and in fact have a gradual age-related high frequency roll-off in my hearing -- but at least the roll-off is consistent in both ears.

The only reason I asked earlier about you own choice of components was to see if you have assembled a low-cost "good enough" system based on your experience that most differences between amps are not audible or whether you have chosen some more expensive gear in selected areas based on features, reliability, build quality, or other issues.

I know that you strongly support blind audio testing and I agree that blind testing can be an extremely useful tool, but I would also love to understand how your audio knowledge translates into choices in audio equipment.

SDL

Why not list the controls you did employ, and we'll shoot from there.

@sdl...

your first paragraph implies that you understand that controlled testing is required...

your second paragraph, however, implies that you don't understand controlled testing is required...

you can't have it both ways...

That seems fine and good if you are able to apply what you have learned in other fields to audio.


This makes your work different from that which goes before you how?


Nobody does until they put forth the effort...


This make you different from the average audiophile how? How are you benefiting from what you say you know about doing reliable subjective testing?


One of the key relevant aspects of power amplifiers is that we know what an ideal power amplifier does, and we know how to build systems that deliver the sound we like without burdening amplifiers with deviations from that ideal.

So, I'm trying to figure out how you knowing what sound you like related to doing effective subjective tests of amplifiers.

And yet when it comes to audio you seem to forget these very basic things:


______________

1. I use uncompressed digital audio (or, when necessary, compression no worse than 256 kbps), and I don't sweat about the DAC used to convert this.

2. I ensure that I have enough amplifier power to drive my chosen speakers to the volume I require without audible distortion.

That's it. Beyond that, I buy what I like for reasons other than sound quality.

SDL - You mention "knowing about" proper controls in experiments and blind testing, and then go off and ignore it all and come up with a bunch of uncontrolled and unrepeatable, highly biased results, and claim that because in some other venue you know about this, it puts you outside of the normal "box" in picking audio components. If that's the case, I've got a few cures for a few major diseases I'd like to discuss with you - I've done no clinical trials, but I've spent time living with them, and know they improve the situation...

Do you see the logical problem? (By the way, I'm not a medical doctor).

As you were told - it can't be both ways. There is truly a point where good enough becomes good enough, and it's usually about 1/100th as much money as whatever other wackadoodle hi-fi rag tells you need to spend. And even that's not really a fair statement, because we're still assuming that cost matters for anything but telling us how much trouble we'll be in with the better half at the end of the month!

This is very helpful -- and is close to what I do as well. (I'd still love to know more specifics about your equipment, but that's fine if you don't want to share those details.)

It doesn't surprise me that you and several other "objectivists" on this thread have chosen to find my statements about blind testing vs. extended listening as inconsistent. I expected that response after reading the comments leveled at SAMIK earlier. Your criticism of me would be true if your chosen listening strategy was without flaws. However, both rapid A/B comparisions and extended subjective listening are flawed assessments -- though they are each flawed in different ways. I will continue to take what I can from what I read and what I hear and make audio choices that are practical and satisfying for me, using a variety of methods to get there. I will continue to look for accurate measurements of equipment, blind product testing, extended listening sessions, and advice from experts to inform my decisions. That's why I joined this discussion of DACs in the first place.

Thank you again for sharing your perspective on outboard DACs -- it really was helpful. Of course, it would have been a lot more fun if I hadn't had to listen to lectures about blind testing while extracting the helpful bits from the discussion.

SDL

There's no need to get defensive about the topic. Just to speak in generalities, usually people have their minds made up one way or another before wading into such a discussion. And it usually leads to hurt feelings and a very defensive tact being taken by the "audiophile" side because their opinions and beliefs are usually very personally qualified - "I heard this and that's how it is!" and anyone who disagrees is (necessarily) challenging their face. With the "objectivist" tack, it's more of a "I disagree with your facts/data/findings/summary and here's why..." which usually (at IME) doesn't result in as many hurt feelings. So basically what I'm saying is - one may participate all they like, but if that's going to just be nit-picking any "objective" test/data/number/figure/value to death and ultimately declaring it an opinion to equalize it with their own observations, I don't see them getting very much out of the discussion. I'm not trying to "single out" anyone here, just sharing an observation I've made based on these kinds of threads.

This is like saying that randomized control drug trials are flawed in their own way. There's a gold standard for drug testing, and everything else is bulls**t. And there's a gold standard for listening tests, and everything else is bulls**t. That doesn't mean that even the ideal test can't produce an incorrect result. But nothing else is even worth discussing.

The tests I rely on come reasonably close to that gold standard. And, more to the point, no "tests" that point in the other direction come anywhere close. I'll leave it to others to draw their own conclusions about your approach.

Buy and listen to what pleases you. That's what this hobby is about. However, if you (or others) are going to post audibility claims about component "x" over component "y", it's helpful to back it up with evidence. Note, evidence and personal opinion are not same thing.

Wow, it's always interesting how people who are being defensive about their firmly held beliefs seem to feel that it's only other people who are being defensive and taking things personally. I'm not trying to single out anyone here, but some posters on this thread seem to believe that they are simply stating facts and that anyone who questions those "facts" is not being logical or "can't handle the truth." They also seem to have little appreciation for how insulting their comments become when they completely dismiss anyone else's perceptions as b.s. I've always thought that it was a good thing to be able to appreciate objective data and combine it with personal listening experience, but apparently I was mistaken.

Just to remind some of you who might have forgotten, I have never claimed to be able to perceive differences between DACs. I haven't had the opportunity to even do any comparisons (blinded or not). I simply asked the question whether differences between DACs do exist, and I mentioned that I have on occasion formed opinions about what sounds good to me without doing a matched, blind test. The rest of this b.s. comes from other posters.

For anyone who cares about some of the actual DAC issues discussed in this thread and who doesn't want to re-hash a lot of tired arguments about critical listening strategies that have bounced around the internet and audio mags for decades, here's what I've learned:

1. All modern DACs sound the same -- at least that's the consensus of the loudest voices on this thread.

2. Don't waste money on an outboard DAC unless you like the box it comes in or it makes you feel better for some reason that has nothing to do with how it sounds.

And you guys didn't think I was listening!

SDL

The key is to understand those occasions when your personal listening experience conflicts with the objective data. And the correct understanding is almost certainly going to be that you should toss out your personal listening experience.

When you get an unexpected result, it usually means you screwed up the experiment.

I would like to modify that:

All modern DACs should sound the same. The ones that don't are probably because the manufacturer didn't care what they sounded like.

Did someone say you did?

Given the sensitive responses to my prior questions, I'm a bit hesitant to ask, but which manufacturers' DACs don't sound as good as the ones produced by manufacturers that "care"?

SDL

You're making interesting choices. I would have thought that objective data could help narrow down those component options that might be most likely to make a difference for you, but that the ultimate decision should be based on your personal listening experience.

On the related issue of flaws in both blind tests and subjective listening, I dug up the link to a classic discussion that sheds some light -- and a lot of heat -- on the arguments: http://www.avguide.com/forums/blind-...itorial?page=1 (I even noticed comments by Arnyk in that discussion, and many of you have probably read it.) I would guess that many posters on this thread would take a particular side in the debate, but I would expect that most objective readers would be persuaded that both approaches have significant flaws.

SDL

Not quite. It should be based on personal preference. Note that I didn't mention the word "listening". Why? Don't forget, there are other senses such as visual, psychological, social (some like to show off)...etc involved when owning an electronic component.


I had a feeling you were thinking of that as soon as you mentioned "flaws in both blind tests and..."
Is it safe for me to assume that you know about the author?

Without naming brands/models, I'll give one example - those with poorly implemented and usually unnecessary tube stages at the output. These can degrade the SNR and add measurable distortion artefacts.

Again without naming brands/models, those cheap, near $30 "adapters" from Asia that you might find on Amazon. The small dongles with an S/PDIF in and a pair of RCA out, massed produced for the lowest price.

I had one that was reviewed well, but whose output clipped when fed a full-range signal. I changed a resistor, and it was then fine. I assume it was designed with components of nominal tolerance, and went too-hot when a component went too far from nominal. Simply a lack of attention to detail.

I suspect that in what you might find a surprising number of different forums, you would be guessing wrong. In many situations blind tests are the gold standard for subjective evaluations.

Your post starts out with an implicit false claim that blind testing and subjective listening are two different, even opposing things. Did you intend that? ;-)

I should invoke TOS 8 based on listening tests using a general run of commercial recordings. ;-)

Here's the reason why - clipping or at least significantly higher distortion just below FS is not that uncommon. However, its relevancy to listening enjoyment is most people's primary concern.

For example, AES standards for testing digital converters in PCs put referene levels a dB or more below FS.

Sure there are recordings out there that run right up to 0 dB FS and just sit there and bang on it. Thing is, they are often clipped in the same region, even when played back without distortion.

If you are listening to quality sources, most if not just about all of them never get within even a few tenths of a dB of FS.

I like to see quality design as much as anybody, and I don't like to see equipment that clips below FS. That all said I know from practical experience that with almost all recordings, clipping in the last 0.5 dB below FS is generally sonically benign.

Now if your resistor change addressed clipping several dB below FS, then work well done! And if you made the change out of a pursuit of truth or goodness, no problem. But its audible consequences for clipping in the last dB before FS? Not so much!

At this point DACs with audible flaws are usually found in places where people seriously don't care, like penny-cheap digital players and the analog audio outputs of CD and DVD drives.

In contrast even low cost but quality alternatives like a Sansa Clip (ca. $30) have really good converters. Other than its tendency to run 0.5% fast with the OEM firmware (but not Rockbox) Clips and Fuses are really very sonically transparent. And iPods are even a little bit better on the test bench.

One thing that a little study of psychoacoustics has taught me is to be very skeptical of drawing conclusions from my own listening experience. Human beings are not calibrated test instruments, not even close. They shouldn't pretend to be.

Now, obviously there are some things that cannot be left entirely to objective measurements—speaker preferences, room set-up. There you have to trust your ears. But you should also be honest about their limitations.


I wouldn't call them "objective." I'd call them "people who don't know any better."

Look, there really is a science behind all of this. The fact that all properly designed DACs and amps (with some assumptions in the latter case) sound the same is so accepted in that field that it can be found in college textbooks. Believing in anything else is like believing in laetrile.

And, Harley??? You're citing Harley??? Remember when I told you to ignore Atkinson when he's opining, as opposed to measuring? Ignore Harley when he's breathing.

Really?

Cites?


Yes, I love John. who could be confused with the most schizophrenic (as in conservative with test equipment and radical with his ears) reviewer in the history of audio.


LOL!

I was surprised to see that Harley didn't remove my posts to that thread. What did I say wrong? ;-)