External DAC with Connect



Show first post
This topic has been closed for further comments. You can use the search bar to find a similar topic, or create a new one by clicking Create Topic at the top of the page.

275 replies

Some observations:

None of these "boxes" (speakers, amplifiers, players, DAC's, etc.) is perfect and each misbehaves in characteristic ways. In my experience, listeners tend to specialize in picking up the characteristic imperfections in one box while failing to comment on the imperfections in other boxes. And, there is a difference between noticing and caring about the imperfection.

I seems hard for most of us to accept, but there are quite a number of listeners who don't care much about speakers. When presented with speakers in the sub $1K class and compared to those in the several $K class, they'll just blink because there is no significant difference. Yet this same person might be very consistent at ranking and correctly identifying different pieces of electronics.

Young and old? To be sure, they are listening in different ways, but the listening experience is valid for both groups. I learned about this in my youth. My father's hearing was damaged in the military and the top few notes on a piano scale were simply mechanical thuds -- there was no pitch. As he and I shared an aural experience, we had very different opinions about what had happened. While I was distracted by all of the high frequency stuff, he zeroed in on the lows. As devices fail, particularly mechanical devices, they tend to emit low frequency distress signals. My father would quickly pick these up. I find that devices will emit higher frequency noises at the beginning of the failure signature. I could predict failure, he would more quickly identify failure -- this drove him crazy. A poorly lubricated bearing will squeak while a failed bearing will growl.

For some reason it is difficult for humans to accept that, when experiencing the same aural stimulus, there could be another conclusion that is just as valid as our own. The first reaction is that the other person is simply wrong, because the correct conclusion is so obvious. It can be a very emotional reaction. It took me many years to work this out, but it has been a very useful lesson for me as I came to understand the implications of my father's handicap. He never fully came to grips with this.

ABX? It's a great experimental design that is hard to fault, but I think that we need to step back a little and take a broad view. In other experimental settings we are careful to take into account the possibility that our measurement technique accidentally modifies the phenomenon that we are attempting to measure. For example, if we are measuring the temperature of a small sample of near room temperature liquid and we insert a relatively large, cold thermometer, the reading will be invalid. And, if we are measuring the result of a process, the result may never be valid because the thermometer might conduct room temperature into the process. My own experience with ABX has been that the presence of the box diminishes the audible differences that seemed to be more obvious before the box was inserted. Yes, level match is absolutely critical, but this is true with or without ABX.

Most audio showdowns are meaningless, with or without ABX, because they are so poorly controlled. A good experiment is hard to design and execute. Good experiments require lots of time to execute. We are usually too impatient for proper execution. If a few trials seem to verify our conclusion, we terminate the experiment and declare victory. I visualize the typical showdown "experiment" as a pack of male dogs circling a tree, vying for the right to "mark" the tree. I've done my share of "marking".

The jitter alone as measured by expert sites (Empirical Audio) makes it an average appliance for audiophiles.


Since you mention jitter, let's see what Stereophile had to say:

The calculated jitter level was 388 picoseconds peak–peak, which is low, especially considering how much processing is performed within the ZP80 and Sonosnet.


That was for the earlier ZP80. It's certainly possible the the current Connect is even better, but I've not found measurements. The Sonos system continues to be 'enthusiastically recommended' by Stereophile's John Atkinson, and has been on their recommended products list for a number of years.

For a much more objective look at jitter than the makers of dubious add-on products provide, here's a highly qualified engineer's view:

http://nwavguy.blogspot.com/2011/02/jitter-does-it-matter.html

Of note is that Stereophile's jitter test of the ZP80 shows it to be below this engineer's threshold of audibility. Are there DACs that measure better? Sure. Can you hear the difference in ABX tests? Maybe, but probably not.

My own experience with ABX has been that the presence of the box diminishes the audible differences that seemed to be more obvious before the box was inserted. Yes, level match is absolutely critical, but this is true with or without ABX.

Would that not depend on the quality of the box?

But regardless of the answer to that, the other thing is that level matching to the extent needed to achieve a correct comparison is hard to do at home, in practice. It isn't simply a matter of volume control settings since the same setting in an amp will produce different sound levels based on even tiny input signal voltage variations. Precise level matching is less important to establish that one can't hear a difference. But it is important for the vice versa, to be satisfied that any heard differences are "real" enough to decide spending decisions based on how valuable these differences are to the person doing the deciding.

What do you suggest a typical user do at home then, to see if the addition being evaluated is worth the expense? Assuming that the addition is being thought of for only SQ improvement, not for any feature.
Badge
Wow.
With all due respect- some pretty questionable audio advice in the thread.

Short answer- all elements in the chain have an impact. Your system is only as good as your weak link.

A better DAC than what is in the Sonos most certainly will make a significant difference if you have a moderately resolving system (Preamp/amp/speakers).

Even consumer oriented non-audiophile sites like CNET acknowledge the relative mediocrity of the Sonos DAC and recommend an external DAC for better systems.

You can only put so much technology into a $350 box.
The jitter alone as measured by expert sites (Empirical Audio) makes it an average appliance for audiophiles.

That said - it is great for casual listening for most all.

As for the comments that any amp will do and put your money into speakers is absurd.
If you have an average source, and an average preamp and an average amp- your speakers (no matter how good) will never get the music to begin with.

Use your head folks- you get what you pay for and it all matters.

+100!
Glad to see I'm not the only one in the world coming to these findings. However, few individuals here, love to come up with scientific theories and a couple of articles from the stoneage to prove it is all the same, no auditable difference. Although listening on itself has nothing to to with science, but is pure subjective...
One advice, don't try to convince them, but be proud you know better...
Would that not depend on the quality of the box?


Of course. We should be having parallel flaming discussions about the merits of ABX box 'A' and ABX box 'B', but the box is always above reproach -- without discussion. In other scientific inquiries we have these discussions. "How is my measurement technique influencing the quantity that I am measuring?" is a critical consideration. Level match is only one component of the technique.

For the purpose of level match it is not so hard to develop an adequate measurement technique for electronics. In this case repeatability is more important than absolute accuracy. Comparing output levels for electronics boxes is much easier than comparing acoustic output of speakers. The cheap voltmeters sold at the local electrical or electronics store are a poor choice in this context.
scientific theories and a couple of articles from the stoneage

Stoneage? Whatever - if you doubt the validity then offer any properly conducted study that supports your view. But then that's not as easy as just attacking is it?


listening on itself has nothing to to with science, but is pure subjective...


Congratulations. You appear to have completely missed that that is exactly the point being made. Subjectively one particular component may sound different to another to any individual listener, depending on many factors which I can't be bothered to list again. That doesn't mean that objectively there has therefore to be an actual difference which is measurable and repeatable.
+100!
Glad to see I'm not the only one in the world coming to these findings. However, few individuals here, love to come up with scientific theories and a couple of articles from the stoneage to prove it is all the same, no auditable difference. Although listening on itself has nothing to to with science, but is pure subjective...
One advice, don't try to convince them, but be proud you know better...


Yes, be "proud" of being gullible enough that with a little hype and a lot of snake-oil, you too can be bilked out of many thousands of dollars of hard earned cash! Pay no attention to that man behind the curtain!! (not to mention objective science), LOL. :rolleyes:
I'm always amazed at the willful ignorance of the audiophools. Really doesn't take much time to educate oneself on the science of audio reproduction, there is a considerable body of research widely available thanks to the internet. Yet they choose to listen to those whose livelihood depends on advertising dollars from 'high end' marketeers. Baffling.
I'm always amazed at the willful ignorance of the audiophools.
It is a belief system, one can't argue with that. All that can be done on a forum such as this is to present the opposite view as rationally as possible, so other interested people can make informed choices.
Userlevel 1
Badge +3
I just tried this and found the difference to be quite dramatic

http://www.hometoys.com/emagazine/2013/10/audiocontrols-rialto-400/2191
Interesting review, but dramatic is subjective unless it is found in a level matched DBT. Comparing it with a Connect Amp also isn't fair, it delivers almost double the power, so speakers that need it will sound better with it than via the Sonos Amp. It also seems to have some signal processing onboard, which makes comparisons of SQ more difficult, since apples v apples isn't present.

Level matched DBTs are admittedly not easy to set up, but that doesn't mean there is a better way to assess any feeling of audibly superior SQ. The human brain hears what it wants to hear. It is also easily fooled, so that just a 0.2 db difference in sound levels will make it think that the louder sounds better.
I just tried this and found the difference to be quite dramatic


No doubt. The "AccuBass" appears to be bumping the bass rather dramatically. Hardly accurate.


If you have the need to drive 4-ohm speakers or need more the 55W per channel, clearly the Rialto 400 is a good fit without even looking at the DAC.


Might want to read Stereophile's review. The ZP-120 can deliver 127Wpc into 4 ohms...
Userlevel 1
Badge +3
No doubt. The "AccuBass" appears to be bumping the bass rather dramatically. Hardly accurate...

I didn't care for the Accubass (personal taste)
Badge
I agree with that wholeheartedly. The Sonos Connect is a relatively inexpensive device, and as such, has limitations. Its DAC is one of them. I use an external DAC, a Weiss DAC202, in one of my zones, and the difference in sound isn't subtle. I think that a lot of the luddites posting here saying that better equipment doesn't matter probably do so in part because they can't afford to find out, or are too cheap to do so, and so they sit back and sanctimoniously slag those of us who have tried better equipment and have found an improvement.
The Sonos Connect is a relatively inexpensive device, and as such, has limitations. Its DAC is one of them.

I think that a lot of the luddites posting here saying that better equipment doesn't matter probably do so in part because they can't afford to find out, or are too cheap to do so, and so they sit back and sanctimoniously slag those of us who have tried better equipment and have found an improvement.

I can't speak for others, but in my case I haven't found any difference using a USD 1500 DAC that was also much heavier than my CD player. It was well built, but had no sonic impact. Now someone might say that I need to use a USD 70k DAC for it to qualify as better equipment...
Generalising on the basis of what something costs, in the area of digital sound reproduction where modern day electronics are concerned, isn't a good idea. Based on my experience of DACs ranging from those in the Connect to the one referred. Age has its advantages perhaps, because I find no difference from even the DAC in an iPod touch playing lossless files, played in my resolving system that costs many multiples the cost of the iPod. Change the speakers or even their placement enough, I can still pick up differences in what I hear.
Badge
Kumar - In my main system, I use a Naim HDX server as my source, most of the time. When I listen to the system (not with Sonos) with the output from the HDX to my integrated amp, the sound is okay. When I listen to the digital out from the HDX connected to my EMM Labs XDS1 CD/SACD player, making use of the DAC in the unit, the sound is much better. I firmly believe that better DACs make for better sound. I have no scientific data to back this up - just my ears. By the way, I agree with you that speaker placement is important.
the sound is much better. I firmly believe that better DACs make for better sound. I have no scientific data to back this up - just my ears. By the way, I agree with you that speaker placement is important.
If it sounds better to you, that's fine. Perhaps it will, perhaps it won't to me - but that is hypothetical and irrelevant.
Consider this though - even a 0.2 db increase in the sound level, needing to be measured by instruments to be eliminated in a controlled blind test, can make the music sound better to any human ear. It is an old trick that every good stereo salesman knows.
I am not saying that this applies in your case, but it is possible that signal level voltage/input sensitivity differences could be causing these in your system with no change in the volume control knob setting.
And better DACs will make for better sound, provided that if you are using the word "better" with reference to instrument measured output parameters, these improvements are within the threshold of normal human hearing. My experiences lead me to think that good enough - in the sense I refer to - is reached quite low down the price point with the resolution of the issues that plagued early days DACs now widely known, and not expensive to eliminate using mass manufacturing state of the art tech.
In digital audio spending more money will get you lots of things - but not, beyond a point quickly reached - better sound quality that your ears will convey to your brain, if you deny your brain every input other than the one conveyed to it by them.
Given that your ears may be better than mine, in your case that point may be higher than mine, but not by a whole lot, I suggest.
Bottom line - if your system sounds great to you, wonderful. It doesn't however flow from this that those who can't hear the sound improvements from a better DAC are Luddites or worse.
A thought experiment for you - if you were to do the sound level matching to less than a 0.2 db difference, and the Connect was to sell for USD 2000, would you think different?
Badge
Hi Kumar - Good points. My ears tell me that the higher quality DAC sounds better. When comparing the Naim unit with my EMM Labs unit, there is a distinct improvement in the presentation of music - I guess you would call it the soundstage. Musical information just seems to open up. I could talk about air around instruments and blah, blah, blah, but that would just make me look like an audiophool, as one poster has put it, and I'd like to think that I'm not a fool. Would I think any different with regard to the Sonos Connect if it cost $2000? Well, if allowed to demo it in my system, comparing it to my other DACs, and if nothing else changed, I would still likely come to the conclusion that its DAC is not as good as the others, surprising given the price. I understand your point. Just because it's expensive doesn't necessarily make it a better product. I've been around a lot of high-end gear over the years and have reached that conclusion long ago. However, having said that, I still think that as a rule, you get what you pay for.
there is a distinct improvement in the presentation of music - I guess you would call it the soundstage. Musical information just seems to open up. I could talk about air around instruments and blah, blah, blah, but that would just make me look like an audiophool, as one poster has put it, and I'd like to think that I'm not a fool.

However, having said that, I still think that as a rule, you get what you pay for.

The things you describe is exactly what is universally heard when sound levels rise by as little as 0.2 db, unless they were at very loud levels to start with. And exactly the same things are heard to collapse when the sound level drops. It is universal and nothing to do with being a fool.

What comes in the way of getting what you pay for in terms of heard sound quality in this case are the two very different scales at which the speciality hifi audio companies operate compared to volume players like Sonos and many others. For the former all costs per unit are higher, as is the need for a higher per unit profit to sustain the business. That translates to a higher selling price for essentially the same product. For this they may throw in a better looking cabinet, more features, better hardware service years into the life of the equipment - but they can't extract better sound from the commonly used electronic designs and componentry, once some minimum standards are met in this area. This has been true of amplifiers for decades now, while the DAC side of things have evolved to a similar level much more recently.
Different rules apply for speakers, where the rule applies much more because getting/making better components such as cones, tweeters, enclosures, and the other mechanical things a speaker incorporates can be more expensive. Which isn't to say that there aren't any lousy expensive speakers. Or the vice versa. But this tends to be an exception.
I am with Kumar here. I have a Benchmark DAC (with pre-amp) between the Sonos and the McIntosh power amp and Thiel speakers. It sounds great but when I tried it without the DAC the difference was minimal and certainly not worse.
I am with Kumar here.
All amplifiers that have a instrument measured flat response and are working within their design limits, do not have a sonic signature under controlled testing, i.e.,they sound the same. For the flatness to be such this is true to the extent even the best of human hearing can notice that it is not, is a problem that has been solved decades ago. The sonic signature that is heard is because of sound level differences, or because speaker behaviour in terms of demand for power is driving the amplifier to beyond the design limits where the delivered response is flat.
My listening, some reading about digital audio and common sense tells me that the same stage has also been reached by now for thing such as CDPs, DACs and streaming boxes such as Sonos. I would however still not make as categorical a statement about these as I can about amps. Can anyone else here, based on more knowledge/experience? The parameters for the relevant flatness would be different than for an amp, but the concept should be similar. It should actually be easier, because the X factor of speaker power draw will not apply here.
This of course does not mean that USD 70k DACs won't be promoted.
And while a USD 100k speaker may well have the power needs that only a USD 50k amp can fulfil, there is nothing that I can understand will be different on the input side of that amp that will need a USD 70k DAC. I know I would still struggle to hear the difference between that one and a Connect wired to the analog inputs of the same amp. But it would be interesting to know what kind of measurements/thresholds one would look for in a CDP or a audiophile Connect, for it to be good enough in terms of not having an audible sonic signature to the most acute listening for such a system, provided of course that nothing but the ears are in play. And the sound levels are level matched by instruments.

And while a USD 100k speaker may well have the power needs that only a USD 50k amp can fulfil, there is nothing that I can understand will be different on the input side of that amp that will need a USD 70k DAC.

Thinking a little more on these lines - I don't know of any 2 channel speaker in the world that needs more than 1000 watts of clean power from the amplifier. But I do know of a few solid state amps that can deliver this, and they don't cost more than USD10k.
Imagine a system of USD 100k speakers, USD 10k amp of 1000wpc, and the Connect wired to its analog inputs, playing lossless files. Using commodity speaker cables of adequate gauge, and good enough interconnects from the Connect to the Amp - the entire cable part of the system available for USD 200.
A level matched AB test of this system with another costing USD 200k using the same speakers - the money being spent on amplification, DAC and fancy cables that don't have inbuilt sonic signatures - would be revealing were any one to carry it out. Saying anything more would be speculation.
My listening, some reading about digital audio and common sense tells me that the same stage has also been reached by now for thing such as CDPs, DACs and streaming boxes such as Sonos. I would however still not make as categorical a statement about these as I can about amps.

When I compared the Sonos's own DAC with my Benchmark DAC1 HDR, all other things being equal I did hear a difference (or at least I thought I did). The Sonos sounded a bit "warmer". In view of all the other parameters that influence our perception of sound, like the recording quality, acoustics of the room etc. I would not lose any sleep over minute perceived differences.

As to amps, Kumar said it all, although I do wonder how loud he plays his music ... 1000 watt is a lot of "spare room" considering that the volume in a living room (at least our living room) rarely gets above 80 Db and a few watts.
I do wonder how loud he plays his music ... 1000 watt is a lot of "spare room" considering that the volume in a living room (at least our living room) rarely gets above 80 Db and a few watts.
I don't need or have a 1000w amp:), I only referred to it as an example of the most power that any home audio speaker in the world will need!
My music listening is also at moderate to low levels, no head banging stuff.
Where the very large majority of modern amps are concerned, all that is needed from a matching/SQ perspective is whether it has adequate power to drive the speaker in question with some power in reserve - if that is the case, it won't have a sonic signature so one can choose based on things like reliability, features and after sales support.

What will be very useful is to have a similar specification parameter for a modern day DAC, such that if that is met/bettered, the same will hold good for it. And use that parameter/s to assess the Connect.

Beyond that differences may exist, but if they can't be heard in a controlled test, they are irrelevant.