Sonos Connect DAC

I know this is long past the time that the decision has been made, but I think it still worth correcting a bit of what jgatie stated. Yes, it’s true that bits are bits, so the music stream is the same whether one chooses optical or coax/SPDIF. However, that doesn’t mean that the connections are the same. John Darko, for example, states that the Connect:Amp digital coax connection is a noisy connection. And it is true that noise or interference induced through that connector could lead to a worse output through coax. I haven’t tested this for myself, and it could also be a unit defect based on his experience rather than general throughout the Connect:Amp range.

Anyway, the stream itself is the same, but that doesn’t mean the signal you end up with is the same if the Connect:Amp is pushing noise through one of the connectors.

Your amplifier’s DAC is likely better than the one in the Connect:Amp, and if Mr. Darko is right, then the optical output is likely better than the coax. But the only way to know for sure in your setup is to try each connection method and decide for yourself which is the best. It will depend on your equipment, but also on lots of other factors like your room and your listening style.

The bottom line, though, is that it likely doesn’t matter that much. Do you like the music and the way it sounds? Then it works for you! And, ultimately, that’s all that’s really important.

Where you say 'Connect:Amp' do you mean 'Connect'?

That “if” is a big stretch; everything he writes is as subjective as is that of everyone testing and talking about home audio kit on the net and print media.

No one reports on the basis of rigorous and therefore scientific testing/listening. Because if they did, they would not have much to write about where modern audio kit is concerned.

John B: Yes, sorry, Connect. I have both and misspoke.

Hi, Kumar. Thanks for the input.

I have no idea whether the user’s Connect has a noise problem in the SPDIF. That’s why I think any Connect user should buy at least one optical cable and one coax cable and try them both out to decide which to use. (And an RCA cable, I suppose, although the DAC in his/her amplifier is almost assuredly better than the one in the Connect. But, why not try it out just in case?)

I wonder, though, do you really think that subjective interpretations of audio equipment are not valid? I mean, sure, one could measure the amplitude of reverberations vs. direct sound of a specific recording on two loudspeakers, for example, but isn’t it just as useful, and also faster and easier, for a reviewer to listen to the speaker and describe it as having either more clarity or fullness? Especially considering that our preferences and experience of listening to music are entirely subjective?

Perhaps we’ll have to agree to disagree on that. But, I hope that we can agree on this: none of that matters one whit -- what’s important is that each of us is enjoying the music that we listen to, and that it brings joy to our lives.

Thanks again for the comment!

I did not say that they are not valid, just that they are subjective and therefore not to be relied upon except as a valid opinion of the reviewer, which may not apply to you.

And I think that listening is just as important as measurement for speakers, but speakers can sound vastly different from room to room; leaving aside the problem that how it sounds is also driven by how it looks, and even by how much it costs - these are psychological biases that all humans are susceptible to. In the case of speakers there is also a subjective preference for the sound of one speaker over another that does not mean one is superior; and the preference may easily be reversed for someone else.

Further, for kit other than speakers, listening isn't as important as measurements - it is impossible for an amplifier that has identical calibrated instrument measured specs as another to sound different, but the Darkos of the world would have us believe that to be the case. Which is what allows them to wax lyrical when this years model of the amplifier is released, having done exactly that for the prior iteration. 

As to the music, of course I agree. It is the proverbial moon, and those that obsess with the kit end up focusing on the finger pointing to it.

And my problem with the quoted, starting with the last bit. On what basis do you suggest that the DAC in the amplifier is assuredly better than the one in the Connect. DACs are cheap commodities now, and their implementation is also widely understood by audio engineers. So why would one be assuredly better? Whether that one be the Connect DAC or the other one.

Unsurprisingly, there is not a single level matched double blind listening test anywhere that establishes this claim you make. Not one. Not even with a DAC unit that costs USD 10,000.

Ditto with noise problems in SPDIF. The only reason for that would be the rare defective Connect, or the not so rare defective cable/its connection at either end. Choosing which one to use, for the vast majority of cases, boils down to what jacks are available on the partner unit and what cable is at hand.

Kumar: Mea culpa. I’m sure there are plenty of amplifiers out there with worse DACs than the Connect. (I qualified my statement with “almost” but you’re right, “almost assuredly” overstates the likelihood that the Connect has a worse DAC than the amp.) And furthermore, it runs contrary to my overarching belief. Even if a DAC is objectively “better,” (whatever that would mean), that doesn’t mean it sounds better to you.

I think I’ve muddled things too much in this discussion. Let me leave it with a clear five statements to declare what I meant to get across:

1. While digital signals are less susceptible to noise/interference than analog signals, it’s still possible for them to be degraded/affected by outside factors.
2. It’s possible for two different digital outputs on a single component to have different amounts of noise/interference induced into them. (Whether this be a design flaw or a defective unit or other circumstances.)
3. We cannot know the specific condition or characteristics of a user’s own component in his or her home.
4. The way a person experiences music is highly subjective, subject to characteristics in his/her listening room, the types of music he/she likes to listen to, musical history, hearing condition, mood and temperament, and yes, even the way he/she feels about his/her equipment.*
5. If a person wants to experience the music in the best sounding way to him/her, ideally he/she should try out for himself/herself each of the output methods (in this case, digital optical, digital coax, and analog RCA) and make his/her own decision.** (Or sure, if he/she is happy with it, just use whatever cable happens to be lying around.)

*This is why there’s no need for a listener to have a friend come over and give him/her a blind listening test.*** If he/she believes the music sounds better through, say, a digital coax cable, so what? He/she is experiencing the music as “better.” Who are we to judge?

**As I recommended from the beginning. I.e., listen and decide for yourself. (Because there’s reason to believe that there might be a perceptible difference. I never guaranteed that there would be.)

***Although, if that’s the way you like to experience music, go ahead! Whatever brings you pleasure. They’re all equally legitimate.

With reference to trying out different output jacks and cables, there are very few scientifically proved cases of these making a difference once psychological factors are not allowed to come in play. One can still do this if one has the time and the inclination, not to mention money, but it certainly isn't required, using your words, “ideally, to get the best sounding music”. Note here that it is close to impossible to do comparative testing at home in a manner that rules out biases as well the psychoacoustic effect of even small levels of difference in sound levels with louder sounding better. There are orders of magnitude more chances for being misled by these, than for the difference to be an objective one.

And it is this “trying out” that leads people down the garden path that marketers of expensive but otherwise no different interconnect cables would love them to take. For no audible benefit, but with considerable hits to the bank account. Again, fine for you if you have fun doing this and can afford to, but certainly not universally the ideal thing that must be done to get the best sounding music.

What may give rise to defective sound from a chosen for convenience option will be down to rare cases of defective kit, or cable/connection faults. But these will be clearly audible, very rarely subtle. And if subtle in the rare case, why worry about them if one is enjoying the music?

I disagree. I am sure that in the vast majority of amps that have DACs, the latter are just as good as that in the Connect. And even for those that are not, or are better in the spec, the difference will not be audible in a blind test in a vast majority of that small minority.

And the minority of the minority is something that is very small even if it exists, because no one has found it yet in a way that is scientific.

Curious, if all modern DACs are essentially equally competent, why Sonos should claim a 'richer' sound from the new Port.  Is this achieved by baked in DSP settings or by the selection of a 'better' DAC (that will, at a later date, be revealed as better than 24/48 capable - only to keep up with competitors, you understand, not because hi-res can sound good)?  

And what is a 'richer' sound anyway?  Less clinical, more ... er ... analogue?

It’s marketing speak. Undefinable in scientific tests. 

@BrianJ :

I was very curious to see if Sonos will follow the herd in this aspect and was disappointed to see that they did just that, with this “richer” nonsense. It also can't be by any baked in DSP settings because as a source component, such baking is bad design if done where the EQ controls are set to flat.

About the only reason that remains for it to sound richer is because it looks richer - as in with newer looks - than the Connect.