Hi-Resolution Audio and Sonos

Show first post

172 replies

Userlevel 7
Badge +21
Are they truncating the 24bit to 16bit as a way to reduce network overhead?

No, but my guess is the hardware only contains 16-bit DACs.
For all the people who have quoted Nyquist–Shannon, here is my complete and full statement on why Nyquist–Shannon actually proves that high-res audio works, and Nyquist–Shannon can also be used to prove that CD resolution does not fully and accurately reproduce the original audio analog signal:


Lol. This guy is an incompetent nitwit. Looks like just about any know-nothing can post on Medium.
I'd be amazed if the DAC, even in the old ZP80/100, wasn't 24-bit given what was commonplace at the time.

One of the posters here did some tests a while back which suggested that 24-bit FLAC/WAV content was truncated, presumably near the start of the audio pipeline. The irony is that the pipeline is mostly 24-bits wide, with the digital volume control operating in 24-bits. It was confirmed long ago that the digital outputs were 24-bit, with the lowest byte becoming non-zero when volume was reduced from full scale. In other words the 16-bit content was shifted downwards, without loss of music information until about -48dB.
Userlevel 7
Badge +21
ratty's explanation is much more likely than mine. My audio hardware knowledge is decades out of date...

Well, that graph is very clear yet it doesn't make the point,  the subject is to prove should oversampling and using more bits than actually needed, in few words having redundancy digital input can be actually identified by trained ears as presenting a more lifelike experience.

I am in the case as it was Sonos and only Sonos that forced our household to jump onto HD audio through Amazon Music HD, now as I am not able to play more than cd quality I cannot say is there’s any difference. I am sure that if there was any wtahsoever It would me minimal in comparison to the jump from Spotify up to CD quality

However I have read an expert opinion on that:


What do you think of Mr martin opinion?

Userlevel 7
Badge +20
Gentlemen, differing opinions aside, let's reel it in, shall we? Clearly this is a topic that you all feel passionate about but try to keep personal attacks out of it. Thanks.
Userlevel 4
Badge +5
Don't give up 'todayamerican', some people are well known to be more interested in putting people down than listening to good music on good equipment. They probably know perfectly well that ABX testing is thoroughly discredited. They keep advocating ABX even though it only shows up gross differences. People who listen to real instruments know that it takes quite a while to appreciate what something sounds like. Would someone really buy a new guitar having tried 10 guitars in the shop for a brief period each?

People who think that 16/44 is 'perfect', the pinnacle of music reproduction, are rather like the idiots who believe that all amps sound the same, or cables, or DACs etc. You might as well argue that all violins sound the same, or guitars, or pianos. Even drums sound quite different from each other. Buy an amp on the basis of ABX testing and it will probably give you a headache after an hour's listening. You need to listen for a long period, with an open mind, and don't allow yourself to be blinded by theory. ONLY the sound matters.

I attended a PROPER test of hi-res a few years ago. 12 of us went into a large room at a recording studio where a string-trio were seated between some large speakers. The speakers and musicians were facing the audience. Before anything was played, we were blindfolded. Music was played for 5 minutes. It was then repeated 4 times. One was live, one was 16/44, one was 24/192 and one was analogue. We didn't know which was which.

ALL of us identified the 16/44 as being a recording - without hesitation. About half of us could hear a very slight hiss on the analogue recording but otherwise it sounded live. NONE of us were sure which was the live music and which was the 24/192 recording even though we were all musicians or audio technicians. There is undoubtedly something about 16/44 which robs the music of the impression of being real. The analogue and the 24/192 recordings were clearly better. Both give a far better illusion of real instruments than 16/44 and that is exactly what I want from my hifi.

If you only listen to processed pop music, hi-res is probably irrelevant because you have no idea what the original music sounded like. If you listen to classical or 'un-plugged' music, hi-res makes a slight but noticeable difference but only if played through GOOD equipment. Try it with budget equipment, or a receiver, or an AV amp and it will probably be lost on you. I should add that I cannot identify hi-res via headphones, not even good ones. Other people claim that they can hear a difference on headphones - and who am I to claim they are wrong. We all hear, see, taste, smell and feel things with different levels of ability and discernment and I feel sorry for those who can't hear what I hear.

By the way, the string trio sitting in front of us played absolutely nothing; they were only there to distract us. The live music actually came from musicians playing in the studio next door so that everything we heard came through the same microphones, amps and speakers. NONE of us noticed that.
My school must have been retarded because we did not dive into Nyquist–Shannon in year one.

It is very unfortunate that digital audio is typically discussed in an incomplete manor. We present the stair step analogy for the analog to digital conversion part, but we forget to include the integration block on the reconstruction side. I guess that this is somewhat understandable because it requires some thought to grasp. One must thoroughly understand Nyquist–Shannon before realizing the implication of that integration. Further, most of the conspiracy theorists have already left the lecture screaming "jaggies" before the reconstruction part is covered.

Misinformation is available at all levels. At a convention of audio engineers a presenter was making his case for narrow bandwidth (stopping at 20KHz on the high end) and a wide band challenger claimed that his spectrum analyzer was capturing information from his Phono cartridge at 200KHz, therefore the audio chain must be able to pass this. Behind me was a disc cutter designer who remarked that "we can't put 200Khz on the master, he must be seeing the consequence of some sort of resonance".

While I can't refute your observations in this test, there are some loose ends in the testing methodology. First, how was the 16/44.1 version generated? There are good and not so good methods. How thorough was the sound isolation between the studios? There could have been some subtle sound conduction between the rooms when the live musicians were playing. Also, the live renditions will be different from each other and the recordings. You also did not specify if the recordings all started from the same session and master file. Was the analog tape recorded live or derived from a master file? For example, the master file could have been sampled higher than 24/192, then down sampled to the formats used in your listening session. I'm also not clear if there were four or 16 listening trials. And, the order could be important. Why? The recent noise exposure of the listeners can effect their perception. For example, if I've been in a vehicle of some sort or exposed to loud music (at a venue), I avoid critical listening for a while. In this context, the discrimination thresholds of the listeners could have changed during the test.

There are so many details that need to be controlled. I'll simply relate an experience that I had a few years ago. I was walking on a city street (light traffic) and my brain suddenly received a "message" -- live band. I couldn't see a band. I continued walking and a couple doors later I observed a small group rehearsing inside a restaurant. The doors were closed.
And of course nevalti will be publishing his "PROPER" test of Hires in a peer reviewed scientific publication real soon.
Badge +1
So, I recorded at the studio last year. I initially got the mix in CD quality. I can assure you folks that the 24 bit, 48 KHz mix of the same music sounded a lot better than CD quality, which loses all the warmth.

I think people who have learnt atleast one musical instrument can make out the difference in resolution as we are trained to hear the real instrument over time.
Userlevel 1

The Truth About High-Resolution Audio: Facts, Fiction and Findings

Mark Waldrep


My enthusiasm for high-resolution has diminished in recent years. After reading numerous studies and articles on the topic, I'm inclined to agree with those that believe Redbook CDs are sufficient to capture all of the fidelity we need when listening to recorded music.





Excellent article, and some comments as well are worth reading!

2 years back, i was on the market to buy a multi-room system.

3 contenders:

  • Sonos
  • Denons Heos line
  • Blue Sound

Like many people, I was excited by the His Res audio capabilities of the two last systems. But I was less enthusiasmed by their price tag..

After much reading, I came to the conclusion that, my ears, combined to the conditions of listening, the source of my music (despite having a bunch of 96/24 FLAC) and the money I was willing to invest, it was not worth the money.

I even ordered an Play 1 and a Denon Heos 1 to make side by side comparison (and I played some random FLAC 24/48, 24/96 to compare with the 16/44) and I ended up returning the Heos and ketp the Play 1 and never looked back.

I am perfectly happy with my current Sonos system, eventhough is it only 16/44 ! Maybe some extravagantly costly system would sound better than Sonos, but I just listen to music, I am not conducting surveys for a living…

If Sonos brings Hi-Res audio in Sonos Controler 2, the better, as I have not converted my 24/48 library into 16/44 yet, an di am lazy….


Userlevel 7
Badge +21
Sadly folks think all kinds of things but when presented with a double-blind A B test can't correctly guess either way.

There are a few exceptions when over-sampled files cause audio artifacts that make the music sound worse.
Here todayamerican, have a read:


Particularly the bolded in this (parenthesis added by me):

Strictly speaking, the theorem only applies to a class of mathematical functions having a Fourier transform that is zero outside of a finite region of frequencies. Intuitively we expect that when one reduces a continuous function (i.e analog audio) to a discrete sequence (i.e. digital audio) and interpolates back to a continuous function (again, analog audio), the fidelity of the result depends on the density (or sample rate) of the original samples. The sampling theorem introduces the concept of a sample rate that is sufficient for perfect fidelity for the class of functions that are bandlimited to a given bandwidth, such that no actual information is lost in the sampling process. It expresses the sufficient sample rate in terms of the bandwidth for the class of functions. The theorem also leads to a formula for perfectly reconstructing the original continuous-time function from the samples.
Badge +1
And some people blindly believe what the books say without trying out themselves.

I used to believe the books too. Until I tried it out myself to see if high res made any difference and sure thing it does.

I an an electronics engineer myself. The way digital music is stored, it does not capture all the information. There is definitely loss based on sampling rate. This I learnt at college.

Some say you can’t make out the difference. But so many people can make out some difference. I think you just have to hear it out yourself and figure out whether you can.

I come with no bias. I heard it myself and made the determination that I can hear better at high-res. My own music, I play back at high-res only because otherwise it loses a lot of the “warmth” = mid & low frequencies. I can only make out the difference in that. Everything else feels the same between CD quality and high-res.
Userlevel 4
Badge +5
And of course nevalti will be publishing his "PROPER" test of Hires in a peer reviewed scientific publication real soon.

There we are 'todayamerican', you can see exactly what I mean.

It's not worth bothering with people with cloth ears and cr a p equipment. Just keep listening and ignore people who kid themselves that everything can be determined by measurement. It has been said many times before but... Above a basic DIN standard hifi level, you can no more determine sound quality by 'measurement' than you can determine wine quality by measurement. Human beings are far more subtle than that and what each of us perceives as 'quality' is very likely to vary from person to person. For example, I went to a concert today ((Tchaikovsky 5th) and it was almost ruined for me as I had to sit in row J instead of my preferred row M. Many people will not have a clue why row J is inferior yet other people will understand the problem immediately.

Anyone who wants to try the test I describe can easily replicate it if they have half-way decent equipment and an acoustic instrument or even just a voice. Anyone who doesn't bother can hardly expect to have a valid opinion. I tried it at home with a humble Sony M10 and could easily tell the difference between 16/44 & 24/192. One sounds 'real' the other simply does not.

In response to the sensible question, there was only one performance and first generation recordings at 16/44 and 24/192 on identical machines plus a 2" analogue recorder - all using a simple crossed pair. No downsampling, mixing or other processing.

I'd love to watch @nevalti post his nonsense over on hydrogenaudio. The industry experts over there would chew him up and spit him out, lol.

Hydrogenaudio requires A/B/X testing, and all their super powers disappear when faced with A/B/X. A/B/X is like Kryptonite to the Golden Ear types.

If three rows make such a vast difference, is it not conceivable that the acoustics of your living room have a much greater impact on 'sound quality' than whether the music is 16 or 24 bits? An honest question.

A friend moved his legacy separates set up to an acoustically better space after some years of ownership, and was astonished to discover that he had owned a HiFi system all those years. It can be a day/night difference, the speaker placement/room acoustics. Tuning like Trueplay can help, but getting the acoustics right is a much larger part of the solution.

The wine analogy is actually useful here, where psychoacoustics - not an invented word by the way - are concerned. There is a lot of rubbish surrounding wine tasting also, down to experts being fooled by a red coloured white to claim that it goes better with a steak than an uncoloured white. And ditto about the use of expensive crystal to drink it in. That said, given a choice, I would always drink mine in a decent glass than in a steel tumbler, simply because the experience is beyond just what is tasted and/or smelt. A similar logic can apply to sound from a better looking speaker that is identical in sound to another in a blind test. And as long as this aspect is appreciated, I can understand people that don't like blind AB/ABX testing which restricts the listening experience to just that provided by the ears. Instead, silly and unscientific claims are made for trained ears, better kit, of red herrings like Hi Res - and of course supposedly scientific explanations as to why blind ABX is flawed.

As to Hi Res, I suspect that the internal componentry that is mass produced today and is therefore the cheapest to buy, has Hi Res capability by default, and if Sonos were not concerned about music play stability or backward compatibility, that could be enabled in all presently manufactured Sonos kit without any extra hardware costs.
What is stopping Sonos from supporting 24 bit 192 khz?
What is stopping Sonos from supporting 24 bit 192 khz?
Userlevel 4
Badge +5
Ho, ho, that got the pack excited. 'Oooh look, another victim to pounce on'. Feel free to join back in 'americantoday' but it seems that I will take their flack for a few days.

Notice that not one of them even tried to replicate the test I described. I doubt that any of the jackals really LISTEN to their music. If they are only using Sonos - I'm not surprised. Sonos is a truly excellent system for user convenience and it is more than adequate for their stand-alone speakers. Sonos do not however offer anything like top quality audio, not even for 16/44 recordings, and nor do they claim to any more. They did to start with and they also claimed that their system delivered a bit perfect stream to the hifi system - but it doesn't and they no longer make those extravagant claims nor do they try to compete with the more modern systems. The jackal pack hanging around the Sonos forum conveniently ignore all that.They have invested so much energy in defending Sonos (I wonder why) that they probably wouldn't dare to sit back quietly and LISTEN. Even if they did, few of them would admit to hearing what they have been missing because it would ruin their hobby of forum pouncing.

I fell for the hype initially and used Sonos for hifi for a while, feeding a 'Connect' into a Benchmark DAC2, a Bryston 4BSST (or a KT88 triode amp) and PMC OB1i speakers. When I wanted to try hi-res for the first time, I hooked up my old Squezebox again. I was quite shocked to hear how much better the old Squeezebox sounded using exactly the same DAC and hifi system. It even sounded noticeably better in a blind ABX test - which only identifies big differences. Both the Squeezebox and the 'Connect' were fed into the same Benchmark DAC2. The Sonos system and the Squeezebox had ALL signal manipulations including volume control disabled. They were both playing exactly the same 16/44 files and delivering the supposedly bit-perfect digital stream into the Benchmark DAC2. I could switch instantly from one to the other whilst playing the same track and the difference was not subtle. The illusion of reality was far better with the Squeezebox - and that is the whole purpose of hifi to me.

Having realised that the 'Connect' was simply not delivering what I wanted, I tried one of my old ZP90's -with similar results. The older ZP90 was actually very slightly better. I gave my 'Connect' to an electronics engineer who worked, at the time, for Naim and asked if he could test it. When he returned it, he told me that it was 'far from bit-perfect'. The rest of his explanation went over my head.

Since then I have been experimenting with alternative systems and found that the Qualcomm system is better but frustratingly unreliable. I also found that the humble Rasberry Pi using JRiver Id software sounds significantly better than Sonos and, set up ready to play, it is less than a quarter of the price. The JRiver IdPi system can multi-room in hi-res too. It is nothing like as user-friendly as Sonos though.

This leads me on to another surprising discovery. Whilst LISTENING I noticed, much to my surprise that even the Windows programmes I used made a difference to the sound. I had been using MediaMonkey and was astonished to hear that JRiver Media Centre sounds noticeably better! I wonder if that can be measured. Given that surprising 'discovery', I now wonder if the Sonos software is partly responsible for the degradation that I found unacceptable in my hifi system.

Well to answer the Original question, I am one of the few one that has a use to the Sonos High resolution. I play sometimes files ripped on my Sony PS-HX500. They sound excellent and if the original album is good they sometimes sound more pleasant (Pleasant rather than detailed) than CD quality streamings from Amazon Music, in the real life sometimes I have had to do the opposite, down convert some ripped files to AAC because the album is sibilant or has some distortion that my Sonos amp can show. When I stream the same file on a Bluetooth speakers from my iPhone this distortion is hidden by the digital filters, hence yes I can notice a different but I have had sometimes to do the trip in the opposite direction to lower resolution files to hide some bad harmonics from the original Vinyl. On the other  there is higher resolution than CD and this is why they are called masters, modern Vinyl is cut from digital masters made at 24/96. The point is if this makes any difference in the Home Audio my answer is probably not but the conversion of any file to higher or lower certainly does even if just altering the loudness. I myself had to buy an app for my mac “bit perfect” to make sure the mac DAC is set at the same rate than the file being played as it completely changes the loudness or the way it sounds. Sonos cannot be different but before adapting S2 to streaming to higher quality I’d rather have my amp reading less compressed files like Dolby Digital Plus, this is way more likely to have an impact. I believe that Sonos should update the software to improve the reading of this kind of files even if it has to convert to 24/48. Compatibility and proper reading is more important than the resolution itself. The ones who writes this has bought more CD’s in this Year than LP’s so I am not sold to the Hi-Re hype but find incomprensible that SONOS doesn't have a wider file compatibility as the Hardware had the capabilities   since the Original Play:1 Play:3. 

Well we can at least agree on something then. Sonos isn't an audiophile system. It is not priced as an audiophile system. I, thank goodness, am not an audiophile.

So, lacking @nevalti's sophisticated appreciation of music and audio quality, I still manage to enjoy my Sonos system somehow.

I need a drink. Pass me a glass of Blue Nun, please, someone.
Userlevel 4
Badge +5
I need a drink. Pass me a glass of Blue Nun, please, someone.
I do appreciate the joke but Blue Nun is actually a very good illustration that all wines are not even similar. Same applies to audio equipment.

"Sonos isn't an audiophile system".
You certainly don't need an 'audiophile system' to get good quality reproduction - just look at my mediocre system. Sonos is however a premium price 'product' and should be judged as such. It is a very good system for anyone who is not interested in listening carefully. You all know that as well as I do but you still feel the need to attack anyone who questions it's performance. Mr 'americantoday asked perfectly sensible questions and was attacked rather than helped. Shame on all the jackals.

Has anyone actually LISTENED YET? Are all the jackals just yapping the same old, same old?
So that would be a big "NO" on performing your test using foobar2000's A/B/X function @nevalti?

So tell me again, who exactly is refusing to "LISTEN"?