Hi-resolution support required (eg: AIFF 96kHz/24bit)

  • 13 August 2013
  • 80 replies
  • 18007 views

Userlevel 3
PLEASE add hi-resolution file support (eg. AIFF 96kHz/24bits) In a world where CDs are becoming obsolete, I believe a good share of Sonos customers are still looking for hi-fi quality, not only the mediocre (format) quality provided by most digital music services/stores. It is rather purpose less to invest in the acquisition and maintenance of such a high quality product as a Sonos system and, not being able to play the best sound quality available with it! Thanks for kindly prioritizing such improvement :o)

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.

80 replies

Sonos has chosen to offer other features in preference to hi-res support.
It appears that it wasn't a case of trade-offs. Sonos were smart and went with the scientific consensus:
http://www.whathifi.com/news/sonos-plans-brighter-and-brighter-wireless-music-future
Hi-res numbers don't add up
While we’re talking about the competition, we can’t help but ask Sonos about its feelings on hi-res music. No matter who we asked, the answer was the same – don’t hold your breath.

“We’ve looked really hard at it”, says Tom [Cullen]. “Of course we want to make sure we’re not missing anything and we feel pretty good that we’re not. There are arguments you could make about deeper bit depth, but we are unable to make a meaningful argument on sample rate. We tried, we can’t – the math just isn’t there.”

Giles Martin agrees: “There’s a huge, cheap miscommunication about high-resolution music. Manufacturers try to put a number to it, like 0 to 60 in a car, and it’s not that simple. I think we need to improve the overall quality of listening full stop. When everyone is listening to the quality of music they were listening to 15 years ago, then we have a discussion about pushing it further.

“Once we get to that stage where there is a huge consumer demand or it, we should accept it. For now, there are neither enough tracks, nor enough value in it for the majority of consumers for us to chase it.”

We ask about downsampling, which Sonos currently doesn’t offer - meaning any high-res music you own is completely unplayable on Sonos unless you downsample it yourself.

Jon Reilly admits: “We want to play everything, so we should do and that’s fair feedback. It’s something we’re considering.”
and 24 bit isn't just about dynamic range... its dynamic accuracy too. 24 bit recording sound smoother and at 24bit 96khz are considered equivalent to DSD recordings.


Nonsense. Any differences in quality heard (even the highly subjective, audiophile-speak ones like "smoother"; you might as well say things like "syrupy" or "orange peel" :rolleyes) have been proven to be purely due to mastering differences. And the ability to hear a difference in the dynamic range between 16-bit and 24-bit is equal to the ability to detect a pin drop over a jackhammer in the same room.

Things like high-hats, strings, vocals sound clearer, smoother and closer to how they sound on good vinyl recordings.


So a higher dynamic range makes it sound like a format with a much, much lower dynamic range? Are you sure this is what you want to say?
Hi Bernhard

Would you please explain further how it is that playing hi-res files across a multi-room setup would be challenging to the Sonos components? Surely each Connect or Play acts individually, even in a multi-room setup and therefore decodes its own data stream, regardless of what else is happening on other Sonos components within the same system?

I can appreciate that there might be a strain on the home network, especially using wireless connection rather than Ethernet. But given a hard-wired setup I cannot see why Sonos itself would find it difficult. I have all my CD collection currently encoded as FLAC, presumably as 16/44.1 as they were ripped direct from CD. But I have a small collection of 24/96 files on disc, and some legal downloads from hi-res sites, and having to down-convert them all first to allow Sonos to play them is a pain.

To Guy's point - I agree that "good enough" quality is NOT good enough. Good enough for whom? Once upon a time there was a common conception that MP3 @ 128 was "good enough". Back in the turn of the century, wax cylinders were no doubt "good enough", but that didn't stop them being replaced with shellac discs. And then with the microgroove record. And then with CD (Better? Separate discussion entirely). The point is that time and technology advances the art of what is possible, and companies should not be dismissing these advances.

Playing to the lowest common denominator is not acceptable in these days of choice. Saying that there isn't a market out there for higher resolution and quality replay is simply not true - audiophiles and music-lovers all over the world still pursue their ideals of perfection, with whatever means suits them - physical media or data files.

Current streaming solutions from traditional hi-fi manufacturers such as Naim, Chord and so on prove that hi-res replay and the convenience of streaming are not mutually exclusive. Sonos generally beats them all hands-down on the usability experience however, with simple multi-room setup, and a (generally) great user interface. It's one of the reasons I chose it after all. So it's a shame that Sonos doesn't see the need (benefit?) for hi-res support.

What proportion of Sonos users need to get behind the request to make the company reconsider? 

Just as an aside - I think this issue is as much about the convenience of not having to generate multiple copies of files in different qualities just to be able to support different equipment in different rooms, as it is about the actual replay quality through Sonos. I appreciate that the rest of the equipment in a Sonos replay chain may not necessarily be able to make the most of the higher-resolution file, but being able to just have ONE file in the highest quality available, and to allow all kit in the house to be able to handle that would be a massive benefit. So even if Sonos has to downsample for replay, fine, as long as the same file can still be accessed by a (non-Sonos) audiophile streamer in the main listening room and played at full quality. Having to manually make copies of files is time-consuming and can take up a lot of additional storage space on a NAS
Userlevel 3
Badge
At this point the only real issue left is the bit depth which does make a difference in dynamic range. More accurate than 16 bit and a difference many audiophiles can hear.

CLAIM to hear...

The dynamic range of 16-bit is 96dB. This is WAY more range than any recording actually uses, especially when you consider the current mastering practices that massively over-compress the dynamic range of the music. Even non-overtly compressed music is still typically under 40-50dB of dynamnic range. Greater than 16-bit makes sense in the recording/mastering pipeline, but 16-but is perfectly fine for the final delivery.
PLEASE add hi-resolution file support (eg. AIFF 96kHz/24bits) In a world where CDs are becoming obsolete, I believe a good share of Sonos customers are still looking for hi-fi quality, not only the mediocre (format) quality provided by most digital music services/stores. It is rather purpose less to invest in the acquisition and maintenance of such a high quality product as a Sonos system and, not being able to play the best sound quality available with it! Thanks for kindly prioritizing such improvement :o)

Lacking some music formats (e.g. also High Res FLAC) is one of the reasons why i will sell my Sonos products again...
Too expensive for the provided functionality!
Userlevel 1
Badge +1
So my question. I have ~ 500 Hi Res files the Sonos won't play. I can Listen to them in MP3 quality via Apple Music. How ever, If I want to listen to them in CD quality I need to convert them to 44.1KHZ/1411KBPS? If I do this will i have two separate files? what is the best way to achieve this? Thx
24Bit/96khz support should be implemented, hopefully! Most audio manufacturers are providing higher resolution support or have at least announced it. Some new technologies in use like Tidal/MQA looks promising. High bandwidth Wi-Fi has been widely available for a while so everything is in place except no news or information from Sonos. The DAC chips utilized in the Play 3 and Connect ARE definitely capable of 24bit/96khz but the problem may lie somewhere else than the DAC.

At least support hi res support for music that people already own. The higher sampling rate would definitely help to eliminate some of the harshness in the music playback.
+1 for FLAC Support - if my BMW can play it, I don’t see why Sonos has an issue... Disappointing to get an error message saying unsupported for half my library. Definitely holding off on purchasing additional speakers until this is added. Might even consider returning the two I just bought. Alexa on Sonos One is not very stable btw, and half the Alex features are ‘unavailable’. not sure why, is there some special hardware the echo dot has that Sonos can’t put in a $200 speaker?.........

Sonos does play FLAC files:
https://sonos.custhelp.com/app/answers/detail/a_id/80/~/supported-audio-formats

Sonos restrictions apply to the sampling rate and bit-depth. It supports CD quality, because higher resolution than this provides precisely zero additional audio quality (for humans, anyway).


Sorry, I meant 96k... Sonos is the only device I can’t play this on...

It seems we are stuck in an unproductive "do loop" ... Yes, 44.1 kHz is a little over twice the auditory limitation of 22 kHz for the average young human, but that is not the point I am trying to make. Also, the market for audiophile equipment might be small, but those folks do influence others making audio purchases and to the average consumer "more is better".

More bits and a higher sampling rate are part of the belief that "more is better". As a separate note, that "dead" analog technology called vinyl record sales (new only) has gone beyond 10 million units in annual sales and appears closer to 20 million. Why would any "smart" consumer want that?

Back to my suggestion, I am only recommending a simple software "feature" that would skip over requested files higher than 16-bits sampled at 44.1/48 kHz. Seems like a trivial exercise to a "dumb" marketing person.
Userlevel 3
Badge
Audiophiles will use something like Roon to manage their music, and these music management tools also do automatic sample/bitrate conversion on the fly when playing to Sonos speakers. These tools also get around the 64k track limitation that Sonos' native playback system has. And the UI (at least in the case of Roon) blows the doors off on Sonos' UI.
Please add 24bit / flac support in Sonos ☺ A workaround is to play from a PC (Windows 10 built-in music player), then select 'play on device', then select the Sonos player from the PC. The Sonos controller app must not be active for this to work.
In the recording studio there are advantages to using 24/192 (and higher), but there is no proven advantage for distributing more than 16/44.1.

It's a numbers game. 24 MUST be better than 16, right? The golden ear types will always claim that things sound better when they know the music is rendered at 24/192, and it is a great story, but to date no one has been able to demonstrate the ability to identify the 24/192 streams in blind, well managed studies. The results are no better than 50/50.

These studies are very difficult to setup. Many times the 16/44.1 source is simply a CD bought at retail, while the 24/192 file is a special mix and mastering -- and the 24/192 version really does sound better for many reasons. However, if this same 24/192 master is carefully down sampled to 16/44.1, then the "obvious" differences disappear into chance selection by the listening jury.

If you have access to some of the Mobile Fidelity Ultra Disc releases, compare them with regular retail CD's. Even though these two products originated from the same session and master tape, they can sound starkly different. In some cases I had to listen, and re-listen for an excruciating number of trials before I was satisfied that the two releases actually came from the same session. You can experience similar differences when comparing a well mastered 24/192 with its mass release retail version. These differences are not due to the 24/192, more careful attention to detail makes a difference. Really, these products were prepared for different audiences. Retail versions are designed to be relevant in a noisy portable environment while the 24/192 is targeted at audiophiles and better listening environments.

There is nothing wrong with purchasing a 24/192 version of something, then down sampling for the SONOS system. This will give one access to the high quality mastering while still allowing the advantages of practical multi-room wireless music at home. It is much more difficult to stream 24/192 at home. It is unlikely that the home wireless environment would reliably support more than a few streams because of the large amount of data to move around. While the current 16/44.1 SONOS streaming is very robust, there are always a few complaints centered around difficult wireless environments that keep SONOS on edge. In a 24/192 environment there would be exponentially more of these complaints. Since no one has been able to consistently demonstrate that 24/192 rendering offers any advantage, it seems foolish to drag the system reliability down, just to show better, but irrelevant marketing numbers. Look at the competing hardware that can render 24/192. How many simultaneous 24/192 streams can they support? (not many) A SONOS system can support up to 32 16/44.1 streams.

Could we stream higher bit rates? Certainly. If you dig around you can find ways to use networking to stream multiple 4K video sources to multiple TVs. (dozens of sources and dozens of TV's) This represents much more data flow than 24/192 audio. However, everything is wired and this cannot be done with networking equipment found at your local big box retail outlet. Unless you are a network pro during the day, professional installation is recommended. This approach is quite different from picking up a couple SONOS boxes at retail and setting them up in a few minutes to run over the home networking that is already in place.
While they are being silly, I don't understand why the audiophile community is not demanding 384/32. It is out there and it must be better, right?
Lol, just give them some time.
Badge
We use just connects mostly on hard line network. And then send that audio into a URC Total Control System.
So the Hi-Res for us would not present an issue for Sonos.
They are looking for it to work and have not real need to justify the risk of that feature.
Guess I need to down covert the files
Badge
The whole point of higher sampling (96khz and up) was to move the noise generated from digital sampling higher in band... above most humans ability to hear. The reason was at the time not all DAC's were made equal. Many of the methods used to filter out the digital noise caused more problems than they solved. By moving the noise up the band you were left with a cleaner signal. The technology finally caught up so that most consumer DAC's could filter out the sampling noise and sound good with standard 16bit- 44.1khz audio. The only thing sampling has domain over is the frequency range captured. At 44.1 you'l get 0-22.5khz well above the 20khz most systems and even most people can hear. Many enthusiasts will argue that you can hear above 20khz... but this is another topic entirely.

At this point the only real issue left is the bit depth which does make a difference in dynamic range. More accurate than 16 bit and a difference many audiophiles can hear. The issue with applying it to Sonos is: A) do their DAC's support 24 bit playback in the first place. Not sure they wouldn't but I'm not aware of what DAC's they're using. 😎 Adding support to the firmware ($$$) C) Adding support to down sample media to the hardware specs. To point C the big issue here is where the down sampling occurs and is a major software rewrite. Is the feature only supported on desktops or do mobile devices with minimal hardware resources need the feature too? Down-sampling in its own right introduces noise and the methods to minimize that noise (called anti-aliasing) is an art left to a professional sound engineer not automated software on consumer speakers. I'm pretty sure there would be an uproar on the forums if such a system was put in place and everyone would have an opinion on the best solution to that.

I don't think Sonos ever billed itself as a hi-fi platform. They certainly don't compete with the high end audio market. I think for the sake of play-ability it would be wise for them to support 24-96khz as these files are used more often. But to be honest if you're looking for a hi-fi experience with these files you should have a better hi-fi set-up than Sonos.
Badge
I agree that 16bit is fine for listening. But I do have ahi-fi setup and have A-B 16 vs 24 and there is a difference. I'm not arguing that Sonos should support the feature to appease audiophiles. My only point would be to support it for convenience. If you're an audiophile you appreciate 24 bit... and 24 bit isn't just about dynamic range... its dynamic accuracy too. 24 bit recording sound smoother and at 24bit 96khz are considered equivalent to DSD recordings.

Here is something else to consider... I DJ and most high end DAC's have dynamic range of 105db and up. 24 Bit files sound great especially on pro sound rigs and where they really help is in the high end. Things like high-hats, strings, vocals sound clearer, smoother and closer to how they sound on good vinyl recordings. I do believe there is a place for 24bit... Higher sampling I could live without, but since its here and most DAC's already support it, Sonos could easily add support for the file formats in a newer generation of products. Again not for the sake of appeasing audiophiles since its a consumer speaker, but for the sake of convenience.
Badge
OK so if 16 is accurate enough why is there a solution for aliasing and the noise it creates... Subjectivity to what one hears aside, there is more to how noise is generated in a digital recording and bit depth plays into that. Your focus is entirely on the metric of loudness and ignores noise in digital recordings. I'm not sure who you think has proven anything. There's more to a DAC than just its specs, the circuit, the room, everything has an influence in what you hear. If I download a re-mastered album in both 16 bit and 24 from the same release... same remaster, play it on my rig and hear a difference... then I hear a difference. The why doesn't matter. You can't take what I hear from me and this whole argument has nothing to do with the point I'm trying to make. There are 24bit 96khz files, they exists. For Sonos to not support it because the science says there's no difference might be logical, but they might also lose the hearts and minds of consumers shopping around. Sonos has its competitors and most consumers don't purchase things informed...

For me I enjoy the sound of 24 bit recordings. You don't have to believe me I don't really care. When I play festivals and put on older recordings that have been remastered to HD and are amplified with tens of thousands of watts it sounds amazing. You can believe what you want, I know what I hear. Could it be the mastering, or is it the DAC sounding better because it was designed too, or is it the format. IDK I don't care. The why doesn't concern me anymore, if a 24 bit file sounds better to me I'm going to play it.

I'm also not trying to take the fact away that mastering doesn't play a part. I've listened to digital recordings and vinyl and even though digital is far superior on paper I've heard vinyl releases that have sounded amazing vs their digital counterparts and vice versa. There are many variables as to why... and I've spent countless hours recording these things and listening on various DAC's and analyzing the difference... it is a rabbit hole. I don't suggest becoming an audiophile to anyone.
it is a rabbit hole. I don't suggest becoming an audiophile to anyone.


On this we agree. The rest is subjective, except when you make definitive objective statements on the superiority of hires like "24 bit isn't just about dynamic range... its dynamic accuracy too" or "More accurate than 16 bit and a difference many audiophiles can hear." No, 24-bit is about dynamic range and dynamic range ONLY. There is nothing about 24-bit which increases "dynamic accuracy". And no, 16-bit vs. 24-bit, when made from the very same masters is not "a difference many audiophiles can hear." It's simply not so.

As to the noise and aliasing, nobody here is saying there is no use for 24-bit/96KHz in the mastering and recording process. Matter of fact, everyone with any knowledge knows the only use for higher frequency sampling was in the early days of the digital mastering process because it allowed extra padding above the audible spectrum for the anti-aliasing filters to do a more efficient job . This has less than zero to do with playback, and only confused** audiophiles who equate benefits for recording and mastering to be equally beneficial to playback would ever say differently.

**Actually, confusion is being polite. In my experience, most "audiophiles", even after being repeatedly corrected on things like "it's good for recording and mastering so it must be good for playback" or "the higher frequency samples fill in the jaggies in the sine wave" BS continue to repeat it long after they know better. This is not a mistake, it is a willful dismissal of facts. Sony got sued and lost for the same willful dismissal of facts when they used a "fills in the jaggies" diagram in an advertisement. If only there was a way to sue "audiophiles" for the same.
Badge
[quote=jgatie]


Things like high-hats, strings, vocals sound clearer, smoother and closer to how they sound on good vinyl recordings.


So a higher dynamic range makes it sound like a format with a much, much lower dynamic range? Are you sure this is what you want to say?


Yes it is because of the noise that was introduced in the digital format. You can't compare things on spec alone. Even though vinyl has less dynamic range, when compared to a particular DAC I was sampling didn't suffer from the noise that the DAC or circuit introduced and sounded clearer than the digital recording...

Yes it is because of the noise that was introduced in the digital format. You can't compare things on spec alone. Even though vinyl has less dynamic range, when compared to a particular DAC I was sampling had less noise and sounded clearer than the digital recording...


Nonsense. This is audiophile drivel. The "noise that was introduced in the digital format" hasn't been a problem since the early days of the format, when analog engineers were trying to transition to digital and used their old techniques which didn't translate well. You are about 30 years too late with your statements, which is typical of "audiophiles" who keep cutting and pasting the same things for years with absolutely no knowledge about the actual science, or any advances which may have occurred with it.

To quote Monty:

This means we can use low rate 44.1kHz or 48kHz audio with all the fidelity benefits of 192kHz or higher sampling (smooth frequency response, low aliasing) and none of the drawbacks (ultrasonics that cause intermodulation distortion, wasted space). Nearly all of today's analog-to-digital converters (ADCs) and digital-to-analog converters (DACs) oversample at very high rates. Few people realize this is happening because it's completely automatic and hidden.

ADCs and DACs didn't always transparently oversample. Thirty years ago, some recording consoles recorded at high sampling rates using only analog filters, and production and mastering simply used that high rate signal. The digital anti-aliasing and decimation steps (resampling to a lower rate for CDs or DAT) happened in the final stages of mastering. This may well be one of the early reasons 96kHz and 192kHz became associated with professional music production


https://xiph.org/~xiphmont/demo/neil-young.html
Badge
Yes I completely agree. But this is all theory. In practice you can get wildly different results. I have a modern new DAC, I've listen to vinyl, then recorded the vinyl and in theory it should have sounded the same. It didn't. Nor do a lot of recordings. So this is perfected (I think not)... and the more consumer grade you get the more variation. I can listen to one DAC and it sounds tinny and distorted... and I can hear another one in a studio and it sounds great. If all these problems were solved in the DAC 30 years ago why are there still differences... Is it just the DAC, no it could be the circuit or it could be a lot of other things.

Don't hold my descriptions on what I hear to some high and holy white paper on the theory of how things should be.
Badge
Have you heard the turn of phrase "in theory this should work" when referring to the science of something. Also implying that the practice of the science can be incomplete or incorrect. You are wack and I'm out.
Have you heard the turn of phrase "in theory this should work" when referring to the science of something. Also implying that the practice of the science can be incomplete or incorrect. You are wack and I'm out.

Theories are accepted explanations of a phenomena that are proven by rigorous observation, experimentation, and duplication, The question "In theory, it should work" opens up said observations (i.e, the theory is not working) to further rigorous observation, experimentation, and duplication. So far, the theory that 24/96 hi-resolution audio can be distinguished as superior to 12/44.1 has not held up to these rigorous observations, experimentations, and reproductions. Until someone does a rigorous experiment that is reproducible which proves differently, then the current theory stands, regardless of your own less that rigorously attained "theories" to the contrary.
I'm fully aware of bias and placebo effect. That doesn't mean there are real differences out there.

Freudian slip? 😃