As We See It

The Beatles' Last Stand

Jim Austin, Tom Fine  |  Nov 16, 2023

On September 27, 2023, executives from Apple Corps and Universal Music Group held a press event at the Dolby Theater in Manhattan. The event included Dolby Atmos demos of forthcoming Beatles releases. It included some big news—although the biggest news wasn't obvious at first.

The obvious headline: The Beatles are releasing a new song. It's called "Now & Then," and all four Beatles play on it. You've probably heard about it by now, since there's a massive marketing campaign. UMG is calling it "the last Beatles song." For more information, see this article.

Those attending the press event also learned that on November 10, Universal and Apple Records will reissue the Red and Blue albums in expanded, 50th Anniversary form, as 2-CD and 3-LP sets and for streaming, with "demixed" remixes of every song from Magical Mystery Tour backward. (The remixes of The Beatles—aka the "White Album"—and Abbey Road have already been reissued, and some of the Let It Be multitrack tapes and live recordings were "demixed" for the Let It Be reissue.)

For audiophiles, the most important information revealed that day was not really about the Beatles. It came in response to a question Stereophile directed at Apple Corps CEO Jeff Jones. (Apple Corps Ltd., the Beatles' umbrella corporation, is of course not related to Apple Computer's Apple Music, which is also involved in this story.)

A bit of background. Soon after Apple Music's 2021 press event announcing its embrace of Dolby Atmos and "spatial audio" (also lossless stereo, though that was deemphasized), one of us (JCA) poked around to see what he could learn about the technology. JCA learned that Atmos in its lossless, hi-rez "TrueHD" form is capable of excellent technical quality—but Apple Music's streaming version of Atmos is quite lossy, maxing out at a bitrate of 768kbps for loudspeaker delivery—roughly equivalent to one channel of CD-quality audio—and a disappointing 256kbps if you're using headphones (footnote 1). When you consider that Atmos is a multichannel technology—the specification allows up to 128 audio channels for input—you realize how lossy it is in this distribution format.

We're primarily two-channel guys for music listening, but we're open-minded. TF enjoys his collection of four-channel "quadraphonic" recordings. We've both long appreciated the theoretical advantages of multichannel audio and the experience of well-produced multichannel music.

Those lossy-compression rates, though, are scary. If we want to experience these new Beatles records in Dolby Atmos—that's all they played at the press conference—in real high fidelity, where should we turn?

It has been clear for a while to anyone paying attention that elements in the recorded-music industry are pushing hard for Atmos. Audio engineers provide an Atmos mix as a deliverable for many new albums and remasters. Many of those mixes are probably very good, and no doubt they are mixed, mastered, and archived in lossless form—even in high resolution. But there's a serious distribution problem: The only way most people can access an Atmos mix is via a streaming service—mainly Apple Music, although Tidal, Qobuz, and Amazon Music offer some Dolby Atmos tracks. As described above, the streaming version of Atmos ishighly lossy. We've seen very few sources of high-quality Atmos downloads, and anyway, downloads appear to be going away and thus are not the long-term answer. How, then, can we access better immersive versions of this music—better than what's provided by the streaming services—now and in the future?

One of us (JCA) has been asking this question ever since Apple Music's Dolby Atmos debut and has rarely received a straight answer. Jones—the Apple Corps CEO—provided one. In the past, higher-quality Atmos files (and other multichannel formats) were stashed on Blu-ray discs in a few deluxe edition box sets, including some of the earlier Beatles 50th anniversary "super deluxe" boxes. Jones's news: Those Blu-ray discs are going away. Why? Because they raise costs hence the retail price, and, as Jones put it, "very few consumers care." The streaming version of Atmos spatial audio, Jones said, "made the Blu-ray obsolete." Neither the new Beatles tune nor the new remixes will be available in high-quality Atmos. What you stream on Apple Music is what you get.

Jones is probably correct: Few consumers care. Streaming Atmos is good enough for most folks. Older audiophiles have lived long enough to remember previous generations of record executives telling us that no one cares about better sound. We, of course, are those "very few consumers." We do care.

Jeff Jones doesn't speak for the whole music industry. One suspects, though, that the opinion he expressed is widely held, and he seems to be right about Blu-ray discs: They're hardly thriving as a music-distribution format. (We're less sure about movies.) Except for vinyl, physical formats in general are fading (footnote 2). The only thing likely to be left standing is streaming—plus, maybe, vinyl.

How much does this matter? The key thing for us is that stereo versions will continue to be available at the usual high quality, streaming and otherwise. Indeed, recent Beatles reissues have streamed at 24/96.

We suspect—but of course we can't be sure—that Apple Music's lossy Atmos will slowly fade away under the weight of higher production costs, lack of consumer interest, and inferior technical quality in this distributed form. Experience shows that people don't exactly notice a reduction in quality. They simply stop listening.

Jones's comments made one thing clear: For those of us who care about perfectionist audio, Atmos, as conceived by record-company executives, is not the answer. We should hope for its demise.

Footnote 1: See professional.dolby.com/events/dolby-atmos-music-specifications/#gref. Atmos also has other disadvantages. It doesn't "fold down" to stereo very well; a dedicated two-channel stereo mix is superior (especially when it's lossless). And Atmos is proprietary, not an open format. Those who use it must pay royalties to Dolby. If you were opposed to MQA on those grounds, you should, for consistency's sake, oppose Atmos.

Footnote 2: See riaa.com/u-s-sales-database. If anything, downloads are fading even faster.

COMMENTS
Archimago's picture
Submitted by Archimago on November 16, 2023 - 8:10am

While we can appreciate that audiophiles prefer lossless ideally, I don't think it's fair to say everything lossy is a problem. Certainly not multichannel, and certainly not EAC3-JOC 768kbps (with Atmos) which is a very respectable bitrate.

If we are to be honest with ourselves and perform a controlled, blinded listening test using say 256kbps MP3 encoded with a modern encoder like LAME 3.100, I believe that you will find the vast majority of audiophiles will not be able to tell the difference. Some audiophiles might want to believe otherwise because they've experienced 128kbps MP3 back in the day, but in 2023, much has evolved if we go above 192kbps encoding even with the old MP3 codec.

As for Atmos, as I highlighted a couple weeks back, the standards for multichannel mixing are better than what we see with typical compressed stereo these days:
https://archimago.blogspot.com/2023/11/musings-multichannel-mixes-are-more.html

Thankfully, Apple (and others) have standards in place for loudness. Even if one has 2-channel playback, the stereo downmix of these Atmos streams are more dynamic and I believe audiophiles can appreciate the nuances that come through. While ideally we might still prefer to own the lossless TrueHD-Atmos version, the higher quality multichannel mix actually, IMO, typically sounds better than a compressed CD version or the equivalent 24/96+ HDtracks version.

IMO, you're wrong. Atmos and multichannel streaming will not go away. It is one of the ways we'll be thinking about higher quality sound in the future whether reproduced over headphones or multichannel speaker systems. Production cost isn't much higher and in fact a single multichannel mix can be the universal version that's used to create the 2-channel output.

As for MQA. It's not just about being lossy or proprietary. It's that they lied about being lossless at the start. They also implemented highly questionable digital filtering claiming this improved time-domain performance with zero evidence. Then they claimed that MQA-CD with only 16 bits to play with somehow could represent hi-res. And many audio journalists fell for this BS. All the while lossless streaming became commonplace.

There's simply no use for that kind of codec these days, much less promoted by/for audiophiles.

mieswall's picture
Submitted by mieswall on November 17, 2023 - 10:02am

@Archimago: Your post is simply outrageous and of an astonishing arrogance. Self promotion of your website in first place; but mainly full of wrong assumptions (and ethically questionable btw, coming from the one accusing others of lying), all of them based to the fake "tests" you and your pal GoldenSound did about MQA.

1: You have repeatedly blamed MQA because for having "only" 13 bits of data (78 db) in the case of (non true) MQA files of 16/48; or "only" 17 bits (102 db) in fully unfolded files (suggestion: try to understand the patents before commenting them). But those bits are *free of noise*, and it is an average, since by means noise shaping MQA has a higher headroom where the human ear listens the most; it is in the other three bits where all the noise is located. And that is much better than the FLAC you say the format worsens, as its theoretical 16 bits *includes* that noise floor, that in the best case would be about 30 db (4 to 5 bits at least).
But yet, all of the sudden now we know that according Archimago half a sampling rate of Atmos (a truncation much worse than a reduced bit rate) would be harmless…and that even a MP3 wouldn’t be as harmful…

2: Your conclusions are based in just two “evidences”: those tests you and GoldenSound made. Absolutely no other proof . But, here is the thing, *those tests were done not to the true MQA process, but a mere online “translator”*, that by definition is aimed to convert a NON-MQA file in the format, inheriting all of the input faults (like those you purposely recorded in your test files), and not having any of the capabilities of the true MQA studio process. Anyone with a honest, scientific approach wanting to test the format would know this, as it is profusely explained in dozens of papers and articles. And yet, you drove all of your conclusions about the format based in this completely wrong and shameful start up point.

3: And then you judge MQA folks as liars because of their lossless claims (I guess that include a world renowned scientist as Peter Craven, and the genius mathematician of the late Michael Gerzon, the true intellectual fathers of MQA). But when you didn’t have stakes in this game, and measured the true MQA full process, you found exactly that: so lossless is the format that it was almost an exact copy, not of the simple 16/44 FLAC, but of a 352/24 DXD from which that MQA file was made: http://archimago.blogspot.com/2017/02/comparison-hardware-decoded-mqa-using.html
Among what you said then:
“As you can see, below around 60kHz in the spectral frequency plot, there's really very little difference! What this tells us is as I described last time, indeed, MQA "works" as advertised by creating high-resolution output from the MQA file to a certain extent. Even more interesting is that the amplitude of the difference between the actual DXD playback and the software MQA decode is averaging around -70dB RMS right off the bat! Folks this level of difference is low especially considering that most of it is just the ultrasonic noise above 60kHz! “

Or:
“Wowsers folks, below 48kHz, there's essentially no difference between the "Reference" DXD recording and the "Software MQA" from TIDAL. Even the loudest sample with a difference in sound is down at -70dB and on average, the 30 seconds null out at almost -90dB!”

Your very own words….

4: And when MQA (the company) answered your test as you were asking them to do, they explained you these same things: you can’t drive conclusions using an online amateur tool aimed just to convert music FLACs to MQA, feed with wrong data. And that the faulty square waves reconstruction of your tests was due to this reason. But… you published just the introduction of that answer, and not the real content in the appendixes, where they processed your very test signals and achieved a completely perfect square wave, so perfect in fact that I doubt any other format (except DXD) would be able to achieve.
https://bobtalks.co.uk/a-deeper-look/appendix-4-mqa-encoding/#

Then so, who’s the liar?

Archimago's picture
Submitted by Archimago on November 18, 2023 - 9:22am

Yeah, I could provide evidence and arguments for all those points.

But I think we're generally done with MQA, right?

It's dead and deservedly so. It's a commercial failure, having lost tens of millions of dollars I believe in the company reports. This is not self-promotion. It's what expression of truth looks like and I point to articles so as not to repeat myself. At risk of being called self-promoting, here's my take which I think is the obvious conclusion to the sad MQA history:
https://archimago.blogspot.com/2023/09/rip-mqa-codec-yeah-on-lenbrook-groups.html

(I'm all for good tech that could bring something meaningful to hi-fi. So I'm certainly happy to see SCL6 succeed if it's good!)

I'm afraid objective rationales/arguments seem to burst some bubbles within the "traditional" audiophile community more than in other technologically-based engineered hobbies.

Speaking of Peter Craven, has he ever defended MQA against criticism? If not, why not?

mieswall's picture
Submitted by mieswall on November 18, 2023 - 11:19am

"Yeah, I could provide evidence and arguments for all those points."
Oh, that would be fun!. You have built your reputation tearing down a honest effort of really brilliant people, basically citing nothing else than your own "tests". And, modest as you are, you miss that Mr. Craven -the creator of at least 4 or 5 key technologies of modern audio and image processing- stop working in space science to answer your own tender attempts to simulate audio "science".
Seeing now your gymnastic efforts to refute what you said in 2017 ("there's essentially no difference between DXD and MQA" - isn't that almost the same as saying "master quality", btw?) would be... well, more of the same; explaining why to hide the way you have been refuted by MQA (the appendixes above), etc. That will be a show! You build the "truth", you judge who lies (by lying yourself), and now we may have the show of seeing you again de-constructing your own previous "truths".

Yes, MQA Ltd. lost millions and millions. I give you and your pal GoldenSound credit for most of that, when not only hundreds of websites but even insiders as Paul Gowan (shame on him...) cites your completely amateurish tests as evidence of MQA faults. It's hard to battle against lies repeated over and over again, even more when protected by anonymity (btw: are you and GoldenSound really different people?).

Many people think this industry needs to get rid of liars. I agree. You convinced the world that MQA were among them. I think it is exactly the opposite.

Anyway, I think the news of MQA death are greatly exaggerated. We'll see.

SoundGoods's picture
Submitted by SoundGoods on November 19, 2023 - 9:57am

No further need to belabor MQA.

When MQA was new, and bandwidth was relatively expensive, it had a place. Maybe.

Now, bandwidth is cheap and lossless FLAC is the choice of discerning listeners. Full stop.

It's not that MQA was a fraud, or a cool gadget, or whatever the argument is. It's just no longer needed.

Why are people still arguing over this? Nobody "killed" MQA. It died by natural causes.

mieswall's picture
Submitted by mieswall on November 19, 2023 - 10:25am

Imho, it is still a very important development, either if it dies or survives. And I'm not sure they are done yet.

File size is still important for anything above 2X redbook-size. MQA allows you to have files of even 8X packed in roughly the size of a common flac. In fact, as the very Archimago demonstrated, it allows you to stream an almost exact copy of the studio master DXD, a thing still unthinkable without bandwidth optimization.

But the true relevance of MQA is that it avoids the ringing artifacts typical of any brickwall filters present in any pcm file except DXD (necessary to avoid significant aliasing). A full MQA process (specially when used in the full analog-digital-analog chain) is able to avoid it in ways much more sofisticated than any alternative filters available in DACs that can work only in the last step, thus inheriting the time smearing already recorded.

Is it audible? Yes, I think it clearly is (and so believe most of the reviewers of Stereophile or TAS, btw). In fact, some argue that the absense of this time smearing is the only remaining advantage of vynil. People uses to think only in frecuency-domain terms; time resolution is as important, or even more so.

Probably all of this, as you say, it's history. Still, having a real understanding of the technology would be important for the audio industry. And, btw, the one bringing the theme up here was precisely the one obsessed in tearing it down, as you may check in the previous posts.

SoundGoods's picture
Submitted by SoundGoods on November 19, 2023 - 10:42am

"But the true relevance of MQA is that it avoids the ringing artifacts typical of any brickwall filters present in any pcm file except DXD (necessary to avoid significant aliasing)."

How does DXD avoid filter artifacts? Isn't DXD a high-rez PCM file?

rt66indierock's picture
Submitted by rt66indierock on November 20, 2023 - 5:22pm

Can you imagine if Lenbrook had disclosed that it acquired MQA Technology for $75k, SCL6 for $25k and some equipment for $25k from the administrators of MQA Ltd? We were going to find out anyway.

The folks who created and marketed MQA wrote about it in the audiophile press and consumers who like MQA could not convince the requisite number of people MQA had value.

People like me did convince the requisite number of people that MQA didn’t have value. It is no harder than that.

rt66indierock's picture
Submitted by rt66indierock on November 20, 2023 - 5:00pm

The answer to your question is on WBF November 13, 2023. My question to you is why do you consider me natural causes?

Archimago's picture
Submitted by Archimago on November 21, 2023 - 1:46pm

Let's just start with a definition:
Amateur = a person who engages in a pursuit, especially a sport, on an unpaid rather than a professional basis.

Yes, I am an amateur hobbyist who runs a blog and writes about audiophilia as a participant who's interested in getting to the truth about matters as pertaining to the hobby. As someone who doesn't work in this area professionally, nor derive significant income (other than a little bit of AdSense revenue & Amazon gift certs which I roll back into buying music or products), this gives me freedom to test independently and express my findings/beliefs. To be anonymous is simply a right we all can have as participants on the Internet expressing our free speech. Some of what I have said and can say would probably not be palatable for those with commercial interests nor would be expressed in the pages of Stereophile.

I trust that is the reputation my blog has.

In all these years, MQA and their supporters like yourself have not released results that would refute what I've written. MQA had an opportunity to refute the Audiophile Style review article in 2018. That same year, representatives were at the RMAF session given by Chris Connaker and other than ridicule and some faithful supporters banging on tables, there was no attempt at technical, or rational, refutation of the information. Over the years, others on Audiophile Style and GoldenSound have found similar issues about MQA. I have to reference my own work much of the time because the "pros" such as the editors here at Stereophile have not done their own independent research. (BTW, no I am not GoldenSound. I think we disagree on some things if you look at my writings and his YouTube posts.)

I am honored that you think my little blog and comments have had an effect on this outcome for MQA. What if the demise of the MQA codec is just a reflection of audiophiles appreciating the truth when they see it and rejection by the market?

Correct, there's no audible difference between MQA and Hi-Res that I could find in the blind listening test. That's the same result as from the McGill listening study. However, remember that the earliest posts on my blog were of a listening test between MP3 (high bitrate ~320kbps) and lossless FLAC, and the audiophile public could not discern the difference there either! That doesn't mean I advocate for perfectionist audiophiles to just grab lossy MP3, nor that there isn't any objective difference. So what if some can't hear the effect of MQA (subjectively good or bad)? Human hearing isn't always perfect (ie. not everyone has "golden ears") which is why for the highest fidelity testing, objective instrumentation is important. I am totally accepting that not everyone desires "high fidelity" sound either.

You are free to believe what you want about MQA and whether they're liars or not. You are also free to enjoy your MQA-encoded recordings. Nonetheless, the facts remain that MQA is not all that we as audiophiles were "sold", we can disagree with people like the JA's who advocated for it.

BTW: Pre-ringing artifacts we see on the impulse response is not a problem with properly bandwidth-limited material. It's not audible because it's just not there in playback for well-mastered recordings and definitely not an issue with high samplerate material since Nyquist would be well beyond audible. For more than a decade, many DACs already have minimum-phase filters. Note that this is not without imposing a time-domain phase shift and MQA was never needed for pre-ringing anyways. For more:
https://archimago.blogspot.com/2018/01/musings-more-fun-with-digital-filters.html

This is IMO basic stuff that should be written about in audiophile magazines. Don't know why it's such an issue given that there's no great mystery to digital low-pass/anti-imaging filtering.

SoundGoods's picture
Submitted by SoundGoods on November 24, 2023 - 12:11pm

1.) No amount of blogging or nay-saying would cause a legitimately superior audio process to fail. MQA failed because there was no longer a need for limited-bandwidth transfers. Lossless FLAC (et al) were embraced not because MQA was flawed, or some blogger said this or that, but because bandwidth got cheap. I didn't know Lenbrook bought MQA for $75,000. That seems like a steal given that there are still dozens of DAC makers paying MQA royalties (?)

2.) Arcamigo is correct. Ringing artifacts are simply not perceived on properly filtered conversions of 88.2 or higher. And on 44.1, OK, maybe a rare artifact, on rare program with a properly-trained golden ear listener doing fast (no gap) ABX comparison on a pristine headphone system. And, even then, probably >98% of sources will be a null. Yes, we've done tests.

3.) Journalism with paid advertising has limitations on what it can say, who can say it, which products get reviewed and talked about, etc.. That's not a slam on for-profit journalism, it's just a fact of life. I think Stereophile does a really good job, considering their fiduciary constraints. MQA was used by many, many Stereophile advertisers. QED.

mieswall's picture
Submitted by mieswall on November 26, 2023 - 10:59am

"In all these years, MQA and their supporters like yourself have not released results that would refute what I've written."

Twisting the facts once again Archimago. The impressive refutal came from the very MQA ltd (and you or your pal GS hidden it). Although in strict sense, there is no refutal needed of your tests, because neither you or GoldenSound has ever tested the real MQA process. And please be serious and don't cite other websites other than yours, because all what they have done is republish your comments and graphs, without a single other proof.

What you both tested was a MQA translator . And knowing in advance the limitations of that tool (because of its very focused purpose), you prepared a test thought to find the faults, not of MQA, but of that online translator provided for indie musicians.

You DID made a kind of proper test to the real MQA in 2017: comparing the music content (a DXD master, nothing less) with the outcome provided by MQA. And YOUR FINDINGS were that they were almost identical. Let me repeat this again: YOU TESTED A DXD MASTER AGAINST ITS MQA FILE, AND FOUND THAT BOTH WERE VIRTUALLY THE SAME . You also found that no other pcm format was able to do that. And this is the real test to be done, as this is an algorithm specifically designed to process music, and not photos sent from Mars.

Those findings were not from Bob Stuart, or Jim Austin, or Jason Serinus, or Robert Hartley, or Michael Fremer, or any other MQA advocate (of whom you have the guts to doubt of their honesty). They weren't also from dCS, CH, Weiss, Mytek, Meitner or any of the world-class DAC's that still provide MQA capabilities. They were from... Archimago.

All what you have done afterwards, including all of your posts here are, in my humble opinion, profoundly dishonest; you perfectly know you have built a lie (and hidden the true facts in the process), and have endlessly profited from it ever since.

End of discussion.

cognoscente's picture
Submitted by cognoscente on November 16, 2023 - 10:53am

yeah yeah, of course, I'm going to buy all these songs for the fifth time, sure, why not?

Archimago's picture
Submitted by Archimago on November 16, 2023 - 11:52am

The multichannel versions are not just remasters of the same recording with just a bit of EQ here or a level change there.

They are remixes that change the sound of the parts and the 3D soundscape significantly. Some might like it more or less, that's subjective, but there's no denying that they are very different - especially when heard in immersive/surround playback.

Joe Whip's picture
Submitted by Joe Whip on November 16, 2023 - 11:04am

For this listener. When done right, Atmos music provides a listening experience that no 2 channel system can match IMHO. Evelin better, you can get those results for far cheaper in terms of overall cost than a high end 2 channel rig, and you get movie es to boot. Not all Atmos mixes are great but the same has always been true for stereo. Listen to the Atmos track on the 4K Blu ray release of Bob James Feel Like Making Live. It is simply a thrilling listen.

JRT's picture
Submitted by JRT on November 16, 2023 - 12:51pm

The blockquote below is an advertisement copied from page 24 of the September 1981 issue of Studio Sound magazine.

Encoder completes UHJ system

You may already know that the Calrec Soundfield mic is somewhat unusual in that it can be used as a stereo microphone in which such parameters as distance from the source, polar diagram and orientation may be varied after the recording - the signal from the mic may be recorded, and then modified later with the control unit provided.

A new unit from Calrec now makes it possible to utilise the microphone for its prime purpose - originating material for Ambisonic surround - sound reproduction. The Soundfield mic enables four channels of Ambisonic information to be recorded: this 'B- Format' signal consists of an omnidirectional mono signal, W, and three difference signals, left minus right, front -back, and up -down, referred to as X, Y and Z respectively. The new Calrec Ambisonic UHJ Encoder takes three of these signals -excluding the Z (height) information, leaving horizontal surround information only -and encodes them into the 2- channel UHJ format which is compatible with stereo and mono replay, yet reveals full horizontal Ambisonic surround information when the encoded material is repro- duced via a suitable UHJ decoder. Four UK companies already, or are about to, market domestic decoders: Minim Audio, IMF Electron- ics, Meridian and Integrex. Minim have also announced a professional Ambisonic decoder for studio and broadcast monitoring use.

The Calrec encoder is a small rack -mounting unit with no controls other than a power switch. It operates at standard line levels, requiring a 3- channel B- Format input and offering a 2- channel encoded output suitable for recording, direct disc -cutting or broadcast operation.

Use of the encoder is not limited to the Soundfield mic as a source: any B- Format signal may be used, allowing the use of the mic for Ambisonic capturing of the soundfield at a live performance, or the encoding of an artificial Ambisonic soundfield created by mixing down a conventional multitrack master via a mixdown console equipped with B- Format localisation controls (' Ambisonic panpots').

Calrec Audio Ltd, Hangingroyd Lane, Hebden Bridge, Yorks HX7 7DD, UK. Phone: 0422 842159. Telex: 51311.

13 months later, on October 01, 1982, Sony released their CDP-101, and that was the first CD player released for retail sales.

UHJ could be decoded to as few or as many discrete loudspeaker channels as desired using analog filters, without DSP, or could be decoded in digital representations of those same transfer functions in DSP, as DSP later became more widely available at lower costs.

So UHJ encoding could have been implemented for broad consumer use in early CDs and CD players.

JRT's picture
Submitted by JRT on November 16, 2023 - 7:24pm

DVD video discs and players first became available in the commercial retail market in Japan in November 1996, and in the US in March 1997.

The linear PCM digital audio portion of the DVD video specification accomodated up to 8 channels (more than 6 was unusual) of either 16_bit or 24_bit dynamic range at 48_kHz or 96_kHz sample rate, in various combinations up to a limit of 6,144_kbit/sec.

Notably missing was the option of 64_kHz sample rate, which would have allowed the option of 4 channels at 24_bit at 64_kHz, at the limit of 6,144_kbit/sec. And that 4 channels would have accomodated full periphony (3D) UHJ Ambisonics, which at playback could have been processed to as few or as many discrete audio channels as desired for width, depth and height.

Included in the initial release were 8 different variations of media, including two diameters, 12_cm and 8_cm (mini-DVD), single side and double side, single layer and double layer. Note that mini-DVD didn't gain traction as a consumer audio medium, but would have fit in a shirt pocket making it that much more conveniently portable than the standard 12_cm CD which does not fit in any of my shirt pockets.

A double side single layer 8_cm mini-DVD has capacity of 2.92_GB, for more than 63 minutes at 6,144_kbit/sec, 105% of one full hour. And as compared to Redbook 44.1_kHz sample rate, the 64_kHz sample rate would shift linear distortions associated with anti-aliasing filters to higher frequencies, more than 45% higher (64/44.1).

To me it seems like a missed opportunity, full periphony high resolution Ambisonics in a disc that would fit in a shirt pocket in the mid/late 1990s.

SACD and DVD-A came later, and at 12_cm did not have the conveniently more portable size of 8_cm mini-DVD, could not improve on the dynamic range of 24_bit audio in DVD-V, and did not provide full periphony Ambisonics.

judmarc's picture
Submitted by judmarc on November 17, 2023 - 1:59pm

Not all that many years ago, lossless and hi res two-channel streaming didn't exist. Now we're almost spoiled for choice. It seems foolish to bet against the same thing happening with multichannel.

So rather than hoping for multichannel's demise, perhaps we can continue hoping and pushing for lossless and hi res in that new arena. As streaming speeds inevitably increase, I think we will eventually have it.

jimtavegia's picture
Submitted by jimtavegia on November 19, 2023 - 8:43am

It has everything to do with a gimmick and not taking full advantage of the high bit rates and sample rates that are available while we WAIT for streaming to get better. I am happy for those who can afford to run their streaming through a DCS product.

The analogy of people loving 4k and 8K video, and improvement in picture quality is obvious for sure, yet folks think that a SACD player is a bad investment. This only proves that people are more visual I guess. This is comparing the price of a Big 8K TV to an SACD player.

I have tried Atmos as some of the available streams are in Atmos and easy to try and are fine. This is still the industry choosing a gimmick over maximum bit rates and sample rates. We have to live with what they give us.

The fact that you chose not to do vinyl I would guess is a matter of convenience over ease of access to streams. Everyone gets to pick their preferences. Here we have the industry denying us access to better sound and choosing to do something less.

My newest CD only player as a USB port on the front for a thumb drive and the industry could deliver albums on a thumb drive up to 24/192. That will not happen ever I am sure. I just bought a Yamaha CD S-1000 SACD player that will be my last one. My old Yamaha S-1800 still sounds great playing cds and SACDs.

It is all marketing now and not all science. We went from 7.5 ips to 15 ips on tape and then masters done at 30 ips and then multi-track. We have LPs that are cut at 33.3 and 45 rpm, all an effort to improve quality and enjoyment. My point is that we are just not seeing that same effort for music on the digital side where convenience reigns.

Joe Whip's picture
Submitted by Joe Whip on November 19, 2023 - 1:26pm

I prefer the sound of digital for many reasons, having heard studio master tapes. Atmos when done well is amazing offering a soundstage that no 2 channel rig can match IMHO.

rt66indierock's picture
Submitted by rt66indierock on November 20, 2023 - 4:42pm

If we learned anything from the Mobile Fidelity class action lawsuits is the audiophile market does not move the needle. ATMOS will live or die based on ordinary people finding value in it.

PeterPani's picture
Submitted by PeterPani on November 20, 2023 - 11:26pm

with higher carrier frequency and audio only. Such an analog format would lift analog reproduction above vinyl and reel to reel and every digital format, for sure.

Archimago's picture
Submitted by Archimago on November 21, 2023 - 1:40pm

Yeah, probably would be better than LP.

The analog audio tracks on an LD uses CX noise reduction and provides I believe around 80dB of dynamic range. Better than most analog but not an improvement compared to digital, even 16-bit.

PeterPani's picture
Submitted by PeterPani on November 22, 2023 - 3:25am

had to tackle with the problem of separating video and audio tracks and overlapping interference. An analog audio only Laserdisc could be made with a much much higher carrier frequency and so become easily the top analog carrier of our time. With no new analog media technology around we get stuck with vinyl and tape. a pure audio Laserdisc would open a window for future development of analog.

jimtavegia's picture
Submitted by jimtavegia on November 21, 2023 - 2:53am

The owners of the media are turning the music business into the cellphone model of subscribers, except no one is giving away free streamers where cell phone companies will give away phones for contracts. I suppose this is the point of making music access more affordable. The WiiM Pro will help with that I guess.

I will be leaving this alone from now on as there is not point to it all. Many SACDs I would like are often sold out. The LP equal is $150 in a Deluxe pressing, $20-$50 regular. We will all just pick what we like and let it go based on what the labels offer.

The former Dolby products from B, C, S, and A for the pros all served a purpose in removing tape hiss from our decks and master tapes, which was greatly needed. Dolby Pro Logic served the home theater market well, as will Atmos I am sure.

Sal1950's picture
Submitted by Sal1950 on November 22, 2023 - 2:25pm

"Jones's comments made one thing clear: For those of us who care about perfectionist audio, Atmos, as conceived by record-company executives, is not the answer. We should hope for its demise."

Rather than wish for a SOTA lossless Atmos stream as we did for many years with 2ch, Mr Austin wishes for Atmos demise. :(
What I would wish for is Mr Austin's position at the helm of Stereophile to have it's demise and someone more akin to J. Gordon Holt to once again take the reins, he was a man with a deep understanding and appreciation for multich music production. Mr Austins view of SOTA audio is way too narrow minded for this job.

DH's picture
Submitted by DH on November 22, 2023 - 5:26pm

Jim Austin -
Do the ATMOS versions streamed to an ATMOS system sound good or not? If they do, what's wrong with it? Many people think they sound great and add something to their listening experience that hi-res stereo versions don't.
Streaming lossless might be very slightly better. Then ask for that instead of wishing that a useful format will disappear.

Joe Whip's picture
Submitted by Joe Whip on November 22, 2023 - 6:42pm

Can and does sound fantastic. There of course are some dogs too, but so what. Nothing new there. My listening tell me true HD Atmos sounds better than the DD+ streams but to wish for the demise of Atmos music is absurd. Does anyone going to a good movie theater wish for the demise of the Atmos tracks heard there because they are MC 24/48 and not 32/384? Of course not. I just do not get much of the audiophile “press “ with views like this piece. Utter foolishness.

barfle's picture
Submitted by barfle on November 23, 2023 - 7:39am

At least, that’s how I see it. Streaming is how a lot of “new” music is discovered, which is what us geezers used to use top 40 radio for. And streaming is one way I find new (to me) music worth purchasing. For that purpose, it doesn’t need to be particularly high quality (remember AM radio?).

I knew lots of kids who were perfectly happy to listen to whatever was on the radio, and never bought a record in their lives. Since I’m reading this forum, it’s obvious I’m not one of them. I’ve been collecting audio media for about 65 years, and I still do that, although a lot of what I “discover” is long out of print and commands prices like a Fatboy tonearm.

True, most music companies care for their bottom line means their focus isn’t on premium quality physical releases, although Rhino and SDE seem to be announcing new re-releases almost weekly. I will continue to support those physical releases whenever I can, because it’s important to me.

Rossini100's picture
Submitted by Rossini100 on November 26, 2023 - 2:51am

We should hope for its demise? Why, because it is not lossless? Apple has progressed from selling 128kbs MP3 files with DRM to offering true high resolution streaming for not much money.

In time, I am sure the Atmos offering will develop. It is pretty much inconceivable that it will be identical to as it is now in twenty years time. So why not hope it develops in a manor favourable to audiophiles, rather than simply hoping for it's early death? In time, it may develop into something truly wonderful. This is far from inconceivable, as it is pretty good right now.

X