Audio Myths – Benchmark’s John Siau Takes on DSD
You don’t have to drill very far to realize that John Siau, the chief designer at Benchmark Media, is a very smart guy. Over the years since I first met him, he has become a friend and one of my goto guys when it comes to the technical merits of this or that system, format, or accessory. He’s not a marketing guy, he doesn’t spin facts to his advantage, and he doesn’t need to write articles on the things he knows about but he does. And thank goodness, he does. Because the audiophile world needs more technically capable people writing clear and concise papers about things that make a difference in our listening pleasure and the pursuit of ultimate fidelity.
There are far too many writers making “subjective” pronouncements about equipment, recordings, tweaks, and technical issues and not enough competent, experienced technical authors sharing their expertise to balance the scales. I know a number of these people and they just sit back and laugh at the stuff that appears online and in the audiophile publications.
This myth goes something like this:
“DSD provides a simple and direct digital path between the A/D and D/A.”
“DSD is simpler than PCM.”
“DSD is not PCM.”
Anyone who’s read the “Reality Check” article on the front page of this website knows that John is not a supporter of the “Direct Stream Digital” format (You can join the over 10,000 readers of that piece by clicking here). I happened to notice a ping back email the other day from the guys over at Computer Audiophile linked to a discussion about an a new article that John posted on the 27th of August. The title of the article is, “Audio Myth -“DSD Provides a Direct Stream from A/D to D/A”. Perfect.
Needless to say, some of the “experts” over at CA has some pretty harsh things to say about the piece. They claimed that his claims weren’t true and that John was doing some heavy spinning in favor of his PCM-based DACs (which also is a first class 1-bit DAC for DSD lovers). DSD advocates simply refuse to grasp that it is never going to be a mainstream recording and delivery format.
One thread of the discussion focused on the fact that there are now available tools to do all of the needed postproduction processes (editing, fading, EQ, reverb, compression etc) on quad DSD staying in the native 1-bit format. I can’t say definitively that such a system doesn’t exist but I’m highly skeptical. The leading system for working in DSD is the Pyramix system from Merging Technologies and they chose to convert everything to DXD (which is high rate PCM) to do all of the postproduction processes. I was encouraged that CA founder and editor Chris Connaker had issues with the “compete post production system” claim by trying to describe the functions that audio engineers need to be able to do with raw recordings. Tascam was mentioned but I haven’t heard anything about a set of tools for working in native 1-bit DSD. I promise I’ll take a look.
You absolutely must read John’s well-written article on why the claims about DSD are pure myths. He careful to say that there are some really wonderful sounding DSD recordings but they are great sounding recordings because of the production methods of the team of producers and engineers and not the characteristics of the format. His position is one that I share. I think my own AIX Records tracks sound they way that they do because of my recording style and less because they are made using 96 kHz/24-bit PCM. The sound of the Christian Jacob “Beautiful Jazz” tracks coming from analog tape sounded really good yesterday. But don’t even think about asking for any additional copies unless you’re will to spend a ridiculous amount of money.
Here’s the link to the article, “Audio Myth -“DSD Provides a Direct Stream from A/D to D/A”
Please don’t forget to check out the “Music and Audio: A Guide To Better Sound” website. Click here to learn more and leave your email to be kept up to date with the approaching campaign.
29 thoughts on “Audio Myths – Benchmark’s John Siau Takes on DSD”
Yes the “CA” club and their poster child Auralic products. There’s a couple bloggers there that make sure they “beat” their pseudo science till all the sheep fall in submission. It’s laughable how cults grow and even more laughable how often they switch gears based on the latest and greatest. Once “invested” the sheep must manifest stories to justify their endless purchases. “Merry go rounds” are expensive, who could blame them, poor things.
Yes the core group there have drank the kool aid and will attack any one that doesn’t play march to their drum in a very strong and nasty manner.
The most vocal supporters of whatever technology on CA are all selling products based on said technology, so their attitudes are hardly surprising.
That’s unfortunate. Although, I must say that Chris was politely challenging the statements about the full editor.
Chris doesn’t appear to believe in the miracles of DSD, so he’d be challenging of that. He may be a follower of some other type of mysticism, but he’s generally polite.
I find it amusing how any discussion about DSD/PCM always turns into some kind of no true Scotsman argument. “Oh, it’s not PCM if it has only a few bits and you don’t downsample.” “Ah, but *that* kind of editing is unnecessary.” Etc.
I hope that during one run of the analog copying, you also made a PCM copy as well, for release digitally!
As accurate copies of the first generation masters, it’s arguable whether they qualify as first or second generation – since their inherent accuracy makes them better than second generation tapes.
First and a half generation?
I did make a PCM copy.
WHAT, no DSD? 🙂
On June 17, 2015 Tascam released a free Hi-Res Editor that supports both PCM and DSD, but based on reading the user’s manual, the editing functions are extremely limited. Besides changing sampling rates and conversion between DSD and PCM, the only editing functions are extracting a portion of a file, combining files of the same format and short fades. Here’s a link to the user’s manual, all 14 pages of it.
Thanks for saving me the time Mark. I figured as much…the advocates for DSD don’t want to acknowledge that you just can’t do anything with 1-bit DSD. My point is why bother? Other than extract additional revenue.
“There are far too many writers making “subjective” pronouncements about equipment, recordings, tweaks, and technical issues and not enough competent, experienced technical authors sharing their expertise to balance the scales.” Amen.
Yes, there are. The subjectivists will even begin attacking the idea of science if it doesn’t support their beliefs.
I met Chris at a Music Matters event at Audio Advice in Raleigh NC a little over a year ago and asked him directly about DSD and he is not supportive and doesn’t think it’s viable over PCM was my take. At the time I was exploring DSD for knowledge since all the buzz was ramping up. I have a few SACD’s that I play on my Oppo 103 and can hear no advantage over the same disc red book rip to my NAS streaming via my Naim NDX (which does play DSD 64 via DoP). I have only free samples in DSD and it’s good but the next holy grail it’s not in 64 to my ears. I agree with the notion it’s another push to repurchase in a different container as Mark says. The PCM technology is fine. As many, I have CD’s as well as LP’s that were recorded and mastered properly and are stunning sonically and fun to listen to. That is where the focus needs to be and not on the medium. I have about 1000 LP’s that I never sold and a very good TT and do enjoy the good ones and those from RR, Sheffield etc even with the known limitations.
Mark’s take here (as with his appeal to Authority [as long-ago deconstructed by C.S. Peirce]) is the same here as in all his online offerings. “DSD is No Good”, “My PCM recordings Sound Great”, repeat.
Yet, these DSD-bashing pieces are exactly the same thing as the Auralic vested-interest over at CA that you are so critical of, and you entirely miss the point of support and interest in DSD. So what if you can’t de-ess, merge, fade, bounce, compress, etc., and all the other actions that the PCM Loudness mess that recorded music today passes for today allows? Answer me this. How many of the albums that people treasure were recorded digitally?
Sgt. Pepper? LZ IV? Ziggy? Five Leaves Left? Foxtrot? Unknown Pleasures? Reggatta de Blanc? Dark Side of the Moon? Tubular Bells? WYWH? Dire Straits? For Your Pleasure? Remain In Light? Crime of the Century? etc., etc. In fact, you can insert almost anything by the artists that sell shed-loads of product and it was all recorded analogue — and it is this that people (audiophiles/music lovers) are interested in, not some futzed-with PCM mess. The stated simplicity in the DSD-chain is taken advantage of by those companies who transfer flat from the master tape straight to DSD, such as Universal Japan, and here the transparency of the system is palpable.
The Point is that the vast majority of DSD supporters are music-lovers who find in these DSD releases a way to hear their music with a degree of fidelity / absence of fatigue long-lost since the demise of the LP, which remains the acme for those with ears to hear.
Of course, there is little point in telling this to MW as he is one of those with an business interest in the opposing school who believe that CD is superior to vinyl.
Fortunately, not all of our Ears are cut from the same Cloth…
Buy yourself a copy of ‘Flowers of Romance’ by PIL on SHM-SACD and compare it with a CD of any vintage — it sounds like analogue, and that is entirely the Point.
I can appreciate that many people, including yourself obviously, enjoy the classic recordings of the analog tape era. They were done using analog tape because that was the state of the art at the time. I recorded using analog tape for 25 years! However, great sounding recordings didn’t cease when PCM digital came along. Yes, it took some considerable effort to get it right but PCM digital is the worldwide format used to make virtually all released recordings.
“So what if you can’t de-ess, merge, fade, bounce, compress, etc…?” If it’s not possible to work with a format during a post production stage of a project, it places an unnecessary and artificial burden on the artists, producers, and engineers working on it. You’re right that many projects use too much processing (analog and digital) and make records that sound crappy. This is not the fault of PCM…there are plenty of crappy sounding DSD recordings as well.
How many “treasured” albums were recorded digitally? The question doesn’t really apply…I’m more interested in advancing the fidelity of recordings. But certainly one of the most cherished audiophile recordings of all time is the “Famous Blue Raincoat” by Jennifer Warnes that just happens to have been recorded digitally using standard-resolution PCM.
High-resolution PCM is not “futzed-with” any more than any other format…in fact, it has the potential to eclipse every other format currently available. You’ve undoubtedly heard of DXD? It’s PCM used as a master format from which other formats are sourced. This includes DSD and hi-res PCM.
The insistence that analog and the vinyl LP are the “acme” of the science and art of recording is a myth. I present factual information about audio engineering, production, formats, and fidelity. If you are among those that believe vinyl LPs are have greater fidelity than the potential of a well done CD, then continue to enjoy your DSD files and vinyl LPs. Everyone is entitled to their personal preferences. But you’re factually incorrect…the specifications don’t support your position.
I don’t want my music to sound like analogue…I want it to sound like live musicians.
Great reply, and let me add that for a serious disccusion (not only DSD vs PCM or even analog vs digital) based on technical facts, you need to address scientifically minded people; otherwise you end up talking to “believers” in those absolutely abstract and subjective concepts like “the soul of vinyl” and BS like that.
davip, reading your post, you seem to be particularly irked that the truth about vinyl was outed 30 years ago, and after fighting that battle on the wrong side for so long, you now find the truth about DSD is being outed, and still you find yourself on the wrong side. How annoying!
But, I must say, such is the fate that awaits people who have so prejudiced themselves against PCM digital that they cannot knowingly listen to well mastered PCM audio without, ahem, ‘hearing’ every ‘issue’ that has been fed into the discussion space by people who these days might be described as the Vinyl Mafia and the DSD Mafia. LOL
Bottom line: any and all of those precious analog master tapes can easily and efficiently be ripped ‘unedited’ into PCM audio and will be audibly indistinguishable on playback from the original master tape. Or from a straight DSD rip. They will have 100% ‘analog sound’, because the digital information container is bigger than the analog information container (tape or direct cut vinyl), and simply ‘contains’ it and all its audible properties. There is no ‘digital sound’ with modern lossless techniques, so all you will hear is analog sound.
Thanks Grant…well stated.
DSD or PCM bashing, the former apparently being a hallmark of your blog is, IMO, both stupid and unhelpful.The choice of format is an artistic decision made by the producer. Outstanding recordings can be and are produced in both formats. The paying consumer should be able to choose either or both formats, according to his or her preference..You state that your revered guru, John Siau, whose anti-DSD stand you are so happy to share and publish, “doesn’t spin facts to his advantage”. Yet, in a remarkable show of hypocrisy, he requested to be included on a panel entitled, “DSD Downloads, The New High-Resolution Standard: A Major Update”, at RMAF 2013, where his anti-DSD posture was exposed by a member of the audience. Not surprisingly, he was not invited to return for a similar panel at RMAF the following year.
Robert, you’re obviously a person that believes that DSD is worthwhile and you’re certainly entitled to your opinion. This blog presents factual information about a variety of topics…and the myth of 1-bit encoding is one of them. Just today, Jason Kennedy of HiFI+ wrote the following in a review of the Wadia 321 digital converter, “Editing is ‘difficult’ in DSD, which is part of its appeal to the purist of course, but in practice it means that any commercial release on the format will have been in PCM at an early stage of its creation. Moreover, DSD can have high frequency noise issues, and even if the actual noise is out of the audioband, as high sampling rates have shown us, what goes on up there can be heard by mere mortals who have maximum 20kHz high frequency hearing. The reason it doesn’t offend many is that it is harmonious, and a bit like the harmonic distortion of valves; some feel that it adds to the end result.” There are serious problems with the format. The format used by virtually all engineers and producers is PCM…not because of the sound but because it allows them to actually produce commercial products.
IT has been shown that listeners can’t tell the difference between a native DSD recording vs. a high-res PCM recording. Maybe you and others are the exception to the rule. Choose what you like but realize that there are very few native DSD recordings…less than 2500. The rest are been through PCM stages or originated as analog tape.
John Siau is a world class designer of analog and digital electronics for audio capture and reproduction…he is in fact an acknowledged expert…a bona fide guru. I don’t know where you got your information but John was invited by the moderator to be part of the RMAF panel. I’ve been on a panel with that group as well…at their invitation. It’s laughable that they titled their panel “The New High-Resolution Standard”…it’s certainly not true in the music business at large. It is a niche format in a niche audiophile market.
The DSD mafia is alive and well…that will not change. However, it’s important for consumers to get the truth about the format, the spin, and the misinformation being spewed by those who bought into the Sony initiative. Bob Stuart, John Siau, Jonathan Reichbach, and many other very smart people understand that PCM is a much better format for recording and reproducing audio. DSD will remain probably stick around but will never replace PCM.
IMO, your blog is the oracle of misinformation. Statements like, “It can be shown that listeners can’t tell the difference between a native DSD recording vs a high-res PCM recording” fit into the same category as, “It can be shown that people can’t hear differences between cables”. Experiences and opinions differ and “what can be shown” regarding controversial issues in audio is too often the position of the proponent. Can’t you see obvious contradiction between your statement and your assertions regarding the audibility of the effect of high frequency noise?
What is most puzzling is why you feel the need to continually bash DSD. It can’t possibly have escaped your notice that whereas DSD capability in DACs was rare only several years ago, virtually every DAC on the market now supports DSD. Ironically, Bryston’s Jerome Tanner used to be a fellow member of the DSD bashing club. Curious how his tune has changed since Bryston proudly announced its new DAC that supports DSD256.
BTW, my information regarding how John Siau came to participate on the 2013 RMAF panel comes directly from an unimpeachable source. What is both laughable and hypocritical that John Siau would even present himself to the public on such a panel. Integrity apparently doesn’t hold the value it once did in certain quarters.
I have nothing against PCM. I don’t consider DSD to be the holy grail. What I and most other audiophiles want is to be able to purchase our music in the format that sounds best to us, regardless of whether that format is PCM or DSD. For us consumers, your conclusion that DSD will never replace PCM is totally meaningless..
Robert, if you regard the information presented in my blog as “misinformation” then perhaps you should cease reading the articles I write. If a rigorous study done in 2004 and written up in the AES Journal doesn’t measure up to your standards, then OK fine. The study determined that advocates like you and others couldn’t tell the difference between two recordings done at the same time through the same microphones in PCM and DSD. Go figure.
Maybe sharing a discussion that I had with an audio professional might help clarify my reasons for advocating for real high-resolution audio. My friend asked why I bother to record at high-sampling rates when the content that is captured is beyond the “audio band” of human hearing. I told him…and have written repeatedly in the articles on this blog…that it doesn’t matter to me whether we “hear” the ultrasonics or not. They are present in the room where the musicians were performing so in the interest of fidelity, why not capture them and reproduce them? It’s not difficult or expensive. And we have technologies that can deliver them. That’s good enough for me and certainly not a contradiction as I see it. You obviously think differently.
I don’t bash DSD or its associated 1-bit format. I have presented factual information about its limitations (the inability to do any post production in the native format for example) and believe as others do that it is a seriously flawed format (even Sony finally gave up on SACD after spending $750,000,000). If you buy into the DSD format and believe the information presented by 1-bit advocates, that’s certainly your right. You’re not alone. Perhaps you prefer the sound of DSD…you’ll get no argument from me. But you should know that it’s not the DSD format that created that sound.
Makers of high-end DACs…including those designed by people and made by companies that know the realities behind the format…have no option than to include and endorse a format that is heavily marketed and reviewed (quite falsely) as somehow “closer to analog”, whatever that means. James Tanner, John Siau, and Robert Stuart haven’t changed their beliefs, they simply understand the realities of the audiophile marketplace. I’ve had the discussion with each of them…DSD is a market reality but a sonic and theoretical mistake.
Having a panel extolling the virtues of only DSD or SACD is where integrity can seriously be questioned. In discussing high-resolution audio, it seems more than a little suspicious that show organizers (under pressure from DSD advocates) would continuously present a panel that presents only one side of the issue. Where’s the equivalent panel on the advantages and fidelity of PCM? There is nothing “laughable or hypocritical” about inviting John to be on a panel about DSD. He’s designed one of the best implementations of DSD that you can buy.
No one is stopping you from investing in the format you believe sounds best…least of all me. However, it is important for purchasers of music to know what it takes to produce the content they purchase. Despite your comments to the contrary, DSD is a very rare format in the music industry and will remain so…if it survives at all.
You are presenting only a portion of the facts by looking at things from a recording perspective only, while ignoring playback considerations. More specifically, the vast majority of today’s DACs employ delta-sigma converters which convert PCM to DSD internally prior to analog output. This requires additional processing of the data compared to data provided in native DSD format. Such additional processing can negatively affect sound quality and is the reason that many people use software to convert all music files to DSD prior to input to delta-sigma DACs. In fact, a not insignificant number have found that conversion of all files – PCM, DSD64 and DSD 128 – to DSD256 prior to input to DSD256 capable DACs provides the best sound quality, The latter is, of course, hardware/system dependent.
Your comment regarding DSD being marketed falsely as “closer to analog” is your opinion and should not be presented as fact. Ed Meitner, Andreas Koch, and others would strongly disagree with you from a technical rather than commercial viewpoint.
The recording process is where fidelity is established…there is not enhancing it after the fact. You can change the fidelity to your own liking. You’re right that delta-sigma conversion is the standard way of converting analog to digital and digital to analog. But contrary to your claim, DSD goes through conversion as well. The native stream is not what you hear from your DAC. Conversion to DSD at any rate may result in a sound that you prefer…your own “best sound quality”…but any conversion diminishes the accuracy of the original recording.
The marketing of DSD as almost analog is a myth, the result of marketing hype by those you mention and other manufacturers. It’s time to move on Robert.
“It’s time to move on Robert”.
I couldn’t agree more. And that includes your anti-DSD evangelism.
You can certainly start you own blog and evangelize for DSD all you want. Letting my readers know the truth about the format is important and appreciated…the truth isn’t always in line with your opinions.
And your opinions aren’t always in line with the truth. Bye.
That’s why I’ll be sure and let you know when I offer them. However, facts speak for themselves.