The Continuing Myth of DSD
Let’s start with a few things I read online concerning DSD and DXD (which an alias for PCM):
“With a resolution up to 256 times better than CD, your ears will immediately realize the full potential of your playback system. DSD 256 is the highest audio resolution format available today.”
“Direct Stream Digital (DSD) actually has been around for a while already, but it has been so married to a physical media, SACD, that it has yet to receive the attention from audiophiles that it deserves. It is only recently with the growing interest in downloading high resolution audio via the Internet that DSD surged to the surface of news coverage. What were compelling reasons to use this encoding scheme for SACD over 10 years ago are now becoming convenient truths for the new era of high resolution internet audio.”
“A digital recording at 352.8 kHz sounds closer to “real analog,” which needs no anti-aliasing filters at all, and the listener is better able to capture full ambient information of sonic events with a 352.8 kHz sampling rate. Even though, higher resolution, higher sampling rates sound better than lower resolution, lower sampling rates, 192 kHz, 96 kHz and 44.1 kHz/24 bit all sound more “digital” than 352.8 kHz sampling because of the pronounced effects of anti-aliasing filters in the lower sampling rate systems.”
These particular marketing messages came from a site that sells digital transfers of old third generation analog tapes so it is perhaps surprising he’s pitching digital standards that far outstrip the fidelity of his analog tape masters.
Does a quad rate DSD recording have “up to 256” times the resolution of a Redbook CD? Of course not! Is it the “highest audio resolution format available”? Also not true. And anyone that would write and post such ridiculous claims makes me want to run for the exit door as quickly as possible.
The stuff about recordings sounding closer to “real analog” and lower sampling rates sounding more “digital” than higher sample rates are also meaningless statements. A duplicated consumer quarter track analog tape or a 2-track stereo “flat” master doesn’t benefit from moving to Ultra HD-Audio sample rates or quad DSD rates. Analog tape isn’t a high-resolution audio fidelity format to start so why transfer tapes to gigantic bit buckets?
Posting these messages are meant to attract uninformed customers who are being regularly pounded with DSD hyped articles and papers. They continue to confuse and confound intelligent audio enthusiasts. DSD is hot because panels of advocates and company representatives have grabbed headlines but they only present half the story.
How about reading an article written by a couple of very well respected designers and makes of equipment…including some of the best DSD converters?
I received an article from an audiophile associate based in Spain. Juan makes the incredible “aria” music server and other wonderful pieces of hardware and software. The article is called “DSD Myth” and was authored by the two principals at Grimm Audio, Eelco Grimm and Peter van Willenswaard (retired).
Here’s what they wrote as an introduction to the paper:
“In recent years the DSD audio format of SACD has seen a big revival in the form of high res downloads. We noticed there’s a lot of confusion about the 1 bit DSD format. Debates about the virtues of ‘128fs’ and ‘256fs’ DSD often lack background knowledge of what is really going on inside the AD and DA converters. Eelco Grimm and Peter van Willenswaard wrote a ‘DSD Myth’ white paper to offer clarity in the discussion.”
Read “DSD Myth“. It cuts through a lot of the nonsense that you find in other white papers, on websites, spouted by panelists at audio shows, and by vested equipment manufacturers. Bravo to these two engineers on writing a thoughtful and accurate paper about the DSD format…and its myths.
Enjoy the sound of DSD if you want…but please stop denying the myths associated with the DSD the format.
17 thoughts on “The Continuing Myth of DSD”
Hi, Mark –
Thanks for sharing the white paper. Unfortunately, it’s a rather unprofessional document, one that’s badly in need of editing–especially the paper’s completely incorrect use of apostrophes.
Also, thank you for a great blog!
Brian…I’m willing to give the authors a break writing in English. They are from the Netherlands and are engineers not writers…the information and message is clear. I shutter to think what would happen if I had to write a paper in another language.
I really am totally over this entire high resolution nonsense altogether now. We are being fed so much BS at all levels and from all and sundry, that I simply want to go back to buying CDs again.
Just about anything from the last decade sounds like crap because of poor recording/engineering/mastering, regardless of the resolution being put forth.
Simply give me a really well recorded/engineered/mastered CD and it is happy days, all the other nonsense is just that, nonsense.
I think also that nine times out of ten when we do get a good sounding high res file it is good for the simple reason that it has not been messed with in terms of dynamics etc, not necessarily because it is high resolution!
I think I’m with you on this. The problem is actually up stream from the delivery platform. Still…I’m stuck on surround and CDs just can’t do that.
Jazz At The Pawn Shop is just about to celebrate it’s 40th birthday and it sounds better than 99% of everything recorded since. It’s more about microphone technique than just about anything else.
Marks recordings sound amazing because he uses good technique combined with SOTA equipment.
I agree on you and with this posting (The Continuing Myth of DSD). This could call “The Continuing Myth of HiRes…”. I’m getting backing to CD’s also. I won’t pay high prices for SACDs or DSDs, when the most important thing of the ring is the recording/mastering of audio.
I sense a lot of stress in your post, don’t let them get to you, your health is more important. As I added to yesterdays blog, its all about who can spent the most $ on PR. 24/192 is all we need, probably more. But it’s old news and the only way to sell new software/hardware is for those in the business to make more $ is to spend a lot convincing consumers that there is newer-better tech that has to be bought into.
Hard to oout market Sony and it’s associates.
Thanks Sal, I’m back to my running…8 miles tomorrow morning…so I’m thinking I’m healthy. The power is in others hands but it still bugs me.
Alright, l’ll Tell What Is Fact. I Own A DA-3000 AND THE ONLY FORMAT THAT ARCHIVES VINYL INDISTINGUISHABLE FROM THE RECORD, IS DSD.
Buy or borrow one, “hear for yourself”
Thanks for the comment. You’re certainly welcome to enjoy your DSD transferred vinyl LPs. And yes, I’ve have heard it and compared it to 96 kHz/24-bit PCM. But because vinyl LPs are standard-resolution media full of distortion and noise, there was no difference. In fact, a CD can adequately capture most vinyl without loss. DSD is a highly flawed technology and unusable for most production purposes. As a professional audio engineer AND audiophile, I wouldn’t record or distribute with DSD. Sony was providing the equipment for FREE back in the early days of SACD — after researching it, I passed.
You missed the point ENTIRELY. Music, noise, distortion, whatever. Recorded in PCM, the source IS DISTINGUISHABLE from the recording. The same is not true for DSD. If you actually did try to RECORD the same source… lets call it the “analog master tape” in PCM, and couldn’t here a difference… then yes, 44.1 will serve you well. No point in “digitally recording” a digital source. If you’re one of those that insists digital (a numerical representation of reality) trumps analog, you’re fighting against the only format that offers any hope for your argument. Quad DSD could be the final word.
You’re certainly entitled to your opinion…personal tastes trump technical realities every time. DSD is a highly flawed method of producing high-end audio recordings. The ultrasonic noise and lack of tools make it unusable in most recording situations. If these formats sounded so different, why did a rigorous study determine that listeners couldn’t distinguish between them. I’ll stick with the best digital encoding scheme yet devised…PCM.
What an obvious paid shill.
CD can capture the resolution of Vinyl my ass.
Actually, CD exceeds the resolution of any analog format including analog tape and vinyl LPs. That fact is undeniable but obviously uncomfortable for those who prefer the “sound” of vinyl. I’d be curious who you think is paying me to state realities. It’s time to study the subject and recognize that dynamic range, frequency response, distortion, speed fluctuations, etc are all better with PCM digital.
Hi Mark, thanks for the original post which still garners hits if you Google “DXD Vs DSD”. The link to Grimm Audio’s DSD Myth is helpful.
It is worth noting that the DSD Myth white paper admits that a properly constructed DSD chain “can sound stunning”; their criticism is that much of what is called DSD has been reconstructed from PCM and worse, many DSD DACs do internal conversion of DSD to PCM, so little DSD playback achieves the claimed quality.
Finally I question your categorisation of Tape and Vinyl as “not High Res”, they are both Analog and thus capable of near infinite resolution – the problem is both formats suffer noise and non-linearity but are capable of sound quality better than 44k CD quality. It is well documented that compromises were made when CDs were introduced.
There are studies which show that most people can’t tell the difference between MP3 at bit rates as low as 128k and CD quality but this is very different from saying Audiophiles with expensive turntables and MC cartridges are getting the same experience as CD quality.
Thanks for the comment Paul. However I take exception to your claim that reel to reel tape and vinyl LPs are ” capable of near infinite resolution.” That’s simply not the case. I have one of the best analog recorders on the planet and it can achieve about 72 dB of signal to noise ratio…far below the capability of standard resolution CDs. The noise, distortion, crosstalk, and other failings of analog limit it to less than CD fidelity. They “may” quality as standard resolution but they will never be hi-res.