DVD-Audio vs. SACD: A 2004 AES Paper Sheds Light
The high-resolution audio world was just getting started when a couple of students at the Tonmeisterinstitut, University of Music Detmold, Germany decided they would do a rigorous study on the sonic differences (if any) between the two physical formats that were available to audiophiles. This was in 2004. The title of their AES paper is “DVD-Audio versus SACD / Perceptual Discrimination of Digital Audio Coding Formats / Listening Comparison Test between DSD and High Resolution PCM (24-bit / 176.4 kHz)” and it was authored by Dominick Blech and Min-Chi Yang. I admit I hadn’t read it previously but I was impressed with the rigor of the study and the scholarship of the resulting study. It brings into question the very essence of the DSD vs. PCM debate. Here’s the abstract:
“To study perceptual discrimination between two digital audio coding formats, ‘Direct Stream Digital’ and high-resolution (24-bit, 176.4 kHz) PCM, subjective listening comparison tests were conducted with specially recorded sound stimuli in stereo and surround.
To guarantee their reliability, validity and objectivity, the double-blind ABX tests followed three main principles: The signal chain should be based on identical audio components as far as possible; these components should be able to convey very high audio frequencies; and the test population should consist of various groups of subjects with different listening expectations and perspectives.
The results showed that hardly any of the subjects could make a reproducible distinction between the two encoding systems. Hence it may be concluded that no significant differences are audible.”
They used ABX testing between a series of new recordings that they produced themselves in both DSD 64 and PCM at 176.4/24-bits. The entire signal chain between the DSD and PCM recordings was very carefully matched with regards to timing, level, processing (none was done to either signal), and reproduction with the exception of the analog to digital and digital to analog conversion. Obviously, in a test of this type the converters must be different. However, in this case they used a dCS 904 converter, which allows conversion in both formats by the same manufacturer. You can read the paper for yourself but they authors took great care to use the very best microphones, cabling, electronics, speakers. They signal path would quality for the JAS definition of high-resolution…which enhances the frequency response to 40 kHz.
The recordings were done with minimal microphone technique. New stereo and surround recordings were made with Schoeps MK 2s, MK4 and MK 41 capsules (with a frequency response to 50 kHz).
The listening/evaluation period took place over 28 days and resulted in 145 tests taken by 110 participants. There were 43 women and 67 men ranging in age from about 20 to over 75. The mean age of 32.9. The pool of listeners included professional recording engineer, music students, professional musicians, a piano tuner, music reviewer, and other non-music professions.
Here’s the summary of the research project:
“These listening tests indicate that as a rule, no significant differences could be heard between DSD and high-resolution PCM (24-bit / 176.4 kHz) even with the best equipment, under optimal listening conditions, and with test subjects who had varied listening experience and various ways of focusing on what they hear. Consequently it could be proposed that neither of these systems has a scientific basis for claiming audible superiority over the other. This reality should put a halt to the disputation being carried on by the various PR departments concerned.”
You can take what you will from this study. It was done over 14 years ago and many things have changed…improved…over that span of years, but the essential premise of this study has remained the same. When audiophiles, equipment manufacturers, reviewers, and publications say that they can tell the difference between DSD 64 and high-resolution PCM recordings, chances are it’s not the format that they’re hearing but the nature of the production process associated with each version.
Perhaps there is no sonic difference between DSD (at any multiple) and high-resolution PCM. If that’s true, then I believe that PCM deserves the spotlight because it doesn’t suffer from the lack of tools, the ultrasonic noise, and file size inflation. It might be time to relegate DSD to “also ran” status and get back to making better sounding records.
“It might be time to relegate DSD to “also ran” status and get back to making better sounding records.”
AMEN Mark, sounds like a very basic and simple truth to me.
BUT the simple truth doesn’t sell new hardware or resell the remasters of classic recordings in the latest/greatest format.
Wish I had an idea to turn things around but I’m just an angry old audiophile who was lied to and cheated by these people for near to 40 years. I’ve watched the industry continue down a road full of liers and snake oil salesmen who’s only interest is the almighty dollar.
Guess the part that makes me the most bitter is that time and money has been wasted on these things instead of making true advances in the art of audio reproduction.
Nope, what the community really needs to get us closer to the SOTA is another $150,000 turntable / phono cartridge combo. BLAH
You’re right! This whole audiophile market is a business and without new and “revolutionary” things to sell, where’s the upside?
Have any similar comparisons been made between cd (16bit/44.1khz) and hd audio?
Within the audio band (20hz-20khz), how much of what we actually hear reaches the extremes?
If most of the sounds we actually hear are contained, as I suspect, well within the audio band then this must leave considerable headroom in the cd format and makes me wonder – do we actually gain an awful lot by going beyond it? Surround sound may be the way forward in terms of realism but do the dynamics have to be so ultra extreme?
I have gained a lot of insight from reading your daily posts and forgive me if I am missing something but surely the real battle is to persuade the record companies, engineers, artists and the listeners to produce and expect a much better quality product with proper dynamics and mastering instead of the blasts of noise we are daily subjected to.
If hdaudio is the by-product then so much the better.
They have. There’s the often quoted Boston Audio Society study, which was completely meaningless since they didn’t audition any real HD Audio. It’s less about format and musch more about making good records.
“Within the audio band (20hz-20khz), how much of what we actually hear reaches the extremes?”
It depends on the type of music. Music with a lot of high percussion, including cymbals, actually has a significant percentage of energy above 20kHz — the question is whether anyone over age 20 can perceive it. Most classical music has almost no energy above 20kHz, less than 0.5%, unless you’re talking about percussion-heavy works, and this 0.5% is largely buried by the much greater energy coming from the 20 – 8,000Hz range.
There is musical information above 20 kHz and I believe (and the recent Robert Stuart AES study and paper) seems to confirm it. However, it is not a huge factor compared to other production choices…but it is so easy to include all frequencies up to 48 kHz that I would suggest why not record and reproduce them? And there are lots of other benefits from working with higher sampling rates.
Regardless of my subjective perceptions, eliminating format wars and/or the massive amount of resulting confusion is good; thank you Mark. perhaps you could explain (although I pretty much know already,) how we ended up with rates that decimate to 48khz instead of 44.1 To me, 176.4 makes a great deal more sense than 192 IMHO.
Craig
Craig, 44.1 kHz came from the earliest days of consumer digital audio (it was originally 44.056 kHz) and video recorders. The professional world of video uses 48 kHz, which is why the difference.
As I thought. Thanks.
Wouldn’t Sony’s “double-blind ABX” comparison have more validity if it were done by an independent tester, University or professional body? Wouldn’t Sony have a stake in an outcome that is favourable to their commercial interests and ends?
Sorry,… but right now, I’m having visions of Dracula overseeing blood testing at a blood bank for type favourable to his type-A,… or a fox supervising hens in a hen house, making sure they’re laying enough eggs to the fox’s taste.
After listen PCM, from redbook to 24/384kHz, upsampled to Double DSD with JRiver Media Player driving my Oppo HA-1 DAC in Native DSD mode, I WILL NEVER LISTEN TO PCM ANYMORE. NEVER, NEVER, NEVER !!!
I don’t need read a lot of maths and technical articles, my ears, listening garbage PCM since 1982 will not allow me make this sacrilege again…
I’m prepared for receive a lot of bad PCM words, but I will not listen neither reply to them, my eyes only see DSD and my ears only listen DSD.
My next upgrade will be a VALVE PURE DSD DAC that doesn’t play PCM, because I stopped listen to bits and my equipment and began listen only to the music!
Bob, no one should ever tell you that you can’t listen to what you perceive as the best sound. If the upsampling, conversions, and hardware chain that you listed above delivers what you want then I say go for it. However, the process chain that you’ve outlined seems a little convoluted given that the fidelity of the final output is still Redbook PCM fidelity limited by the 44.1 kHz/16-bit source specifications. None of the upsampling or conversions contribute to any fidelity improvement…in fact, they degrade the sound from a technical point of view.
I only wish you could experience my system.
The part that maybe is a great true, is that DSD´s lovers are old people and DSD tend to enhace the high frequency (more Airy…), making more pleasant for this age.
I became a fan of DSD , seems responsive, capture details after main sound is gone… As if DSD present the movie, and CD(44KHz/16 bit ) was only a picture …
Maybe hi definition PCM > 192 KHz , 24 bits it´s good enought , but certainly CD 44.1 KHz , 16 bit it´s not !
No comparison…
You may prefer the “airy” sound of DSD but in fact, there is far more noise in the HF range due to the noise being shifted out of the audio band. If you’re a fan a DSD, then continue to enjoy what you hear. The accuracy of PCM – even at CD specs as my research study showed – is virtually all we need to capture the music we enjoy.