Another Article on DSD vs. PCM…Really?
Whenever a comment is made that references this site (in this case a link to the John Siau interview…which has now received over 5500 views!), I get a notification. Today, someone referenced the interview with John in a series of comments on the Computer Audiophile site. Chris Connaker posted a link to an article written by Charles Hansen of Ayre Acoustics, makers of very high-end equipment. The title of the piece is “World’s First Valid Comparison of PCM versus DSD?”
It begins with:
“Recently the introduction of computers into home audio playback systems has made possible an unforeseen occurrence — the reintroduction of DSD, the modulation scheme used in Sony’s failed format of SACD from the turn of the millennium.
At the end of the 1990s as the CD patents were expiring, so was a huge revenue stream for Sony and Philips, developers of the Compact Disc format. Anxious to replace the CD with another exclusive format that would also generate licensing income, Sony and Philips tried again with the Super Audio Compact Disc or SACD. In the meantime, none of the other hardware manufacturers were having any of it. They all saw the explosive growth of DVD as the wave of the future and wanted to base any new format on DVD. Thus began one of the most bizarre chapters in the history of audio formats.”
Here’s a link to the entire article Read the article
As you would expect, the CA post immediately stirred the pot about this touchy subject among advocates on both sides of this latest “format war”. There are a couple of very insistent individuals that post with way too much regularity on that site about that topic (DSD proponents). If you have too much time on your hands, feel free to sample some of the usual musing on the topic. I’ve pretty much given up trying to make sense of these people.
What I did find amusing was the title of the piece. It’s a showstopper…you just have to read an article called, “World’s First Valid Comparison of PCM versus DSD”. It might make you think that Charles and the folks at Ayre have actually done something new and innovative with regards to DSD vs. PCM. I’ll skip right to the punch line and save you the trouble of reading the article. They didn’t do the first nor did they do a valid comparison of these two formats.
I do applaud Charles for saying much the same thing that I’ve been saying here for these past 8 months. It takes a brave man to publicly take on the Sony juggernaut and the proponents of DSD. The article is accurate with regards to the facts behind the format and the misinformation behind the SACD format and DSD.
But Charles and his team offer up a series of recordings made in DSD and PCM for comparison. You can download the files and listen for yourself from their site. But just like the flaws in the Boston Audio Society study that pronounced CD and HD media as identical, the team at Ayre transferred older vinyl LP through their A to D converters making the test pretty useless, in my opinion. These are the files that you can compare. I’ll download them tomorrow and get back to with my assessment, but I’m already disappointed that the sources are vinyl LPs. Why choose an inferior audio recording format to establish the differences between formats that are superior?
Doesn’t it make more sense to start with a new recording recorded through the same signal path to both DSD and HD-PCM? That’s what a group of us did back in March. I’d say that was a whole lot more valid and it happened nine months ago.
To be continued.
Mark,
I tracked down your interview with John Siau and was appalled at the dogmatism of the DSD proponents in the comment thread. Reflecting on that and the “test” you describe here, which basically asked “which system sounds more like vinyl?” and it seems to me that it’s a bit like baby ducks imprinting on the first creature they see.
They are a holdout generation who imprinted on the sound of vinyl and refuse to consider any other sound to be valid. But the good news is that they’re pricing themselves out of the market with their vinyl demagnetizers and other hocus pocus, and they are simply “aging out.” Ultimately, they’ll be out of the debate.
The current generation has grown up on the 44/16 PCM brought to them on CDs, and the spread of Blu-ray will acclimate folks to 48/24, which is the go-to format for major films and concert Blu-rays (Eric Clapton’s Crossroads series being the rare example of a major label 96/24 Blu-ray disk).
While HDTracks and others may dilute the term HD with 96/24 reissues of analog masters, they’re at least using PCM, and – if done with greater care than prior releases – such releases can sound better than the prior releases. That can help PCM win the war against DSD, which has nothing going for it other than being copy-protected. Think of the reissue sites as showing that since 24 bit PCM can modernize the sound of our musical heritage, imagine what it can do when new recordings are made using that kind of technology.
The longer-range real battle is against lossy compression, since the urge to stream everything will push in that direction. See the article in Wired a while back about “The Good Enuf Generation,” where they played an mp3 and the CD it was ripped from for college students – who preferred the sound of the mp3!
The secret weapon, ultimately, will have to be getting folks to make music themselves – or at least hear it live in intimate performance venues without excessive amplification, so they have a real world standard of comparison.
Phil
Hi Mark
I read the article on CA as well. I started smiling when reading the piece as i was anticipating your reaction. Luckily you read it as well and I am still smiling after your reaction. Fully agree. I most smiled when i read they used a vinyl recording as a master for this comparison.
Its is flabbergasting that (almost) no new recording are brought out in DSD and PCM-HD. And i mean than the more popular music.
I must admit that I always was more leaning towards DSD ( being Dutch, favoring Philips) but now, after reading your pieces (with good scientific evidence) I lean more towards PCM-HD.
Music must touch the heart and can do that it bad and good recordings.
I’m the one who posted the link on your discussion with John Siau. I applaud your efforts to promote high quality recordings. But I’m disappointed in the way you go about dealing with the Meyer & Moran study. You seem fully competent to know the best way to refute a study such as M&M is to perform a double blind test yourself with source material that meets your preferences. I think the world would benefit from such a study. I have no idea what the outcome would be, and nor does anyone else – which is why everyone would benefit from it.
Why not go back to the BAS, use the same ABX box and converter, but with material you prefer? The more convinced you are that the result would be different, the better it would be to reuse as many of the methods and analyses as possible to eliminate new variable influences.
The reasoned, impartial and detailed way you go about in addressing the tradeoffs of DSD and PCM is impressive, and you’ve contributed significantly in making a credible case that high resolution methods in recording and producing are superior to 16 bit/44.1k recording and producing. I can only encourage you to emulate that in investigating the true audibility of 16 bit/44.1k versus higher resolution formats at the point of customer playback with double blind testing.
There were lots of problems with the M&M study. There would be no benefit to using their flawed methodology or hardware. But it’s funny you bring a new research project today, because there is a phone call tomorrow with a prominent organization that will be discussing my proposal to establish once and for all the issue of high-resolution audio and its perceptibility. I have laid out a plan that will do a double blind A | B | X test of real high-resolution materials amongst other things in question.
It’s going to take some time and it’s going to take some funding…but the first steps are in motion.
Great news! A robust double blind study is indeed a great deal of work, so I hope you proceed. In reading your critiques of the M&M work, I’m sure you appreciate the irony of slamming the source quality of the SACD/DVD-A discs: consumers have the same dilemma, and if an average consumer can’t tell which songs have appropriately high quality sources to take advantage of any real or imagined benefits of high resolution formats, then the purchase decision and customer satisfaction of high resolution vs 16 bit/44.1k formats becomes problematic.
No doubt you’re aware of this issue and have better thoughts, but it seems some kind of simple consumer labelling is needed to discern high resolution songs using poorly recorded, mixed or mastered sources from those that follow best practices. A crude example is the SPARS code used on CDs – this was something I could even teach my Mom to look for (and did). Use Vorbis tag field(s) on downloaded files to reflect whatever code is used. Put another way, if practices like those used in the Loudness Wars aren’t also dealt with, many consumers won’t see the point (or more specifically hear the point) of high resolution formats.
Good luck!
Hi Mark
What was the outcome of your test back in March?
Tim
I posted the results in an earlier post> You can read it at http://www.realhd-audio.com/?p=629
How is vinyl an inferior audio recording format? First of all, it’s not a recording format. Secondly and more importantly, it is not an inferior medium. Yes, I agree that Ayre should have used a new analog recording using professional tape recorder. However, a vinyl record played on a quality deck & stylus combined with a good phonostage comes bloddy close to the sound of original recording.
Vinyl LPs are a wonderful format (and yes it actually is a format…there have been hundreds of recordings made using a direct to disc approach) and there are many…likely including you…that feel they represent the sound of the original recording better than any other format. As euphonious as they may be, they are usually created from analog tape masters. The dynamic range and frequency response of an analog tape machine specs out to be less than a well made compact disc…especially with regards to dynamic range. When vinyl LP is mastered (spent some time as a disc cutter), the low frequencies are summed to mono to ensure the groove will be symmetrical and the high frequencies are juiced so that they will come back through the RIAA curve close to right.
However, a high-resolution PCM 96 kHz/24-bit digital recording delivers a more accurate recording…if the goal is to come close to the sound the actual musical performance. I hope you’ve had a chance to download the HD-Audio tracks from the FTP site. I’m sure you’ll find them interesting.
“a vinyl record played on a quality deck & stylus combined with a good phonostage comes bloddy close to the sound of original recording”.
No. It doesn’t. Use difference testing to compare a flat master tape or file with what comes out of a phono stage after vinyl mastering, vinyl production, and vinyl playback. The difference test should yield silence. Prepare for a shock.
Now do the same with an ADD or DDD process.
Can’t do this because you don’t have access to the flat master, or equipment to process a difference test? Then how can you make the claim you did?
The fact is, the vinyl process does a huge number of things wrong, from high distortion, to poor frequency response, poor signal dynamics, high noise, ease of damage (and high audibility of damage), terrible production variability…. it is all completely unacceptable to people interested in accuracy. The other fact is, some people like it like that. It is just a bit annoying when those people insist that black is white and worse is better.
I recently acquired an Oppo BDP-105D and downloaded David Elias’ DSD recordings from Oppo’s website. The one track that is also done in PCM for comparison was at least 5 dB louder than its PCM counterpart according to my Radio Shack SPL meter. I had to increase the volume on the 105 by 8 “clicks” just to get the PCM version to equal the DSD file in volume. As louder tends to sound better, I can only hope that the difference between these files was not intentional. I understand Elias is a strong proponent of DSD recording and playback. After equalizing the volume, both files sounded so close that I can’t pick a winner yet. But I’ll keep listening.
Don’t you know louder is always “better”. I know David and he’s sent his tracks in both PCM and DSD to sell on iTrax. I haven’t gotten around to the DSD files but his productions have been available as PCM files for a long time.