Is there research that shows HD recordings sound better than Redbook (16 bit / 44.1 kHz)?
-- Sun, 10/25/2009 - 21:28
I'm from Missouri (the Show Me State), and I'm a trained skeptic. I also love to listen to music on my high-end stereo.
Can anyone point me to good scientific studies which document that HD recordings sound better than Redbook (16 bit / 44.1 kHz) recordings?
When I say good scientific studies, I mean they use good methodology (such as double blind, randomized) and are published in peer-reviewed journals. Thanks for helping me find published articles on this topic.
The best study you can find is to listen for yourself.
If you don't or won't believe your own ears what do "peer group studies" matter?
I've done my own listening and I KNOW that HD audio sounds better.
And how BTW do you train a skeptic? Is it easier than potty training?
I don't need no stinkin' "scientific studies" Been there, done that, and I got the tee shirt. :)
Steven Stone
Contributor to The Absolute Sound, EnjoytheMusic.com, Vintage Guitar Magazine, and other fine publications
I absolutely agree with Steve Stone. If you don't hear the difference, it might as well not exist. In fact, for you it doesn't. Would you spend a bunch of cash on something that sounds better to other people, but not to you? There are plenty of Hybrid SACDs out there - just compare the CD layer to the SACD.
There are however, two BIG 'caveats':
1. Make sure you listen on a good player! My first introduction to SACD was on a $200 Sony universal player. And on that inferior player, indeed SACD didn't sound much better than Redbook CD. That experience kept me out of the SACD market for almost a year. Then I heard an SACD played back on the original Sony SCD-1. WOW!! I'd never heard anything that good on CD...maybe not even on LP! I started shopping for a player IMMEDIATELY!
2. Some SACDs are not remastered to DSD from the analogue source, but from a PCM 'master'. That usually results in a VERY disappointing SACD, which doesn't sound any better then the original CD. The other side is that some direct DSD recordings are so good that, even after being down-sampled to PCM, they are very impressive, and sometimes sound ALMOST as good as the SACD version. So compare SEVERAL Hybrids to be sure.
Taking those 'caveats' into account, I'd be very surprised if you cannot hear the more natural instrumental timbres and more realistic soundstage of the SACD medium.
BTW, here's some objective info: If you understand how bit-rate works in digital media, you know that a higher bit density allows you to more closely recreate the original analog sine wave, just as a higher pixel density allows a digital camera or your TV set to give you a sharper, clearer, picture. PCM at 16/44K gives you about 704K bits/second. DSD gives you over 2 million bits/second. How is that for a 'measurable' difference?
Hey, Hey...Steven, watch it!....there is a proper school for skeptic training and it is harder than potty training. The good part is that its not difficult when constipated and it doesn't cost $35K a year for a degree in political science. I want to know how the hell an undergrad pays off $140,000 of debt for an undergrad degree, aquired at age 22. Have we gone mad as a country?.......so, Id say go for the skeptic training instead.
You might be interested in an editorial I wrote on this subject along the the replies it generated. The editorial references an Audio Engineering Society paper, but one that is flawed, in my view. Here's the link: http://www.avguide.com/forums/blind-listening-tests-are-flawed-editorial
Here's my own personal experience regarding this question. I recently got the Oppo Blu-Ray player (love it) and the second Blu-Ray disc I got through Netflix was a Japanese language movie. When I first started playing it, I was stunned by the sound of the ocean and wind and birds. When the dialogue started, there were no subtitles and I made several attempts to get the subtitles. At one point the three dimensionality and eerie realism of the sound disappeared. I checked the system and found I had gone from hi res sound to regular Dolby surround. The difference in quality wasn't even close. ( I also found out the first dialgue was meant not to be translated because it was through the ears of a child who did not understand what was being said ((Memoirs of a Geisha))). So clearly there is a difference in good hi res versus regular. However, well recorded CD's on a good system sound pretty close to me to run of the mill SACD's, but really good SACD's and DVD-Audios sound great on my system. As an aside, I trust my ears totally. I also over many years learned that there are reviewers who are really good at accurately describing and expanding the language of description of what they hear. Being skeptical is one thing, being so cynical that one cannot see or hear what is there because of what you think is quite another.
I am a director of sales at Wilson Audio and also a recording engineer who specializes in location recording of live classical music events. I have been doing this professionally for over three decades. Frankly I have never been much of a believer in any kind of double blind testing when it comes to sonic evaluation. What I have in fact done to evaluate all aspects of my recording chain is to simply make recordings and listen to them on my reference playback system in my living room. If for example I am evaluating a new mic preamp, I will make a recording with the same mics feeding the mic preamp under evaluation in parallel with my reference and them listen to both separately via separate tracks in my listening room. Clearly there are differences and I will rely on my experience to determine which is the better of the two for my circumstances. This same procedure is used for evaluating cables, microphones, analog to digital converters, etc. But as far as determining if higher resolution formats offer any sonic benefit, the process of making that determination is far simpler. What I do is play one of my 88.2khz masters at 24bit directly to my digital to analog converter and listen. I also feed a parallel AES/EBU signal from the master to a sample rate converter which provides a realtime 44.1khz rendition at 24 bit and then take that signal and use another device to properly dither the signal to 16bits. I then feed the resultant signal to another input of my dac and switch between the two while listening to compare. In the context of my listening room and on my playback system, the differences are very substantial. There are a million adjectives I could use to describe the sonic differences but the most concise way to put it is to say that at 88.2khz/24bit, the sound is far closer to what I consider to be real representation of what the microphones captured than what comes through at 44.1/16bit. That does not mean that the "red book" derivative is bad or unlistenable. It simply means that higher resolution recordings sound better and in my view any effort made by our industry to get this or even higher levers of resolution in the hands of music lovers is very worthwhile.
I do not know of a paper that proves hi rez sounds better. But, here's a link to one that allegedly says that there is no difference.
http://drewdaniels.com/audible.pdf
Also, here is a thread of hot discussion and refutation of the paper:
http://www.sa-cd.net/showthread.php?page=1
In essence, I believe the Meyer & Moran test is flawed in several ways, as the sa-cd.net thread illustrates. The playlist used was mainly analog remasterings into SACD, rather than DSD-recorded material. There are legitimate questions about the statistical methods used in assessing results. I also have a problem with ABX testing in general, because it requires a different listening skill than we normally employ in music listening, i.e., listening for identification, is X = A or B? So, I think ABX testing is biased against finding the kind of small differences that are prevalent in audio today. Yes, I said small differences, in spite of the rhetoric that you see in TAS all the time..
I believe hi rez is better in several ways, but not hugely except in one way, which I will get to. I am referring to the predominant hi rez delivery medium for music, which is SACD. First, some of the small audible differences are mainly due to the increased bit depth over RBCD leading to improvements in noise floor and low level resolution. Also, hi rez audio is relatively free of artifacts from the RBCD "brick wall" filter just above the audio band, because any filtering is shifted way up in the frequency band in hi rez.. So, people are paying big bucks for, say, the new Meridian CD player to get the apodizing filter that counteracts some of these artifacts. But, high rez listeners have been free of these effects from the get go, because of the much higher filter frequencies used in hi rez. Some people think the much higher upper frequency response of hi rez is a plus, but I do not think that is really audible or provable. Net result of these diiferences is a small improvement in sound of hi rez over RBCD.
So, what's the big difference? It's multichannel, which in the consumer market is only available in hi rez. The difference in faithfulness of a hi rez Mch recording vs. stereo CD or vinyl to the sound of a live concert in real acoustic space is huge. It's such a huge difference vs. stereo that even ABX testing would reveal it. But, it is so obvious a difference that no one would be stupid enough to really conduct such a test. Testing aside, if you love the way live music sounds, hi rez Mch is the closest approach by far to reproducing it.
So, if that's true, why dosen't TAS say so or review much Mch or even hi rez gear? Why don't dealers demo it with music? What don't more hi end manufacturers make Mch gear? Why, indeed!! It boils down to not biting the hand that feeds you. But, that's a long story for another day.
It is a basic, non-disputable fact backed up by a cascade of evidence that human senses are easy to mislead. Anyone doing blind wine tasting will tell you how much the label on the bottle will mislead you. Likewise, paying more for an audio product will color your perception in a positive direction.
Of course, this does not imply that you cannot draw your own conclusions. But you have to listen with an open mind, and blind testing helps you do just that. However, a useful test, be it blinded or not, has to be timed according to the differences of the tested items. When I decided between two cartridges for my new turntable, I had to spend several hours to make a choice between the two contestants. OTOH, some differences in bass reflex alignment in my DIY speakers are sorted out in seconds.
Post new comment