AVS Forum banner

1 - 20 of 204 Posts

·
Registered
Joined
·
76 Posts
Discussion Starter #1
I just read an article that I believe others reading this forum might enjoy. Apparently there is a new study out that compares CD with DVD-A and SACD audio quality.


I haven't read the study, just an article discussing it (URL below), but apparently 2 researchers have shown using extensive double-blind testing that there's no audible difference between 44.1 Hz / 16-bit audio and the more advanced audio available from SACD and DVD-A:

http://theaudiocritic.com/blog/index...Id=41&blogId=1


Apparently the study only looked at 2-channel stereo sources. Here's the first paragraph from the web page I referenced above:


"In the September 2007 issue of the Journal of the Audio Engineering Society (Volume 55, Number 9), two veteran audio journalists who aren't professional engineers, E. Brad Meyer and David R. Moran, present a breakthrough paper that contradicts all previous inputs by the engineering community. They prove beyond a shadow of a doubt, with literally hundreds of double-blind listening tests at matched levels, conducted over a period of more than a year, that the two-channel analog output of a high-end SACD/DVD-A player undergoes no audible change when passed through a 16-bit/44.1-kHz A/D/A processor. That means there's no audible difference between the original CD standard (Red Book) and 24-bit/192-kHz PCM or 1-bit/2.8442-MHz DSD."


Thoughts, anyone?


-Polar Weasel
 

·
Registered
Joined
·
54 Posts
I'm not surprised by that, as 16/44.1 done properly exceeds the resolution of human hearing and is, for all practical purposes, transparent (assuming good low-level linearity in the A-D and D-A conversion stages.) However, it doesn't address the other advantages of the high-resolution formats, namely the multichannel functionality (CD is limited to stereo) nor the extra care often put into hi-res mastering (versus the deplorable lows to which CD mastering has sunk thanks to the labels and their accursed loudness war.)


So yes, 16-bit/44.1KHz can sound absolutely superb when done properly. The problem is that it seldom is these days, at least from the standpoint of major-label releases. Thus we have the ridiculous circumstance that even a vinyl copy of a new release is often much more listenable and dynamic than its CD counterpart.


I've often wondered how hi-res would fare in a double-blind trial with real-world program material...this seems to go a good way toward answering that.


Aczel's observation here is interesting (and it highlights why I record at 24-bit in ProTools) :

Quote:
It should also be pointed out that more bits and a higher sampling rate in recording are still a good thing because they permit a little bit of unavoidable sloppiness, so that you can still comfortably end up with 16-bit dynamics and 20 kHz bandwidth. Meyer and Moran do not say that 14 or 15 bits in a truncated CD are just as good as 20. What they say is that spot-on 16-bit/44.1-kHz processing is as good as it gets, audibly.

TP
 

·
Registered
Joined
·
3,656 Posts
Where's the "popcorn smiley" when you need it?



There are few things that come to mind for me (and I neither have "golden ears" nor I am a general believer in the more esoteric "tweaks" out there--I don't use exotic cables or power cords, for example).


One--I'd like to know what discs were used. The biggest theoretical advantage of hi-res for 2 channel, in my opinion, is the greater available dynamic range (though the audibility of that, all other things being equal, is obviously a point of contention).


Two--I suspect that much of the better quality audio in hi-res owes something to the fact that, generally, those that go to the trouble of releasing in hi-res take extra care to do it right. Far fewer instances of overwrought dynamic compression (a HUGE culprit in the last decade of redbook releases overall), fewer egregiously gross applications of unnecessary EQ, and so on. So while redbook CD MAY (I'm not concluding anything, just stipulating for discussion) be "as good", the people issuing redbook are NOT making the necessary effort for the practical result to be "as good". As such, I will continue to buy hi-res when it is an available option.


Three--and this is purely a personal preference, I buy hi-res PRIMARILY for MCH presentations. Redbook CD is not capable of providing me with that in anything other than a lossy DTS-like format and I can most certainly tell the difference between a lossy DTS-CD and a hi-res lossless presentation of the same mix (I've done a level-matched comparison). As I am a big fan of MCH (discrete) mixes, I will happily buy a lossy DTS-CD version if there is no lossless version available, but I will always privilege the lossless option.


So, this is interesting (and, perhaps, not all that surprising) but as MCH is my preference, this study's conclusions will have no bearing on my hi-res purchases.
 

·
Registered
Joined
·
10 Posts
Im not surprised to see CD audio can sound identical and "as good" as SACD and DVD-A, but as others have pointed out, this is almost never the case in practice. Standard redbook CDs put out by big labels these days are just about always a disappointment to my ears, simply due to the "louder is better" nonsense in the industry. Those manufacturers simply dont care how their discs sound on a hi-res system. So I avoid them like the plague, my limited listening time is worth more to me than that.


Just about the only redbook CDs I'll buy anymore are those put out by MFSL, since I know I can expect the best possible recording on a CD from them.


If there was some kind of online archive detailing the sound dynamics of every CD release, then I might be inclined to buy more. There are some artists out there that will take the care and extra effort to release properly mastered CD material. Bob Dylan and Norah Jones come to mind, their cds mostly sound excellent. But for the vast majority of CD releases, I just wont throw $10-15 away on an overprocessed piece of garbage that my ears will reject. Heard the new Springsteen CD on a hi-res system? Boy I regret that one. Too bad cause the songs are quite good.
 

·
Registered
Joined
·
218 Posts
Bob Dylan has said how poor his own CD's sound verses the studio masters.


Better production, mastering, theoretical/actual 24-bit resolution, 2.0 mix and 5.1 channel mix are all reasons to like SACD and DVD-A.


They are great alternative to the poorly done CD's and MP3 files. Too bad the major record companies could give a sh!t about quality.
 

·
Registered
Joined
·
291 Posts
I could hear a difference between CD, DTS and DVD-A for Porcupine Tree's Fear of a Blank Planet. Assuming that Steve Wilson carefully mastered all three to the best fidelity he was capable of, it doesn't explain the audible difference that were apparent even to these aged ears. And what about hybrid SACDs like Peter Gabriel's UP? The SACD layer, again, sounds superior to the redbook CD layer in my opinion.
 

·
Registered
Joined
·
3,656 Posts
Level matching is the key. Without that, no comparison is really valid. Even a crude level matching with an SPL is better than nothing, though it is still not sufficient for a scientific study.
 

·
Registered
Joined
·
1,771 Posts

Quote:
Originally Posted by Ovation /forum/post/12886783


Where's the "popcorn smiley" when you need it?



There are few things that come to mind for me (and I neither have "golden ears" nor I am a general believer in the more esoteric "tweaks" out there--I don't use exotic cables or power cords, for example).


One--I'd like to know what discs were used. The biggest theoretical advantage of hi-res for 2 channel, in my opinion, is the greater available dynamic range (though the audibility of that, all other things being equal, is obviously a point of contention).


Two--I suspect that much of the better quality audio in hi-res owes something to the fact that, generally, those that go to the trouble of releasing in hi-res take extra care to do it right. Far fewer instances of overwrought dynamic compression (a HUGE culprit in the last decade of redbook releases overall), fewer egregiously gross applications of unnecessary EQ, and so on. So while redbook CD MAY (I'm not concluding anything, just stipulating for discussion) be "as good", the people issuing redbook are NOT making the necessary effort for the practical result to be "as good". As such, I will continue to buy hi-res when it is an available option.


Three--and this is purely a personal preference, I buy hi-res PRIMARILY for MCH presentations. Redbook CD is not capable of providing me with that in anything other than a lossy DTS-like format and I can most certainly tell the difference between a lossy DTS-CD and a hi-res lossless presentation of the same mix (I've done a level-matched comparison). As I am a big fan of MCH (discrete) mixes, I will happily buy a lossy DTS-CD version if there is no lossless version available, but I will always privilege the lossless option.


So, this is interesting (and, perhaps, not all that surprising) but as MCH is my preference, this study's conclusions will have no bearing on my hi-res purchases.

Nicely put! I agree on all points.
 

·
Registered
Joined
·
1,180 Posts
I am not surprised by this but as some have stated 2ch SACDs are bought by most in hopes of obtaining a better master and since in some cases this has turned out to be true I generally prefer to purchase a SACD version if available - just like a Mobile Fidelity release.


No one can dispute the audible differences in multi channel however.
 

·
Registered
Joined
·
1,025 Posts
Another confounding factor is that the actual listening playback was conducted differently than a "real-world" listening session. By interposing another component in the chain (the a/d/a converter), it's quite possible that the addition to the signal path caused some limitations in the sound quality. A better method would have been to have an SACD and a redbook CD of the same recordings to compare, without another "atypical" component in the electronic chain.


I also believe that you must have a system that can perform at a certain level of resolution, etc. to hear much difference.


Equipment:


B&W 801D (L,R)

B&W HTM-1D (C)

B&W CWM8180 (3, FOR 6.1 SURROUND)

Krell Evolution 403 amplifier (L,C,R)

Denon AVR4806 (preamp, processor, surround amplifiers)

Denon 3930CI

Samsung BD-P1000


I can usually identify the SACD vs. the CD on well-recorded material. Garbage recordings are much less reliable.



My two Cents' Worth,



Lee
 

·
Registered
Joined
·
54 Posts
From the article at BAS:

Quote:
When the subject was listening to the high-bit audio alone, whether sighted (with the display showing A) or blind (with the display showing X), our test system added the following components in series with the high-bit player’s output: (1) an ordinary RCA connecting cable 18 inches long, and (2) a switchbox comprising two RCA connectors, a total of 4 inches of hookup wire, and a reed relay with 0.2-ohm dc resistance.

There is absolutely no chance of even slight let alone significant audible degradation with this setup. Any believed degradation would only appear in a sighted test, and would be a manifestation of the placebo effect.


Moreover, comparing the SACD layer to the CD layer can often be apples/oranges, if the label used the standard redbook master for the CD layer rather than doing a 2-ch mix from the DSD master. This particular trial was more well-suited for determining whether or not the 16-bit/44.1KHz system was inferior to higher-bitrate/bit-depth audio.


An even better test would be a live music source recorded on synced DSD and standard recorders, then the comparisons made with the synchronized playbacks. Perhaps a chamber orchestra, and a solo piano with violin would be good source material.
 

·
Registered
Joined
·
391 Posts
I'd like to hear more about the recordings they used. Were they pop recordings? Classical? Jazz? Were they from DSD masters or from PCM masters that were converted to DSD?


I certainly hear a difference in my 2.1 channel system between the SACD layer and the CD layer--and I have done demonstrations for my music-loving (but not necessarily audiophile) friends who agreed. Could it be due to different masters used for the SACD and CD layer? I'm willing to concede that. Still, the best SACDs I have still unquestionably sound better than my best CDs.
 

·
Registered
Joined
·
427 Posts
As the producer of recordings that are actually captured using PCM at 96 kHz/24-bits, I have a difficult time with the work done by the authors of the AES paper. It was interesting to finally have access to a partial list of source materials used in the evaluation...the original paper published in the journal didn't mention any of the recording that were used. I contacted David Moran and asked for a list of the recordings (I never received the list) back in the fall, because it is central to the issue that is raised about whether one can tell the difference between a CD and an HD Audio recording.


If the source recordings were done in the era of analog tape or SD digital (which many if not most of the items on the list were), then it is not possible for them to contain any sonic information beyond the technical specs of the equipment used during the original tracking or overdubbing sessions. As the prior comment points out...virtually all commercial CD releases AND the majority of SACD/DVD-Audio releases are based on source recordings that do not contain frequencies or dynamic ranges that are the benchmarks of the new HD standards (his mention of 60-70 dB is accurate and manageable in 10-12 bits). No wonder the authors of the "study" arrived at their conclusions (I'm not going to get into the limitations of the components of their "state-of-the-art" evaluation system).


To do this sort of evaluation correctly and with some real rigor, several critical things have to happen that didn't happen in the published paper.


First, the source recordings have to be REAL HD audio recordings...meaning they must have been made using a recording chain capable of capturing digitally at 96 kHz/24-bits or better (DSD is probably OK but subject to rather artifacts as a result of the required noise shaping). No older analog tracks or SD digital tracks subjected to upsampling or HD transfer can be used. It is critically important that the source material be tested and be shown to have frequencies beyond those present on a CD (Keith Howard has evaluated many recordings that have plenty of real sound energy above 22 kHz). And the dynamic range needs to be beyond the 60-70 dB range of most recordings.


Second, the playback chain for the A-B testing MUST be completely transparent using only the finest converters (critical jitter specs), cables and amplifiers. The gear has to be able to reproduce the HD Audio. Of particular importance is the speakers, they have to be able to reproduce frequencies and dyanmics that are present in REAL HD Audio recordings. That means dynamics up to 124 dB and frequencies well beyond 22 kHz...I suggest up to 40 kHz.


I should point out that I have a stake in this argument. My little audiophile label (AIX Records) has recorded and released one of the largest collections of HD Audio recordings...and we've won numerous fans and awards for high-resolution audio. I found it interesting that David Moran hadn't used any of my recordings in his test. He told me that people brought in their favorite "high resolution" recordings and they did the test using those. In fact, he told me that he didn't compile a list during the testing period. A short-coming of their "study" that doomed it to failure.


BTW I do not master any of my tracks...they are recorded at 96 kHz/24-bits, mixed digitally without compression, EQ or artificial reverberation and then sequenced before being put on DVD-Audio discs or placed online for download. I have distributed them as DVD-Audio tracks and now make them available as HD downloads.


Our new web site, itrax.com has the same source recordings available as MP3 files, Dolby, DTS, WMA Pro, WMA LL and 96 kHz/24-bits (and in three different mixes). There is a set of free files that can be downloaded and compared for those that are interested. Listen to REAL HD Audio track as compared to a CD version and I continue to maintain that music listeners can tell the difference. The high resolution formats may have failed but it wasn't because the formats were incapable of improving the fidelity of the music.


I haven't even started talking about surround. I don't get to this forum that often...if you want to follow up please go to the community on the itrax site.
 

·
Registered
Joined
·
326 Posts

Quote:
Originally Posted by zoney99 /forum/post/12918012

You should hear what the "Saturday NIght Fever" soundtrack sounds like



..no, you really shouldn't!!!!!!!!!!!

Actually, yeah you should. It was recorded at Criteria, Jeep Harned's own technical test bed. Amazing stuff under all those Bee Gees vocals. Unfortunately, the 2-track master was only done at 15ips (I've held it in my hands.) so the transient response wasn't all that spectacular, but in other respects, it was technically exceptional.
 

·
Registered
Joined
·
16,876 Posts
"Hirez" is better cuz the name sounds cool!
 

·
Registered
Joined
·
4,653 Posts
I've always felt the problems with CDs isn't the frequency response limitation, but the sample resolution. Had we had 24bit/44.1khz CDs from the start, I believe there would be no need for hi-rez audio except for multichannel recordings. I've found that 16bit recordings have a slight harshness or slight un-organic edginess to them. 24bit and DSD recordings are slightly smoother, more natural, and a little richer...granted the difference is subtle.
 

·
Registered
Joined
·
16,876 Posts

Quote:
Originally Posted by Kal Rubinson /forum/post/12921449


I thought Hirez was a root beer.

Mmmm. If only I had some root beer AND some 'nilla I scream. Sadly, I have neither in the house at the time.



Sadist!
 
1 - 20 of 204 Posts
Top