|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.76.104
In Reply to: RE: Agreed posted by E-Stat on June 19, 2010 at 12:20:33
"When every recording engineer has compared the direct feed of a recording through multiple resolutions and found Redbook lacking, why would E Brad ever consider that his test is fatally flawed?"
Most, perhaps. But not all. E Brad is a recording engineer, but I haven't heard any examples of his work.
Sorry to be picky.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
E-stat
"When every recording engineer has compared the direct feed of a recording through multiple resolutions and found Redbook lacking, why would E Brad ever consider that his test is fatally flawed?"
I am not quite sure just what E-stat is saying there, but he seems to refer to all the editing ("multiple resolutions," whatever he means by that)one might do prior to putting a recording on to a CD. I don't think anyone maintains that hi-rez editing is not better than 44.1 kHz digital. The Meyer-Moran paper doesn't address that, just whether the resulting stereo recording can be recorded onto Redbook digital without audible degradation--save for some additional noise audible at very high levels. That's all that was tested, as the very title of the article suggests:
E. Brad Meyer, "Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback," JAES, Vol. 55, No. 9, 2007 September.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Actually, even if you believe their tests were sensitive enough and the listeners well trained (I don't) they did not show that you could record at 44/16 and not lose anything. The reason: they did not actually record anything at 44/16, just converted data in that format on the fly .
Indeed, there could be degradations in the 44/16 equipment that they used that could have gone unnoticed in 44/16 loop back that might have been grossly audible were the device actually used to record and then play back later. For example, if the clock were very unstable it might impart a tremendous amount of jitter on record that would be almost exactly undone when playback took place with the same clock. But if the data had been stored between record and playback any clock errors would not have canceled so neatly.
IMO the entire concept that two sounds that differ by less than the threshold of hearing are somehow equivalent does not generalize in the context of a complete record-playback system. Substituting one "equivalent" component for another may result in two sounds that are equivalent, but if a bunch of equivalent components are all substituted the cumulative result may be well over the threshold. Hence, tests like those that Meyer and Moran ran are relevant, at best, to those who are willing to settle for mediocrity.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Well, the tests dealt with what is possible with the consumer product. If the SACD or DVD sounds different from the CD (and I know there are dual layer SACD/CDs), one has the right to ask why, since at the levels used, no one in the Meyer-Moran tests showed they could, and that includes some "audio professionals."
The Meyer and Moran tests took the analog output of the SACD or DVD players and digitized it to a 44.1 kHz sampling rate and then converted it to analog. I personally wouldn't call that data conversion, as from what I understand that is a purely digital process.
Since you are too cheap to get the journal article, I will provide you with a couple of URLs for the paper and a further explanation:
http://drewdaniels.com/audible.pdf
http://www.bostonaudiosociety.org/explanation.htm
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Thanks for emailing the link.
I am not too cheap. I don't pay for journal articles on principle . At one time journals actually did something: they did editing, typesetting and production. Now the editing and type setting is done by the author and the production is essentially free. As before, the editorial function is done by free volunteers. (I just refereed a paper yesterday.) In other words, journals are a rip-off and I won't support them unless absolutely necessary. (In this case, I already read the article because Brad mailed it to me when it originally came out. I just couldn't find it.)
I don't mind paying money to authors (or in the case of music, artists). It's just being ripped off by middlemen that bugs me. Technical standards are the worst, because all of the people that do all the work to write the standards don't get any money for their efforts. One of the reasons why the TCP/IP based Internet technology won in the marketplace was because all of the standards for the Internet are available for free downloading, encouraging additional contributions from impecunious students, etc...
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Comparing same content of live mic feed vs. recordings at various resolutions, i.e. DXD, 24/192, DSD, 24/88, Redbook, etc.
Example
Another
One from Tony
rw
That is a different situation and not what Meyer and Moran tested. And for this new situation, all you offer is anecdotal evidence. That would be a good reason to do a controlled DBT to see if it's actually so, but you have such a low standard for what you accept as evidence.
I have Telarc Digital Stereo 5039, the famous LP recording of Band Suites by Holst, and music by Handel and Bach. In the liner notes, Stan Ricker said: "The signal from the digital sounds exactly the same as what we heard coming from those transformerless Schoeps microphones." Of course, the Soundstream sampled at 48 kHz, not 44.1 kHz.
So, I can find anecdotal evidence which indicates hi-rez is not necessary to produce recordings indistinguishable from the mic feed. Now, if you could only come up with some controlled DBTs that establish your beliefs, that would be great.
The fact is that the SACD and DVD signals were passed through a 44.1 kHz DAC and were not distinguished by the audiophiles taking the tests run by Meyer and Moran. Coming up with some other situation which you think should be tested doesn't affect the validity of their results.
According to E. Brad Meyer, "no one can tell when an ABX box is in the circuit." To make your objection at all significant, you would have to show that someone, at least, can hear it in the circuit. I've been waiting for years.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
That is a different situation and not what Meyer and Moran tested.
Comparing the original feed is the real world, not some contrived attempt at duplicating the real world.
Stan Ricker said: "The signal from the digital sounds exactly the same as what we heard coming from those transformerless Schoeps microphones." Of course, the Soundstream sampled at 48 kHz, not 44.1 kHz.
Obviously, his opinion changed when he had access to higher resolution:
Here with Meitner
"I have to admit that...
"...the tape which was, I believe, 96 K, 20-bit from a Nagra certainly sounds better than the CD that came from that recording."
I have also heard a direct feed from the original Soundstream recorder when I participated in the ASO recording of the Firebird.
The fact is that the SACD and DVD signals were passed through a 44.1 kHz DAC and were not distinguished by the audiophiles taking the tests run by Meyer and Moran
Yep, that is what happens when you assume a $250 player is audibly perfect.
According to E. Brad Meyer, "no one can tell when an ABX box is in the circuit."
Fine. Provide the details of the control tests he ran in order to verify that assumption. I've never seen that done before.
rw
If we're comparing Stan Ricker anecdotes, which could hear better: the Stan Ricker in 1978 or the Stan Ricker over 20 years later? In any case, what is unreal about comparing the SACD or DVD with a 16/44.2 kHz recording of it? After all, there are dual layer SACD/CDs.
E-stat
"I have also heard a direct feed from the original Soundstream recorder when I participated in the ASO recording of the Firebird."
Nice recording. I have it both on LP and CD. But we discussed this before: as I recall, you didn't compare the direct feed from the mic with the feed from the Soundstream recorder.
E-stat
"Yep, that is what happens when you assume a $250 player is audibly perfect."
This is a fantasy in more ways than one. First, a number of other players were also used, including a Sony, a Yamaha, and a Denon. Second, I should point out that at very high gain levels, the testers could hear some low level nonlinearities in the left channel of the Pioneer player. Hmmm . . . the ABX switcher didn't cover that difference up, now, did it? That's not an assumption. I am also shocked that you characterize the players by their price!
Now, as well, at high gain levels, the noise levels of the CD loop in the test was audible--but the level was uncomfortably high for listening to music. My, my! The ABX switcher didn't cover that up, either. That's not an assumption, either.
In fact, with DBTs employing an ABX switcher, a number of positive results have been obtained, and you haven't come up with anything that the ABX switcher covers up.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
If we're comparing Stan Ricker anecdotes, which could hear better: the Stan Ricker in 1978 or the Stan Ricker over 20 years later?
Apparently not. He likely has a greater depth of experience over those years. I know that my exposure to more live music and higher resolution equipment has changed my perspective. I guess that is a foreign concept for one who stays on the bench.
Hmmm . . . the ABX switcher didn't cover that difference up, now, did it?
That the box reveals gross differences is not under debate.
...you haven't come up with anything that the ABX switcher covers up.
That would be virtually every recording engineer with first hand experience. Speculate on!
rw
E-stat
"That the box reveals gross differences is not under debate."
You have a peculiar idea of what gross effects might be!
Meyer and Moran looked at some of the things asserted by the subjective audio press, audio professionals, and audiophiles, and tested them. You don't seem to include the experience of researchers who do audio DBTs in your concept of experience.
Quote from me:
"...you haven't come up with anything that the ABX switcher covers up."
E-stat comments
"That would be virtually every recording engineer with first hand experience. Speculate on!"
Dream on! Some of the Meyer-Moran tests were conducted at "a CD/DVD mastering facility." "Some of the source material for these trials was a classical music production which was then in process at this establishment."
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Unlike you, many of us have experienced different systems whose performance is beyond the insipid. You're among the proud few who wave the banner of mediocrity high in the name of audio.
Dream on! Some of the Meyer-Moran tests were conducted at "a CD/DVD mastering facility.
Which further illustrates your inability to understand the significance of comparing the live feed and the result. Hint: recordings are not performed in a mastering facility. Perhaps it's time for you to look at the picture and read the text in your link:
The "bulk of the trials" were using an "audiphile grade" system (pictured) using a Pioneer DV-563 player, Adcom preamp, Carver amp and zip cord? Audiophile grade? Is he a moron or just thinks that everyone else is? That's just too funny. I am astounded at the profound ignorance demonstrated by his assumptions - both with the validity of his contrived "this-is-never-how-systems-are-used" *test* and his choice of test systems. You can prove whatever you want with such dumbed down platforms. BTW, you can pick up one of those truly revealing "state-of-the-art" players on Amazon for about $75. Go for it dude. You'll be in heaven. LOL!
rw
One wonders what you are talking about. You have just said that the mic feed and the recording sound just the same based on anecdotal evidence, including that of Stan Ricker, just as Stan Ricker said the Soundfield 16/50K recording sounded the same as the mic feed in 1978 when he was younger and could hear better.
So, do you think the digital masters deteriorated over time? Because unless you do, your objections have no force at all.
Meyer and Moran said:
"The usefulness of the increased dynamic range afforded by longer word lengths for mixdown has never been in question."
Now, in a world of dual layer SACDs and issues of recordings in more than one stereo format, there is nothing unreal about what was tested.
You seem to have no evidence that there is anything wrong with the system described, and your only criticism of the Pioneer player is its price! Tsk, tsk, shame on you.
Again, I just pointed out that no one said the Pioneer player was perfect--indeed, Meyer and Moran pointed out a flaw, audible at very high levels, but which no one managed to hear at listenable levels on music. But then no one showed they heard a difference using a number of other players, either.
You have no objections against the data; you have no substantiated objections against the equipment used.
There have been suggestions that some changes would result in a better test. That's always possible.
You seem to wish something else had been tested, but haven't specified what or how.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
There is a big difference between 50 kHz sampling and 44 kHz sampling. In my experience, going to 48 kHz gives the majority of benefit from higher sampling rates. Beyond that it's a case of diminishing returns, but with modern technology the costs are so low that it makes no sense to go below 88.2 kHz, or even 176.4 kHz. (A few cents per album extra bandwidth and storage cost for an Internet download.) In my experience, sometime you can get good results with 44.1 kHz, but it's a puzzle. There are various choices in filtering that can give you distortion free sound (no aliases), full 20 kHz frequency response, or unsmeared transients (no excessive ringing). Unfortunately, it is only possible to have two out of the three possibilities at this sampling rate. If the recording is such that all three are not really needed, then the final results can be very good, but arguably not excellent.
Meyer and Moran follow an established tradition of audio denialists to dismiss positive findings. However, they are not an extreme examples of this error. There was an AES study that reached the conclusion in the summary that there were no significant differences between high res PCM and DSD, despite the body of the same article pointing out there were a few subjects who could clearly and reliably hear the difference.
If there were money in audio then there would be "scientific" research to "prove" what most of the mastering engineers already know. Most, not all of them, unfortunately. If you can hear the difference you don't need any "authorities" or "peer review" to validate your sense impressions.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"There was an AES study that reached the conclusion in the summary that there were no significant differences between high res PCM and DSD, despite the body of the same article pointing out there were a few subjects who could clearly and reliably hear the difference."
The Blech-Yang convention paper is interesting. I will have to read it more closely.
I have pointed out to E-stat the importance of considering the purpose of a test and the appropriate standards. Blech and Yang say that:
"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."
So, from a purely statistical point of view, there is a significant difference, so the authors do not disagree with you on that. They suggest, however, that the 4 Tonmeister students who did obtain significant results using headphones may have been hearing an artifact of the test process rather than a difference between DVD-A and SACD.
However, it would appear that the authors also applied another standard for some other purpose, I presume a practical one.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
The fact is that if 4 people can hear a difference, then it is likely to be audible. Even likely that the majority will be able to hear the difference once the "trick" of how to do so is explained. If you want to wallow around in the 50% percentile of excellence (a.k.a. mediocrity) be my guest. I prefer to reside in the upper half percent (a.k.a. excellence).
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It still remains to be shown just what it was those 4 heard. The difference(s) may not have been related to the recording format, as the authors pointed out.
I congratulate you on your excellence.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
If the authors had been scientists rather than students or if there had been any money in Audio, they would have gotten to the bottom of the matter.
Right now, I am listening to a Mozart Violin concerto performed by Joseph Fuchs and Eugene Gossens, transferred from the 35mm Everest master, digitized at 96/24 and purchased from HDtracks.com and downloaded while I was out to dinner. Nice. Definitely not a truncated experience. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Definitely not a truncated experience. :-)
is you don't choose to "wallow in mediocrity". :)
rw
...when he was younger and could hear better.
Speak for yourself if you have gained zero experience and have not improved your listening skills over a twenty year period. I'm sorry to hear that - but it is quite expected.
You seem to have no evidence that there is anything wrong with the system described
Not for shallow meter readers who require a test to tell them what they've heard. Only those who have been exposed to far better understand otherwise.
...-indeed, Meyer and Moran pointed out a flaw, audible at very high levels, but which no one managed to hear at listenable levels on music.
This is really getting pathetic. You really have no idea what kinds of audible differences exist between $250 and $25,000 players. Is the Rotel the best you've heard? I remain amazed at how much ink is spilled by those who speculate (based upon their non-experience) there aren't any differences beyond gross measures of level and frequency response among audio components. Your mission to spread the word of mediocrity falls on (not) deaf ears.
rw
Well, if you want to call hearing loss with age an assumption, that's up to you, but don't expect rational people to follow. You make an assumption that listening skills will make up for that. Maybe, maybe not. But still, you have no scientific tests.
"Not for shallow meter readers who require a test to tell them what they've heard. Only those who have been exposed to far better understand otherwise."
Maybe, maybe not. You're making an assumption. The only way to establish the issue rationally is with data from controlled blind listening tests, not sighted auditions. With sighted tests, one can distinguish the DUT without even operating them.
Your quote from me:
"...-indeed, Meyer and Moran pointed out a flaw, audible at very high levels, but which no one managed to hear at listenable levels on music."
E-stat's comment:
"This is really getting pathetic. You really have no idea what kinds of audible differences exist between $250 and $25,000 players. Is the Rotel the best you've heard? I remain amazed at how much ink is spilled by those who speculate (based upon their non-experience) there aren't any differences beyond gross measures of level and frequency response among audio components. Your mission to spread the word of mediocrity falls on (not) deaf ears."
I am not an issue here, neither is my equipment. I did not participate in the tests. The point is that a number of players were used in the tests and one was identified as not as good, using the ABX Comparator--as I already pointed out. Nevertheless, it was not shown that the player was not good enough for the uses for which it was designed.
I would like to see proof that a $25,000 player is audibly better. Have you got any established by scientific methods?
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Well, if you want to call hearing loss with age an assumption, that's up to you, but don't expect rational people to follow.
Neither you nor I can speak to Mr. Ricker's hearing ability, but only the inept would assume that the top half octave is where the heart of music lies. It does not.
I would like to see proof that a $25,000 player is audibly better.
I'm sure you would. That's what folks who never have extended exposure to such gear say to rationalize their inability to hear differences that are obvious to those who have. It is indeed difficult to relate the depth of an experience to someone whose frame of reference is completely absent. Imagination always comes up short.
Do continue flying the banner of mediocrity since that is all you know!
rw
I agree with Pat-D when it comes to $25,000 players. The pricing is just plain absurd, being several times the price of the most expensive converters used to make recordings. Surely far beyond the point of diminishing returns.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
on which you base your opinion? dCS? Burmester? EMM Labs? Esoteric? Others?
Pat's conundrum is that he doesn't (even begin to) know what he doesn't know. It is difficult to base an opinion on that which is completely outside one's frame of reference.
Did you ever ask Brad why he chose cheap consumer units to use as his reference points for "high resolution audio"? Is he even remotely aware of what is available?
rw
It's been years since I was at Brad's place. At the time he was touting how his Carver amplifier was perfect and all CD players sounded the same. He even demonstrated the latter. It wasn't a convincing demonstration, not because the players sounded different but because the music didn't sound as good as I could hear over at his ex-buddy Clark's place.
My comments on converters relate to the pricing of the top pro audio converters, such as the DAD AX24 vs. the high end consumer products, not sound quality. There is a long history of over charging for consumer products, since more money buys more ego satisfaction, something that is less important in the pro audio field where performance and reliability rank first.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It wasn't a convincing demonstration, not because the players sounded different but because the music didn't sound as good as I could hear over at his ex-buddy Clark's place.
There you go.
BTW, the EMM Labs players/DACs are pro units used by virtually all the SACD recording studios, including Sony.
rw
Why the hell would he bother with such a convoluted pipe-content-through-a-series-of-external-converters approach rather than directly comparing the same high quality feed through multiple native recording resolutions?
It makes no sense whatsoever. I am constantly amused by the contrived and indirect "kiss your elbow" test procedures some engineers go to in order to prove a point. His failed.
rw
"Why the hell would he bother with such a convoluted pipe-content-through-a-series-of-external-converters approach rather than directly comparing the same high quality feed through multiple native recording resolutions?"
Maslow's Hammer?
This was done some time ago, before the era of computer audio and widespread hi-res PCM recordings, high quality sample rate converter software, and widespread studio quality DACs. Sony and the RIAA can also be blamed for the SACD copy protection requiring the use of the "analog hole". Also, he had a hardware ABX box.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
This was done some time ago, before the era of computer audio and widespread hi-res PCM recordings
Is that how you would characterize 2007?
We're all deaf
rw
2007 was the publication date. I don't know when the actual experiments were performed. Thanks to the AES I would have to pay to download the copy of the paper that popped up on Google and unfortunately Brad mailed me a copy in hard copy a while back and I've misplaced it. I assume the paper itself will tell when the work was done, or at least the submission date of the paper to JAES.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
E Brad was comparing Redbook to SACD - which has been around for a decade now. If that's news to him, then he's not exactly with the plan.
rw
It was great taking a trip down memory lane! Ol' Andy19191 really cracked me up - and not just me, but some RE's as well. That was a hoot! Thanks! I do need to check the archives more often when I'm feeling blue. That cheered me right up! lol
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: