Hopefully he's learned something since then. I've never found the Redbook standard *perfect*.
-
Feeling lucky? Check out the AK 24th Anniversary Giveaway thread and join in the fun!
World's best drummer explains why analog is superior to digital
- Thread starter PanheadXR290
- Start date
guiller
Toscaninichus Australis
An argument from authority (argumentum ab auctoritate), also called an appeal to authority, or argumentum ad verecundiam, is a form of defeasible argument in which a claimed authority's support is used as evidence for an argument's conclusion. It is well known as a fallacy, though it is used in a cogent form when all sides of a discussion agree on the reliability of the authority in the given context.
Quoted from https://en.wikipedia.org/wiki/Argument_from_authority
Quoted from https://en.wikipedia.org/wiki/Argument_from_authority
KrisM
Lunatic Member
So, I'm not the only one struggling with the whole 'World's best drummer' thing?
wyn palmer
Addicted Member
That was one of the contradictory papers I was referring to, the other is the Meyer-Moran study.
http://hiresaudiocentral.com/the-controversial-meyer-and-moran-study-a-fresh-look/
https://www.sa-cd.net/showthread/58757/58908
See above for an interesting discussion of the paper and its results/defects and prior contraindications.
Please explain why is the real world bit depth inadequate and what, exactly, does "smeared" mean?
This horse is certainly not dead.
The late Charles Hansen of Ayre Acoustics introduced me to that study when it came out.
What a joke for so many reasons. Let's take content that was never recorded at high resolution, play it on a $300 Pioneer unit with no better performance than CD and use it as a reference point. LOL!
Yes it is.
wyn palmer
Addicted Member
By the way Monty is largely responsible for the Ogg/Vorbis codec. The codec largely depends on a detailed psycho acoustic understanding for its compression capabilities and undoubted success, so Mr. Montgomery is clearly not just an ex OS developer.
wyn palmer
Addicted Member
Fine, if you say so. However, it seems that many in the AES community who like to offer a little more, well, analytic substance, to their opinions, disagree with you, so the issue will continue to resurface.
wyn palmer
Addicted Member
Oh, and did you notice- and this is alluded to in the SA_CD.net discussion- many of the Pras and Gustavino results seem to be internally inconsistent and it's by no means certain that they meet rigorous standards of statistical significance. It's always tempting to accept a result just because it conforms to your opinions.
I'm neutral in this. I do believe that higher res matters and that upsampling and the use of intelligent interpolating filters like SOX makes an audible difference to Red book standard recordings, but I haven't been able to confirm it adequately to ensure that it's simply not confirmation bias on my part.
wyn palmer
Addicted Member
And by the way, David Moran strenuously disagrees with many of the objections to their study. You should read the thread on SA_CD.net if you have not.
There's no debating the facts I presented. View the content and equipment used. Dark Side of the Moon was not recorded in SACD. I, Robot was not recorded in SACD. Gaucho was not recorded in SACD. I could go on with others, but why?
wyn palmer
Addicted Member
http://www.aes.org/e-lib/browse.cfm?elib=18296
Above is a link to a download page for my favorite metastudy on the subject, dating from June 2016. It posits that Hi res is better and tries to create a proper statistical framework to assess that assertion based on a number of published studies.
A Meta-Analysis of High Resolution Audio Perceptual Evaluation
JOSHUA D. REISS,
A E S M e m b e r (joshua.reiss@qmul.ac.uk)
Queen Mary University of London, London, UK
It's certainly not entirely conclusive, and, in my opinion, it leans heavily on the Yoshikawa 1995 result which has also been the subject of considerable derision, but interesting nevertheless.
A quote from the conclusions:
"In summary, these results imply that, though the effect is perhaps small and difficult to detect, the perceived fidelity of an audio recording and playback chain is affected by operating beyond conventional consumer oriented levels. Furthermore, though the causes are still unknown, this perceived effect can be confirmed with a variety of statistical approaches and it can be greatly improved through training"
Above is a link to a download page for my favorite metastudy on the subject, dating from June 2016. It posits that Hi res is better and tries to create a proper statistical framework to assess that assertion based on a number of published studies.
A Meta-Analysis of High Resolution Audio Perceptual Evaluation
JOSHUA D. REISS,
A E S M e m b e r (joshua.reiss@qmul.ac.uk)
Queen Mary University of London, London, UK
It's certainly not entirely conclusive, and, in my opinion, it leans heavily on the Yoshikawa 1995 result which has also been the subject of considerable derision, but interesting nevertheless.
A quote from the conclusions:
"In summary, these results imply that, though the effect is perhaps small and difficult to detect, the perceived fidelity of an audio recording and playback chain is affected by operating beyond conventional consumer oriented levels. Furthermore, though the causes are still unknown, this perceived effect can be confirmed with a variety of statistical approaches and it can be greatly improved through training"
wyn palmer
Addicted Member
Again, read the thread if you have not.
Also, to say the least there's debate as to whether SACD is in fact superior to high res or even standard res PCM.
http://audioinvestigations.blogspot.com/2014/03/dsd-how-good-is-it.html
This is definitely not a closed subject.
I'm familiar with that as well:
"Overall, there was a small but statistically significant ability to discriminate between standard quality audio (44.1 or 48 kHz, 16 bit)and high resolution audio (beyond standard quality). When subjects were trained, the ability to discriminate was far more significant."
And you the one to which I linked in post # 145. You'll find additional commentary by John Curl, Ted Smith (of PS Audio), John Atkinson and E. Brad himself.
Do we really need another lossy codec?
wyn palmer
Addicted Member
One other thing, I do not believe that it is possible to mix tracks in conventional single level DSD (SACD) as it's not an inherently "linear" process. As far as I know all DSD recordings that require mixing are either done in analog prior to conversion to a bit stream or are processed in PCM then converted to DSD.
I have a 2nd or 3rd generation copy of Dark side of the moon on 15IPS 1/4" tape that sounds better than any LP or CD that I have heard of that material. I would assert that in that case, the source material was potentially adequate for a comparison.
wyn palmer
Addicted Member
Well, yes we did, and do. It's apparently audibly superior to the MP3 high quality one and is open source, so it has been enhanced and it comes royalty free. Remember, not everyone can either hear or if they do hear, care, about these trivial (to them) differences in sound. I have a 15 year old daughter that will not listen to my main audio system but prefers to listen to the 160kHz Ogg-Vorbis output on Spotify over her headphones even though she knows that it's very noticeably inferior. I recently pressed her to use the 320kHz codec version which is available on the premium tier (which I pay for) and she now does and she can tell the difference- but she really does NOT care and is just doing it because I insisted.
Last edited:
That is why Pyramix developed the DXD (352/24) format for editing DSD content.
Feel free to use circular reasoning.
wyn palmer
Addicted Member
And by the way, I had read the thread from Ayer etc. It's all very well, but ultimately it's a lot of opinions- on both sides- with very little in the way of scientifically valid evidence and conclusions. Again, I'm an agnostic, leaning towards the "more is better" side, but I find it fascinating that confirming to a degree of scientifically acceptable exactitude such a small matter has proven to be so elusive, even in the presence of trained observers.
wyn palmer
Addicted Member
And DXD is PCM- and the stream is only created at the end of the processing- so what's your point? And why is referring to one of the most widely accepted great sounding analog recording as a suitable reference point circular reasoning- particularly as this thread started with the assertion that "analog sounds better than digital"?
Only the obvious - several recordings were not originally made using high resolution digital formats.
Look up the definition of circular reasoning - yours is a textbook case.
Fortunately for the rest of us, the question has been put to rest by the recording industry. The vast majority is 24 bit and much of the better content available sampled at 88 or above. My attitude is summed up by mastering engineer Brian Lucey:
"Let’s just sell the 24 bit files at the mastering session sample rate, not higher and not lower, and call it a day? "
I'm over this back and forth. Yours is the last word.
wyn palmer
Addicted Member
Thank you. Let me say that I have no issue with higher than 16 bits being the standard- Monty discounted some of the dynamic artifacts that are created by sigma delta modulators, but I've seen instances where several bits of resolution are effectively lost due to what used to be called "idle tones", but that can be created when the modulator is in unpredictable states and/or the dither sequence is repeating, so 24 bits seems pretty safe. All of my resamplings are done to at least 24 bits, and all of my LP digitizations are 192k/24 bit PCM.
As has been the case, e-stat, it was fun. A good afternoon's entertainment was had.