Jump to content
IGNORED

To hi-res or not to hi-res, that is the question


Part-Time Audiophile
 Share

Recommended Posts

Or is it?

 

I think we, as computer audiophiles, take it as read that hi-res is, ceteris paribis, of superior audio quality than mere redbook. The devil is in the details, of course, and reading out everything that is meant, intended, implied and entailed with the "all things being equal clause" might take some serious unpacking before we can all agree, but I'm glossing over all that for now.

 

Because perhaps it doesn't matter at all:

 

http://theaudiocritic.com/plog/index.php?op=ViewArticle&articleId=4&blogId=1

 

I've never heard of this site (The Audio Critic), but this is just great reading. Anyway, if you can assume that a peer-reviewed (?) journal article by a couple of serious-minded folks with empirical bends can lend some weight to their conclusions, well, then I think we have arrived (again) at a bit of an impasse. Perhaps. Maybe. YMMV, &c &c. ;-)

 

Ok, now before everyone goes all googly-eyed, let me paraphrase from the clipping's conclusion: sound quality varies not with sample rate (Nyquist is as good as we mere humans can perceive, like it or lump it), but rather with recording quality, which as we all know varies wildly. Interestingly, and perhaps coincidentally with SACD (and other hi-res material), there is more focus on and attention paid to the quality of the recording of the material -- and its the result of this care that produces the biggest (and perhaps only!) gains to be found in the realm of hi-res.

 

Food for thought, no?

 

Link to comment
Share on other sites

....may well be able to help! They offer downloads of the same recorded material at different resolutions and some of it makes for very interesting listening indeed.

 

Personally, I prefer the 24/96 material but it is sometimes not as clear cut as I would like to think it should be! The recordings are exceptionally clear and well mixed and it makes for some very illuminating comparisons.

 

Link to comment
Share on other sites

So they took the analog signal from an SACD disk, reconverted to Redbook digital, and back to analog and hear NO difference? They must have a set of perfect converters, or a playback system incapable of exposing those differences.

 

Link to comment
Share on other sites

Hi-rez is sonically superior to 16/44.1. Many have posted on the CA site and elsewhere on this very subject and I believe if you research the provided documents and listen for yourself on a quality audio system you will concur.

 

On a more basic level I just want the same master files that the recording studios have. In the old days it used to be the master analog tapes with first generation copies or discs coming in a close second. If the recording studio masters are 24/176, DSD or DXD that to me is the best available. DXD downsampled to RBCD will never be as good.

 

The above not withstanding, implementation is everything. So if your hardware is not up to the task for hi-rez playback then for you perhaps the answer is not to hi-rez. Excellent RBCD playback will beat poorly implemented hi-rez playback everyday.

 

Of course the quality of the recording is paramount, garbage in equals garbage out, but look to the masters of recording and mastering for answers to your question.

 

 

Link to comment
Share on other sites

I think the point of the study in question shows something quite remarkably at odds with those opinions you referenced, but regardless, simply repeating those observations does not refute the study. And the study does seem to indicate that a "quality audio system" is immaterial. ABX studies are pretty funny that way.

 

Now, as to the quality of the recording, we are in agreement -- but where I think the you and the study are diverging is why that recording is better, and they submit that the reason is not due to the sample rate.

 

"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." (caps mine).

 

Link to comment
Share on other sites

Everyone's entitled to their own opinions but forgive me if I side with the masters in the recording studios instead of two audio journalists.

 

Are these guys really saying that there is no audible playback differences between the hybrid SACD layers of Red Book and DSD? If so these guys are not only double-blind they are double-deaf. I guess they just want to get noticed.

 

Link to comment
Share on other sites

No one has shown that ABX testing can discern differences in human cognition. There are many threads on this forum and elsewhere that have shown why the double blind tests are flawed. And the reasons why the double blind methodology consistently misinterprets human cognition. *Please refer to those (endless) discussions.*

 

Cognition and perception are incredibly complex and not suited to simplistic empirical measures/ methodology as are the physical sciences. Psychology measures constructs not facts.

 

And engineers/journalists shouldn't be doing psychological studies. Everyone can do psychology, right?

 

The Audio Critic has been beating the same dead horse for years.

 

Over and Out,

 

James[br]

Link to comment
Share on other sites

The link to the article is here:

 

http://www.aes.org/journal/online/comment/?ID=14195

 

And as far as I can tell, the primary objections to the "study" weren't due to the nature of ABX studies (and their endless variations), which tend to be fairly standard tools of psychology. Of course, there are many that say that psychology itself is rather a blunt instrument, but then, to them we must say, if all the knives in the drawer are dull, does that mean no dinner gets made?

 

Some say that humans are bad at ABX testing. Perhaps this is so. I suspect that this is facile, however, and there are rather strong theoretical reasons to suspect that humans are actually quite gifted in this regard. That does not, however, entail that there isn't a threshold of perceptibility nor that there aren't those (perhaps with either special training or natural gifts) that can ace such tests with ease.

 

But yes, a quick search of the site reveals many discussions and/or tangential/oblique references to the strengths and weaknesses of ABX testing. Pretty funny reading, really -- Chris, you really do have a wealth of info in here -- but again, not material to the point the non-PhD-and-therefore-non-credible authors of that paper which is that discriminating between source material sample rates is something of a fool's errand.

 

Believe, or not, as you choose. But no, no one is entitled to any opinion whatsoever. You simply have them or not.

 

As for myself, in my own A/B testing of cabling, I have some reason to think "differences" might be somewhat less than "expectations of differences". And on my own admittedly antiquated Accuphase DP-85, the differences between redbook and SACD are ... subtle.

 

Anywyay, I just thought the reference was interesting. Ahem! But in the interest of not revisiting territory so comprehensively covered, I'll refrain from further equine brutality.

 

Link to comment
Share on other sites

I didn't read the article, but at least I have always been saying that (which ever started at the "Is 192KHz pointless" thread).

 

But it got worse lately in my system, and upsampled redbook sounds BETTER than "native" hires now.

 

With this, notice that in all circumstances I use the exact same DAC at the exact same sample rate and all.

 

I have suggested a couple of times already that something is (going) wrong at the creation of hires. Something like filtering due to redbook is still active, or whatever it is that makes it flaw.

It must be so, since it is theoretically impossible that upsampling (though of a very special kind) is better than reality.

 

Peter

 

Lush^3-e      Lush^2      Blaxius^2      Ethernet^2     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
Share on other sites

Interesting article, but nothing unusual for The Audio Critic.

 

I first started reading The Audio Critic back in 1988 or 89. Peter Aczel takes great pride in cutting through the bull crap dished out to us by the audio industry. Many of his articles supply excellent information based on sound scientific principles and testing methods, while other articles can be highly opinionated rants, but always spiced with a dash of humor. I really like Issue No. 29 from 2000, “The 10 Biggest Lies in Audio.” Audiophiles are still debating the same old subjects today. Interesting and amusing reading for anyone with a passion for audio.

 

Link to comment
Share on other sites

This (the 16KHz roll off) is true for unfiltered - hence unreconstructed material. Worse ... this starts just under 5KHz, but at 16KHz it really gets bad.

 

But theoretically it would be true. Not in practice, unless you use NOS/Filterless.

 

With 24/96, the audible high freq can be reasonably retrieved.

 

Again true. However, this doesn't tell me the result is a reliable representation of true 96KHz sampling.

To my belief the 44KHz DOWNsampling is done in a decent way (by decent (over)developed means). So, never mind the result needs treatment - that treatment can be done because the base is consistent and is a representative from whatever the original sample rate was.

 

Some day we will know and understand.

Peter

 

 

Lush^3-e      Lush^2      Blaxius^2      Ethernet^2     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
Share on other sites

I have 3 versions of Claire Martin's 'Too Darn Hot' track:

1) 16/44.1 WAV ripped from CD (HDCD-encoded) - length 3:52.06

2) 16/44.1 FLAC downloaded from Linn (HDCD-encoded) - length 3:52.06

3) 24/96 FLAC downloaded from Linn - length 3:52.46

 

i've just compared these. To my ears, 2) sounds a tad drier than 1), but otherwise they sound pretty similar.

 

3) is CLEARLY better than 1) and 2). The difference is not subtle. The whole sound is more three-dimensional and organic.

 

@Peter

Quad Arc Prediction improves 1) and 2), but doesn't bring them up to the level of 3).

 

On a final note, you may notice that 1) and 2) are exactly the same length. 3) is not, which implies it has not been derived from an inferior format.

 

Mani.

 

Main: Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima speakers + 2x Rotel RB-1590 stereo amps -> 4x subs
Office: MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers
Vinyl: Thöress Phono Enhancer -> RME ADI-2 Pro

Link to comment
Share on other sites

Hahaha Mike. Sorry. But if I imagine what you have read about this ... no. It wouldn't be the whole story.

 

Anyone knowing about signal processing will show you that no roll of is there once the "filtering" is applied while that filtering is obligatory. Unless you're a true NOS/Filterless believer, *then* it holds, but also : then it is your own fault. So, 44.1 just NEEDS the filtering, or otherwise it would be disobaying some laws.

 

But then of course ... if those who invented the CD wouldn't have been so stupid to use 44.1 (but 96 or so), the whole problem of this "law of reconstruction" wouldn't have been there in the first place. And thus indirectly you are perfectly right. It is a poor solution. Btw why ? because otherwise our known vinyl album wouldn't have fitted on a CD. The guys just couldn't wait until they could do *that*.

 

What a world. :-)

 

Lush^3-e      Lush^2      Blaxius^2      Ethernet^2     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
Share on other sites

WRT my earlier post:

 

It may well be that 1) and 2) were 'downsampled' from a 24/96 master...

 

Mani.

 

Main: Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima speakers + 2x Rotel RB-1590 stereo amps -> 4x subs
Office: MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers
Vinyl: Thöress Phono Enhancer -> RME ADI-2 Pro

Link to comment
Share on other sites

  • 3 weeks later...

Hi to all,I came across this article recently and was looking for some where to make a comment about it

 

http://theaudiocritic.com/plog/index.php?op=ViewArticle&articleId=4&blogId=1

 

Even though this is 2 years old,it is mis-leading and this kind of article gives people the wrong interpretation

 

of things.They say that there is no perceived difference between Redbook 44khz cd and SACD/DVDA.

 

Well I think a point they miss is that when a CD is mastered at 24bit / 192khz for SACD it is then dithered or

 

down sampled to fit the 700mb limitation of a CD. And played back possibly via a 16bit / 44khz processor.

 

I still believe the trained ear can perceive a difference in quality of the recording.And this is where their

 

assumption that there is no difference is flawed,and it is some what similar to the analogy of an AVI vs a DVD/or even Bluray and DVD,it is file size.More capacity more info/higher sampling rates

 

An AVi is aprox 700mb / a DVD aprox 4.3Gb

 

A CD is aprox 700mb / a DVDA aprox 4.3Gb

 

A DVDA mastered at 24bit / 192khz on DVD media Can have a file size of up to aprox 4.3Gb

 

One I did was 5.7Gb and I put it on a dual layer DVDA

 

Now this type of disc played back on quality equipment /via 24bit / 192Khz Processors,has a very real

 

difference to the standard Redbook CD.And when they produced that Article in 2007,even then I had a

 

Processor that was capable of 24bit / 192Khz processing.

 

I would liked to have had a right of reply on that website to the article,unfortunately there was none

 

You can try what I say for yourself (if you have not already done so) here is what you need to do

 

Go to http://cirlinca.com/index.htm and download the free trial ,it will I believe allow you to make aprox 5

 

DVDA discs. Now there are a lot of options with this software/and Iam talking about remastering a Redbook

 

CD 16bit / 44Khz to 24bit / 192Khz Stereo DVDA . You can also remaster 2 channel stereo to 6 channel

 

This is a difference I perceive and Iam enjoying very much,the end result to me is very dynamic and the micro

 

details are so much more noticable /a perceived difference noticeable and real.Iam not one who likes to

 

pretend I here things.If it's not there it's not there/I hope this info is of use to someone/somewhere

 

Cheers Weeone

 

 

Link to comment
Share on other sites

My previous post is a bit all over the place,a little hard to read.

 

That is not how I typed it out /sorry if it is frustrating to read

 

P.s also did not notice the Audio Critic link on the first post

 

Cheers Weeone

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share



×
×
  • Create New...