Lies about vinyl vs digital

[Ralf11]

9 hours ago, sandyk said:

 

.. I doubt that your precious Checksums are anywhere near as precise as you claim, otherwise they would show the differences that I hear, AND have been verified by numerous members.

 

 

You're right and that is why the internet doesn't work

[The Computer Audiophile]

@sandyk enough about your bit identical files, it only leads to you raging beyond your control. This isn't a snark or joke. The past predicts the future. 

[esldude]

2 minutes ago, The Computer Audiophile said:

@sandyk enough about your bit identical files, it only leads to you raging beyond your control. This isn't a snark or joke. The past predicts the future. 

In this case, I fear the future predicts the past as well. 

[lucretius]

13 minutes ago, esldude said:

In this case, I fear the future predicts the past as well. 

 

Travers-able wormholes?

[fas42]

Of interest, I had a listen to the Vega Luka track, just from a YouTube clip - I was never into her music at the time. And it's obvious why people are thrown by it - there appears to be a nominal clash between the intimacy of her voice, and the highly reverberant backing. This was a production decision, and is completely acceptable, as a creative choice by the people doing the album, including Suzanne. So, it's up to the playback system to be able to replay that correctly; to provide the accuracy so that one can appreciate the two acoustic styles being overlaid like that - I have heard plenty of recordings that do similar types of things, that is, layer highly contrasting acoustics; this creates an interesting dynamic to the the sound - but it won't work as a listening experience if the replay system is not up to it ... of course, this means it's easy to use such tracks to test the competence of a rig one comes across, 😁.

[fas42]

Also, had a listen to the decoded Boxer, mp3 and flac ... too much damage has been done to the tone of the saxophone, on the left; listened to a straight YT clip to confirm this.

[John Dyson]

1 hour ago, fas42 said:

Of interest, I had a listen to the Vega Luka track, just from a YouTube clip - I was never into her music at the time. And it's obvious why people are thrown by it - there appears to be a nominal clash between the intimacy of her voice, and the highly reverberant backing. This was a production decision, and is completely acceptable, as a creative choice by the people doing the album, including Suzanne. So, it's up to the playback system to be able to replay that correctly; to provide the accuracy so that one can appreciate the two acoustic styles being overlaid like that - I have heard plenty of recordings that do similar types of things, that is, layer highly contrasting acoustics; this creates an interesting dynamic to the the sound - but it won't work as a listening experience if the replay system is not up to it ... of course, this means it's easy to use such tracks to test the competence of a rig one comes across, 😁.

Youtube has a  lot of feralA on it.  I was surprised by it myself.

 

John

 

[John Dyson]

1 hour ago, fas42 said:

Also, had a listen to the decoded Boxer, mp3 and flac ... too much damage has been done to the tone of the saxophone, on the left; listened to a straight YT clip to confirm this.

Sometimes, perceived damage is correction back to the original sound.  I have had real musicians check the results of feral decodes -- not just listeners.  I don't always get it right -- there are so many variables, and the truth is that the MF correction is very tricky.

 

I just did a careful A/B comparison and correction of the ABBA decodes (compared against actual vinyl), and have had to make a few changes.   Reference copies back to vinyl are critical...  And yes, almost EVERY ABBA CD sounds nothing like the original authors intent vinyl.   Why do I know that it is probably authors intent? -- one reason is because I have multiple mastered copies of different vinyl releases for reference.

 

I do re-calibration back to original vinyl when I start again on another set of ABBA examples, and the effort just started.  The only results which will be different from original vinyl end up being a set of tradeoffs between some CD releases and original vinyl -- because some (one esp, guess...) of the albums on vinyl suck badly.  A few tradeoffs to give in to a bit brighter high-end are based upon some of the highly-variable-quality CD releases.  (Not being able to tell the difference between the myriad of CD releases says -- just give up now!!!  ABBA is higly variable.)   Even some of the Polar CD releases are far, far away from the original vinyl.

 

It isn't ANYTHING about our immediate hearing or 'sounds good to me' as much as it is accurate representation of the original recording.  Sometimes feralA DOES sound better, but is seldom accurate at all relative to the original recordfing.  Sounds good and frequency response/spectral curves are INTERESTING, but only useful when comparing against a stable, properly handled reference source.  My goal is for archives like LOC and others that just might be using the DHNRDS for archival purposes.  We don't specifically have LOC yet, but hoping to make sure that the quality is proven well enough.

 

Suzanne Vega is an example of something that is only slightly more intense on CD than it should be.  Other material, like S&G is sometimes  really mangled.  Hiss should not usually be intrusive even on recordings from the 1960s...  If there is noticeable hiss or room vibrations, like at the beginnnig of S&G songs, it is often feralA compression.  No right-minded mastering engineer would allow the hiss or relatively strong room vibrations as a matter of mastering choice -- feralA is a plausible reason though.

 

Again, it isn't how it sounds to you or me, but it is accuracy relative to the acutal recording going into the DolbyA unit.  most often, original vinyl comes closest to authors intent.   Also, I can hear DolbyA decoding defects, and find that if they are removed -- then it is best chance come close to the original recording.  That is very different than personal taste or 'sounds good'.   'Sounds good' or 'sounds like what someone thinks that it should sound like' doesn't show accuracy.   The recording results really should based upon what a DolbyA encodes.  When we have original  vinyl, it makes the target much easier for the older DolbyA material.  (we often don't have a copy of the pre-DolbyA material -- seldom exists, so must depend on other properly DolbyA decoded examples, like ORIGINAL VINYL.)

 

So, when I decode material without a guideline -- I listen for decoding defects, and not 'sounds good'.  'Sounds good' is based upon personal taste and perception biases, but removing defects is something that any reasonable person prefers.   Hiss and compression effects are defects when it is from undecoded material.

 

 

John

 

[fas42]

Just on the S& G - I have a very ordinary CD of Bridge Over, early but a budget release of it. My version of this album tends to sound hissy, and a bit mediocre when my system is sub-par - it requires a high state of tune for the oomph of these tracks to kick in. When at that level, the emotional impact is truly remarkable ... I mentioned some time ago hearing his version of this album at the audio friend down the road, when his setup was in good enough tune, and the "big" songs on the album were truly cathartic ...

[John Dyson]

4 minutes ago, fas42 said:

Just on the S& G - I have a very ordinary CD of Bridge Over, early but a budget release of it. My version of this album tends to sound hissy, and a bit mediocre when my system is sub-par - it requires a high state of tune for the oomph of these tracks to kick in. When at that level, the emotional impact is truly remarkable ... I mentioned some time ago hearing his version of this album at the audio friend down the road, when his setup was in good enough tune, and the "big" songs on the album were truly cathartic ...

You just might have a feralA album.  They do exist.

I appreciate your emotional response about music, and there can be aspects that I am not involved with -- however, part of our disagreement might be more about our goals.

Accuracy is not always 'prettiness'.

 

My goal is related to the researcher 30yrs from now, with a copy of the vast feralA or true DolbyA archives, and wants to have an idea of the nuances of the music from 20-30+yrs ago (or 60+yrs ago from then.)   That can be a somewhat different goal than listening for things that a person likes.

 

I was just discussing elsewhere, where there are aspects of EQ for mastering that are more complex than just applying 'tone control'.  When mastering (which I seriously avoid, simply don't do actually), the material is modified for EVERYONE who each hears diffierently.  When listening at home, there is so much more freedom to adjust the recording.

 

When working with something like DolbyA, the set of limitations (before decoding) is highly restrictive.  It is important to correct a frequency domain (and phase somewhat) distorted signal into the basic DolbyA signal before decoding.  If those distortions remain or if the correction is substantively in error, then the DolbyA decoding results will be less than optimum.  Also, there are limitations to the useful EQ after decoding, if the EQ is intended to correct a pre-decoding EQ.  Any EQ BEFORE decoding has different effects than the same EQ AFTER decoding.

 

It is a complex thing, and I simply don't have the freedom to use a 'tone control' or use parametric or shelving EQ without following some rules.  Otherwise,the results become impossibly wrong, and can never be corrected with subsequent tone control.

 

If the rules are carefully followed, then the results are more likely to be tolerated by a larger set of people.  I remember when CDs first came out, there were all kinds of complaints about the sound quality, harsh 'digital' sound, 'compressed' sound, messed up highs, etc.  Most of those problems were not resolved -- because most of them were effects of feralA.   Back then, the distributors DID NOT follow 'the rules'.  If properly decoded, then the result is maximally flexible, including using for-purpose compression if the compressed sound is desired.

 

Since my goal is for archives and future research, they are sometimes different than an immediate 'sounds good' to any one person.  Some people enjoy the sound of dynamic range compression -- really, I had discussions even back in the late '70s about that!!!

 

John

 

[John Dyson]

Here is a potentially interesting comparison between 2 vinyl versions, several CD versions, two alternative DHNRDS decodings of a snippet of ABBA SuperTrouper starting at approx 13-14 seconds for about 40 seconds each.  I made these 'flac' so that people wouldn't worry about mp3 issues.  (There are some valid problems with audible mp3 lossage -- esp the timing accuracy of the DHNRDS.)  The 'mellow' DHNRDS is NOT accurate to the original ABBA, but is the direction, maybe a bit more bright, that I kind of prefer for my self.  ABBA is mastered way too harsh and aggressive for the actual beauty and cheer (sometimes sadness) in their music.

 

ADDED NOTE:  these CDs are either correctly mastered or perhaps compressed feral in the case of the remasters.   The original sources that I used for DA decoding were feralA issues selected from the best available quality for decoding purposes. I can provide reference IDs for the feral CDs.

 

The SuperTrouper album tends to be fairly constistent across versions unlike, for example Ring Ring.  Some of my versioning info is offline, but here are descriptions of the versions:

 

CD polar original

CD polydor original

CD discomate original

CD 1997 remaster

CD 2001 remaster

CD TCSR remaster (hyper compressed)

vinyl1 more recent version

vinyl2 older version

DHNRDS

DHNRDS mellow

 

If you compare the DHNRDS vs the others, the big differences is that transients are maintained...  That is part of the DolbyA fog.

But, perhaps more interestingly, the SuperTrouper album is relatively consistent, and you can hear the differences also.  But, in common, there is a latent compression left over in DolbyA because of the slow response through the expansion feedback loop.

The 'mellow' version is an attempt to control the intensity of the ABBA -- but it still doesn't have the right balance -- it needs a bit more brightness.

 

But, always -- I refer to the original versions for reference.   The DHNRDS will never have the compressed sound -- experiments show that the DHRNDS is very very true to the material being encoded.  The DolbyA has a delay in the deocding feedback loop that causes myriad of problems.  ABBA material is especially problematical and one reason why I use it for testing.

 

The maintenance of detail is one reason why the DHNRDS is interesting for archival purposes.  it also maintains quality for multiple sequential encodes and then multiple sequential decodes -- proving that it is accurate, and also it does maintain the details, transients, etc.   Also, it has a similar tonal balance to a true DolbyA.

 

(I was going to attach the files until I realized that they were 5MB each, so here is a pointer to them):

https://www.dropbox.com/sh/sm7ovodlqg0huck/AACMYk-JV64xZ2spSolgzGePa?dl=0

 

John

 

[sandyk]

3 hours ago, fas42 said:

My version of this album tends to sound hissy, and a bit mediocre when my system is sub-par -

Even "The Sound of Silence" on the Sony Blu Spec -Rock Vol.1 comparison album is " feral" Dolby A as can be clearly heard with the huge amount of sibilance evident .

 John verified this an hour or so ago, so perhaps it's no coincidence that your copy may sound a little hissy.

[fas42]

28 minutes ago, sandyk said:

Even "The Sound of Silence" on the Sony Blu Spec -Rock Vol.1 comparison album is " feral" Dolby A as can be clearly heard with the huge amount of sibilance evident .

 John verified this an hour or so ago, so perhaps it's no coincidence that your copy may sound a little hissy.

 

The hiss is that of the tape master. Unpleasant sibilance is one of the most obvious markers that there is excess distortion in the playback chain, and one of the characteristics of the playback quality I pursue is that this issue completely disappears, as something that is subjectively disturbing.

[sandyk]

15 minutes ago, fas42 said:

Unpleasant sibilance is one of the most obvious markers that there is excess distortion in the playback chain,

In which case you have a pretty pathetic playback chain.

Whether you wish to believe it or not, this track was incorrectly decoded, as so often happens with compilation albums , and I have since heard it after correction, with a marked improvement in soundstage as well.  However a small amount of sibilance still remains.

[Confused]

22 hours ago, John Dyson said:

Here is a decoded Luka snippet.  I didn't do the entire album with the new style of formulas, so there might be more quality issues than if I consider the entire album.  (Note that when I do decodes, I am not looking for 'good sound' as much as avoiding defects.)  Sometimes, I don't even remember the material when concentrating for problems -- weird, huh?

Comment -- I do believe that Vega's recording should be a little more hot/brighter than something like S&G, so I tried to make sure that I didn't dull the material too much.  I did a straight -1, -2, -1, -2, -1, -2, -1, -2 EQ -- which seems to be a relatively common EQ value (the freqs are (4250, 4500, 5750, 6000, 8750, 9000, 11750, 12000), all Q=0.50.  Almost all (I mean ALL) decodes need EQ at these frequencies or 0dB sometimes for 4250/4500 and once in a LONG LONG while, 0dB at 8750 on up.  Note that the 250 Hz offsets are totally critical -- amazing, eh?

 

 

Luka.mp3 2.1 MB · 2 downloads

This sounded great to me, but in terms of tonality I personally found it a touch on the bright side, not excessively so, but personally I would have probably dropped the HF a little. 

 

As it happens, I have a copy of this track the "KEF 50 Years" compilation CD that came with my speakers, in terms of tonality, this sounds very similar to your snippet.

[semente]

9 hours ago, fas42 said:

Just on the S& G - I have a very ordinary CD of Bridge Over, early but a budget release of it. My version of this album tends to sound hissy, and a bit mediocre when my system is sub-par - it requires a high state of tune for the oomph of these tracks to kick in. When at that level, the emotional impact is truly remarkable ... I mentioned some time ago hearing his version of this album at the audio friend down the road, when his setup was in good enough tune, and the "big" songs on the album were truly cathartic ...

 

And the high state of tune reduces the "hissy", right? Wrong.

 

The emotional level is in the music, not the sound by the way.

[John Dyson]

22 minutes ago, Confused said:

This sounded great to me, but in terms of tonality I personally found it a touch on the bright side, not excessively so, but personally I would have probably dropped the HF a little. 

 

As it happens, I have a copy of this track the "KEF 50 Years" compilation CD that came with my speakers, in terms of tonality, this sounds very similar to your snippet.

Man -- you get it entirely about trying to decode the material...   It is really hard to do and trying to remove my own taste and to make the 'right' decision is tricky.   Your thoughts about the sound being too bright or too dull hit the nail on the head.

If I can get this decoder into a more feralA usable state, it will make it practical for people to use the decoder for feralA, and if there is some per-user taste involved, then that will be accomodated.

As it is, it is exquisitely easy to use for pro material, but the feralA has an additional layer of filtering that makes the problem tricky at best.

At least, now I have a base set of filters, ESPECIALLY with the ability to play the audio realtime in the decoder, then tweaking in the correct EQ should be a matter of selecting up to 5-6 variations to get a 'good' decode, and perhaps a couple more for precision.   A 'good' decode being a little bit in error is tantamount to a tone control

 

I plan to make it available for consumer gratis (modulo a few restrictions against commercial use and maybe a license trick to support that.)   For commercial use, it is only in the US$300 range for perpetual use as a commercial nuisance fee, and any of my talk here is not of commercial interest, so I don't feel a conflict.  (I try to be ethical, even though sometimes a bit free with 'examples'.)

 

John

[Rexp]

1 hour ago, Confused said:

This sounded great to me, but in terms of tonality I personally found it a touch on the bright side, not excessively so, but personally I would have probably dropped the HF a little. 

 

As it happens, I have a copy of this track the "KEF 50 Years" compilation CD that came with my speakers, in terms of tonality, this sounds very similar to your snippet.

Have you got the vinyl? Would be interested in a comparison. 

[John Dyson]

9 hours ago, John Dyson said:

Here is a potentially interesting comparison between 2 vinyl versions, several CD versions, two alternative DHNRDS decodings of a snippet of ABBA SuperTrouper starting at approx 13-14 seconds for about 40 seconds each.  I made these 'flac' so that people wouldn't worry about mp3 issues.  (There are some valid problems with audible mp3 lossage -- esp the timing accuracy of the DHNRDS.)  The 'mellow' DHNRDS is NOT accurate to the original ABBA, but is the direction, maybe a bit more bright, that I kind of prefer for my self.  ABBA is mastered way too harsh and aggressive for the actual beauty and cheer (sometimes sadness) in their music.

 

ADDED NOTE:  these CDs are either correctly mastered or perhaps compressed feral in the case of the remasters.   The original sources that I used for DA decoding were feralA issues selected from the best available quality for decoding purposes. I can provide reference IDs for the feral CDs.

 

The SuperTrouper album tends to be fairly constistent across versions unlike, for example Ring Ring.  Some of my versioning info is offline, but here are descriptions of the versions:

 

CD polar original

CD polydor original

CD discomate original

CD 1997 remaster

CD 2001 remaster

CD TCSR remaster (hyper compressed)

vinyl1 more recent version

vinyl2 older version

DHNRDS

DHNRDS mellow

 

If you compare the DHNRDS vs the others, the big differences is that transients are maintained...  That is part of the DolbyA fog.

But, perhaps more interestingly, the SuperTrouper album is relatively consistent, and you can hear the differences also.  But, in common, there is a latent compression left over in DolbyA because of the slow response through the expansion feedback loop.

The 'mellow' version is an attempt to control the intensity of the ABBA -- but it still doesn't have the right balance -- it needs a bit more brightness.

 

But, always -- I refer to the original versions for reference.   The DHNRDS will never have the compressed sound -- experiments show that the DHRNDS is very very true to the material being encoded.  The DolbyA has a delay in the deocding feedback loop that causes myriad of problems.  ABBA material is especially problematical and one reason why I use it for testing.

 

The maintenance of detail is one reason why the DHNRDS is interesting for archival purposes.  it also maintains quality for multiple sequential encodes and then multiple sequential decodes -- proving that it is accurate, and also it does maintain the details, transients, etc.   Also, it has a similar tonal balance to a true DolbyA.

 

(I was going to attach the files until I realized that they were 5MB each, so here is a pointer to them):

https://www.dropbox.com/sh/sm7ovodlqg0huck/AACMYk-JV64xZ2spSolgzGePa?dl=0

 

John

 

Correction: I just realized after reviewing my post.  Remember I mention about my hearing adjusting to material?  The polar, polydor and discomate CDs plus the vinyl are feralA!!!   Last night when I did the posting, I attributed the compression sound to dolbyA fog by mistake.  I get foggy myself.  I knew that there was a problem somewhere -- listen carefully for the compression.

 

I fully expect that the reason why the SuperTrouper albums are feralA is that supertrouper is a pain in the b*tt for DolbyA -- it is almost the worst case of all kinds of detail in the background an lots of middle highs that naturally intermodulate.  If one looks at and understands the DolbyA detector front-end, it is easy to see how the intermodulation happens. (The intermod is created by a wobbling gain control -- the lower 3k-9k and 3k-20k detector frequency ranges are slow enough to get through to the diodes and the IMD happens at the diode/output of the detectors.)

 

It is very likely that except the very first album releases & music videos which might have a different path, that the DHNRDS is providing one of the few fully decoded copies of SuperTrouper available.

 

I have added a TRULY decoded copy of the SuperTrouper snippet from ABBA The Hits CD.   Note that it sounds similar in some ways to the feralA, except the dynamics are more pronounced.   The DHNRDS still maintains more detail and clarity.  The DolbyA detector is intermodulation-city, but the DHNRDS DA detector, by virtue of the digital design, allowed me to do an intermodulation-free detector that has the same dynamics...

(Believe me -- I have a LOT of material for reference -- I take this seriously.  Too bad I cannot do as complete research on every group.)

 

Here is the comparison archive again...  The only decoded copies are:  st-dhnrds, st-dhnrdsMELLOW, st-thehits...

https://www.dropbox.com/sh/sm7ovodlqg0huck/AACMYk-JV64xZ2spSolgzGePa?dl=0
Sorry about any confusion.

 

John

 

 

[John Dyson]

36 minutes ago, Rexp said:

Have you got the vinyl? Would be interested in a comparison. 

VInyl can be helpful, but as I just got caught by the ABBA stuff, there IS some feralA vinyl!!!  I am actually disappointed about it..  Well mastered material should be treasured.

 

Finding reference copies is hit or miss, and I treasure any reference snippets that I have access to.  For the 'crime' stuff, I had to rely on some hyper-picky artist/audiophile types to help  zero-in on the correct corrective EQ config.

 

John

 

[John Dyson]

7 hours ago, sandyk said:

In which case you have a pretty pathetic playback chain.

Whether you wish to believe it or not, this track was incorrectly decoded, as so often happens with compilation albums , and I have since heard it after correction, with a marked improvement in soundstage as well.  However a small amount of sibilance still remains.

Your comment about compilation albums being problematical at times is 100% the truth.  Someone who did 'greatest hits' type compilations told me first hand about the problems with the myriad of sources.

 

One reason why the DHNRDS can cleanly process almost anything is that it doesn't create IMD from recordings that have calibration errors (that is the DHNRDS is somewhat robust when it comes to matching the compression/expansion curve.)

* cleanly processing the material does not guarantee accuracy -- it just means that excess distortion doesn't build-up.

 

Sadly, the state of pop recordings is generally poor.  A lot of old master tapes were destroyed, and in some cases, the feralA masters might be the only thing left....

 

John

 

[4est]

On 5/17/2019 at 9:12 PM, lucretius said:

 

That's makes sense, since Alex is Barry's number 1 fan.

Barry deserves a bit more respect than this...

[sandyk]

7 hours ago, 4est said:

Barry deserves a bit more respect than this...

https://forums.stevehoffman.tv/threads/essential-barry-diament-masterings.131468/
https://forums.stevehoffman.tv/threads/essential-barry-diament-masterings-part-2.220334/page-12

[daverich4]

1 hour ago, sandyk said:

https://forums.stevehoffman.tv/threads/essential-barry-diament-masterings.131468/
https://forums.stevehoffman.tv/threads/essential-barry-diament-masterings-part-2.220334/page-12

Thanks for the links, I’m going to try and hunt some of those down. 

[fas42]

11 hours ago, semente said:

 

And the high state of tune reduces the "hissy", right? Wrong.

 

The emotional level is in the music, not the sound by the way.

 

No, correct. A low state of tune means that all the sounds in the recording, whether part of the intended recording. or artifacts of the capture and mastering process merge into a soup, there is poor separation of the elements that make up what you hear subjectively - the "hiss" is part of the music, as far as how it comes across.

 

Improve the state of tune, and the irritating anomalies are now heard more clearly, easier to note as being something "that shouldn't be there!" ... this is the classic, normal, audiophile, "most recordings are crap!" zone.

 

Final and essential, for me, step in improvement means enough clarity for the listening brain to understand, unconsciously, that the hiss is not part of the music, and it effortlessly discards it - this is the cocktail party effect kicking in, making sure that what is irrelevant in the sound field doesn't detract from what is of conscious interest.