Skip to main content

Topic: Waveform Plots Considered Harmful (Read 14407 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Recent talk about mastering quality has compelled me to commit this to my blog:

http://audiamorous.blogspot.com/2008/09/wa...ed-harmful.html

Summary: Anybody who waves an Audacity screenshot around saying how good/bad a mastering is will be summarily punched in the face.


  • Lyx
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #1
I think one needs to differentiate here.

Amplitude graphs are a very good way of explaining the problem - the overall principle. But they may not be a good way of measuring and detecting it in practice.
I am arrogant and I can afford it because I deliver.

  • uart
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #2
So does that mean that when you correctly rip a CD and get waveforms that look like the following, that they may actually be ok?

I ripped this a few days ago and frankly I was horrified. I thought it must be trash, like some sort of audio vandalism where the person doing the mastering must be a clueless loser. Can it really be that the people mastering this stuff do actually know what they’re doing and that they are carefully monitoring the distortion levels that are caused by this clipping and keeping it within some acceptable bounds?

  • Last Edit: 09 September, 2008, 08:43:39 AM by uart

Waveform Plots Considered Harmful
Reply #3
EDIT: Massive self edit to save falling out with Axon. If you are the author of that piece and I haven't seriously misinterpreted the suggestion that deliberately adding clipping to a waveform is in any way acceptable, then you really don't want to hear my opinion on it! All I will say is that it artificially generates a stage of lossiness in a digital system that is perfectly capable of being, to all intents and purposes, lossless.

Cheers, Slipstreem. 
  • Last Edit: 09 September, 2008, 09:43:32 AM by Slipstreem

  • 2Bdecided
  • [*][*][*][*][*]
  • Developer
Waveform Plots Considered Harmful
Reply #4
http://audiamorous.blogspot.com/2008/09/wa...ed-harmful.html

Summary: Anybody who waves an Audacity screenshot around saying how good/bad a mastering is will be summarily punched in the face.
You can hardly tell people not to use waveform plots, and prove the folly of this by using a waveform plot...!

The point is not to misinterpret them. As you have shown here...
http://2.bp.blogspot.com/_biABNEep_j0/SMYs...yendecker-2.png
...waveform plots, carefully examined, can be perfect for teasing out what has really happened. Not always, but often.

btw, I have CDs where the "clipped" peaks look exactly like the clipped peaks from your vinyl example, because the clipping occurred before other processing. I found such "clipped" peaks because I heard the problem, and went looking for it.


I don't have the Bob Katz book you quote, but you appear to misuse the quote itself...
Quote
Bob Katz has commented (in Mastering Audio) that "a rule of thumb is that short duration (a few milliseconds) transients of unprocessed digital sources can be reduced by 4 to 6 dB with little effect on the sound." Even 1ms of clipping is 44 samples of CD audio, and relatively few examples of clipping exceed even that.
"reduced by 4 to 6 dB" does not equate to "clipping".

DC shift aside, runs of clipped samples are effectively muted samples. There is no audio there. So the question becomes, if you hide the DC shift, how many samples can you set to zero before the effect becomes audible. This is quite well researched in psychoacoustics - the answer is totally content dependent, but it can easily be a couple of milliseconds or more. This is for something like white noise, with a single mute - not a repeating pattern.


Quote
Instead, clipping in this case may at worst sound like a change in timbre - annoying, to be sure, but potentially not an issue for most listeners.
Most listeners have speakers that will "change the timbre" dramatically more than any record producer would choose to - that doesn't mean that those of us with decent speakers should also have to suffer.

Quote
Perhaps the most accurate method of evaluating mastering quality, though, is the simplest: Asking the mastering engineer.
Funny, I use my own ears, not someone else's!

Quote
the sound quality of music today is still considerably better than what it was for (most of) the last 60 years.
I think the kind of music that's trashed today would probably have been trashed in a different way had it been made in the past. Many (not most) pop records from the 1960s were fairly lo-fi. However, it's quite easy to find recordings from the late 1950s which sound far better on a decent stereo than typical modern pop recordings. The chance of being able to listen to a pop recording on a decent stereo and actually enjoy it is probably lower now than it's ever been.

Cheers,
David.

  • uart
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #5
All I will say is that it artificially generates a stage of lossiness in a digital system that is perfectly capable of being, to all intents and purposes, lossless.

Cheers, Slipstreem. 


Exactly. And the thing I find hard to understand is why the heck anyone would want to make that compromise for an extra 3 or 4 dB gain when we've now got a thing called an amplifier and volume control!

The crazy thing is, if it was 100 years ago and we were all listening to records played unamplified though a passive horn with no volume control then it would actually make sense to make a compromise like that.

Maybe this is what the modern producers think were now using

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #6
Recent talk about mastering quality has compelled me to commit this to my blog:

http://audiamorous.blogspot.com/2008/09/wa...ed-harmful.html

Summary: Anybody who waves an Audacity screenshot around saying how good/bad a mastering is will be summarily punched in the face.



hmm....manually compressing a face....that's a little extreme   

Quote
Quite simply, the waveform plot is a worthless tool for evaluating mastering quality and its use for that purpose should be banished.


I think it would be more accurate to say, subjectively 'bad sounding mastering' can usually be convincingly correlated after the fact with some quality of a waveform or spectral view; but be careful going the other way -- I have seen people tout waveforms with relatively minor, and likely inaudible, amounts of clipping, as proof that the track sounds bad.

What 'sounds good' is so highly influenced by audio and non-audio factors that I really have to wonder whenever I read rave (or neg) reviews of music these days. Example: since I began ripping DVD-As and 'laserdropping' SACDs, it's been illuminating to look back and see how many of these often highly compressed (though rarely clipped) stereo versions were praised to the skies as the best ever masterings so far...by people who claim to loathe compression when they see it on CD waveforms.  To me, that's gotta be some combination of expectation bias (hi rez = good sound) and actual pleasure with the sound.

As for 'ask a mastering engineer', and using Hoffman as a paradigm, given his lifelong aversion to digital (he wouldn't even make a single digital edit to join the two masters of side one of 'Thick as a Brick', claiming it degrades the sound), and some of the bogosity he's spouted about it, I'd hardly consider him a go-to guy on the matter.  I'd go to Katz, who actually accepts that digitally compressed audio can still sound great.

Have you posted your links and thoughts in a thread on Hoffman's forum?  That should be, um, interesting to watch, if so. 
  • Last Edit: 09 September, 2008, 01:08:34 PM by krabapple

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #7
So I'll start by saying that I wrote that entire post between the hours of 1-5am and I was more than a little punch drunk. That said, I would like to clean it up and distribute it more widely (once a ton of stuff is fixed).

Thanks for the responses y'all. 

So does that mean that when you correctly rip a CD and get waveforms that look like the following, that they may actually be ok?

I ripped this a few days ago and frankly I was horrified. I thought it must be trash, like some sort of audio vandalism where the person doing the mastering must be a clueless loser. Can it really be that the people mastering this stuff do actually know what they're doing and that they are carefully monitoring the distortion levels that are caused by this clipping and keeping it within some acceptable bounds?
It's possible, but a lot depends on the context.

Clipped sine waves - and more generally, periodic clipping - tends to be fairly horrifically audible; the clipping harmonics extend so far out of the original signal band that the sensitivity to that sort of thing is tremendous.

OTOH, clipped transients - in the limiting case, something like a single clipped sinus - are relatively inaudible, because the transient content is so smeared in the frequency domain to begin with. At a high enough level, you can reduce the power enough to cause a dynamic range reduction, and/or change the timbral character of the transient, but it's not exactly something like "shave 0.1db off and you'll have serious problems." Whereas shaving 0.1db off a sine wave is a very big deal.

In the context of your waveform, the big clipping sections are 13ms apart, which sort of implies a large 76hz component - some sort of drum work, I'd completely guess? If it were a bass guitar you'd clearly have issues, but if it were percussion, I could go both ways on this. On one hand, the clipping is only about 1-2ms each time and it looks vaguely about 3-4db down from true peak. OTOH, the low frequencies involved, and the expected resonance of the drum, means that the clipping may be almost tonal in character. And when the clipping does happen, the harmonics will start at 152hz and plow on up from there.... which could totally suggest a sort of "congested" feeling  in the music - which I'm guessing you've noticed  But that could just be expectation bias talking.

I'm switching sides so fast it's making my head spin. One of the points I'd like to make more clear in my post is that there really aren't any silver bullets here. That is, it is plausible that in some cases, the mastering engineer knew what he was doing all along, and that the clipping is not audible. In many other cases, the exact opposite is true. And there's really no completely trustworthy way of telling the difference, with a computer tool, yet. Follow your ears, I guess.

EDIT: Massive self edit to save falling out with Axon. If you are the  author of that piece and I haven't seriously misinterpreted the  suggestion that deliberately adding clipping to a waveform is in any way acceptable, then you really  don't want to hear my opinion on it! All I will say is that it  artificially generates a stage of lossiness in a digital system that is  perfectly capable of being, to all intents and purposes, lossless.
Dude, I was kind of expecting to be walking into a hornets nest on this one.  As I mentioned above, my argument for the inaudibility of clipping is more or less restricted to strongly transient content.
  • Last Edit: 09 September, 2008, 01:20:40 PM by Axon

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #8
That post needs some serious quote-tag validation, dude.  ;>

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
Waveform Plots Considered Harmful
Reply #9
The max is ten quotes per page.  If you want more you can post in pieces and if you do it quickly enough they will be combined.
13 February 2016: The world was blessed with the passing of a truly vile and wretched person.

Your eyes cannot hear.

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #10
I apparantly discovered that independently, greynol.  Thanks.

You can hardly tell people not to use waveform plots, and prove the folly of this by using a waveform plot...!  The point is not to misinterpret them. As you have shown here... [a href="http://2.bp.blogspot.com/_biABNEep_j0/SMYs...yendecker-2.png" target="_blank"][a href="http://2.bp.blogspot.com/_biABNEep_j0/SMYs...yendecker-2.png" target="_blank"][a href="http://2.bp.blogspot.com/_biABNEep_j0/SMYs...yendecker-2.png" target="_blank"]http://2.bp.blogspot.com/_biABNEep_j0/SMYs...yendecker-2.png[/a][/a][/a] ...waveform plots, carefully examined, can be perfect for teasing out what has really happened. Not always, but often.
Good catch. I worded that in a perhaps far more inflammatory fashion than I probably should have. Still, I would like to make it clear that waveform plots are very untrustworthy. Even those zoomed-in plots, at a cursory glance, seem to confirm that the dynamic range is higher on the vinyl version - quite a few clipped regions on the CD do not show any obvious signs of clipping in the vinyl. So I think I can still reasonably end the title with "Considered Harmful".
 
Quote
btw, I have CDs where the "clipped" peaks look exactly like the clipped peaks from your vinyl example, because the clipping occurred before other processing. I found such "clipped" peaks because I heard the problem, and went looking for it.
That's great, but you're also a competent, professional audio engineer who knows what to look for 
 
Quote
I don't have the Bob Katz book you quote, but you appear to misuse the quote itself...
Truth be told, I don't have it either. I got lucky on a google search and came up with the quote in Google Books. I might have a relative who has it though....
   
Quote
"reduced by 4 to 6 dB" does not equate to "clipping".
I will admit to equivocating between "clipping" and "brickwall limiting" here. In my view, they are largely two implementations of the same basic concept - it's just that clipping tends to generate a lot more high-order harmonics than brickwall limiting. Is that kosher?
 
Quote
DC shift aside, runs of clipped samples are effectively muted samples. There is no audio there. So the question becomes, if you hide the DC shift, how many samples can you set to zero before the effect becomes audible. This is quite well researched in psychoacoustics - the answer is totally content dependent, but it can easily be a couple of milliseconds or more. This is for something like white noise, with a single mute - not a repeating pattern.
As mentioned above, periodic clipping is obviously a much bigger issue than aperiodic clipping - and periodic clipping clearly happens a lot in modern mastering, and it sucks - but telling the two apart is, I believe, quite important, and rather difficult to do with a visual analysis.
     
... also, you know, for a psychoacoustic result that is so well known, it's damn hard to find information on it online. I think I spent half an hour looking for documentation on this sort of thing, and the only reason I didn't search for a longer time was that I've done this search before and came up relatively empty-handed.
       
        Any additional information you can provide on the audibility of clipping/limiting would be greatly appreciated.
           
Quote
Most listeners have speakers that will "change the timbre" dramatically more than any record producer would choose to - that doesn't mean that those of us with decent speakers should also have to suffer.


No, but it does drastically reduce the perceived importance of clipping distortion. People care when information that is "rightfully" theirs - the dynamic range of the performance - is robbed from them. I don't believe they care as much when the drum set doesn't sound right. There are tons of successful albums out there that, IMHO, have ridiculously bad sounding percussion.
           
Quite frankly, I believe that dynamic range changes and distortion increases are Big Deals that can persuade people, but timbral changes are not going to persuade anybody outside the audiophile set.
           
Quote
Quote
Perhaps the most accurate method of evaluating mastering quality, though, is the simplest: Asking the mastering engineer.
Funny, I use my own ears, not someone else's!
Bah. Don't make me shout TOS8 at you. ;P
         
  Actually, SRSLY... I do have (legitimate) access to a multitrack recording session of a rock group, and a reasonably professional mixdown and master that manages to squeeze 9db off the peaks. Would you or anybody else be interested in some ABX work to see how much limiting can be applied before audibility occurs?
       
 
Quote
Quote
the sound quality of music today is still considerably better than what it was for (most of) the last 60 years.
I think the kind of music that's trashed today would probably have been trashed in a different way had it been made in the past. Many (not most) pop records from the 1960s were fairly lo-fi. However, it's quite easy to find recordings from the late 1950s which sound far better on a decent stereo than typical modern pop recordings. The chance of being able to listen to a pop recording on a decent stereo and actually enjoy it is probably lower now than it's ever been.
And yet..... it's still done. A lot.
       
That's the biggest paradox in all of this that really makes me question how important talk of the loudness war really is. If it really were as bad as everybody says it is, people would not be listening to modern music nearly as much as they actually do. But the teeny boppers are still listening to the Jonas Brothers and Hannah Montana (at least they do in Texas - I dunno what's going down in the UK), and they seem just as thrilled about them - even in the recorded format - as kids from previous generations have been about "their" music. I have yet to see any kind of conclusive evidence that declines in the music business can in any significant way be the result of compromised audio quality.
       
On a vaguely related note, the cynic in me wants to believe that this is all a product of our Prozac-addled Muzak-influenced culture, and that the reduced dynamic range of modern music is simply a reflection of a culture that demands less emotion out of its music. And that that's what we're really railing against.
  • Last Edit: 09 September, 2008, 01:22:43 PM by Axon

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #11
I have Katz' book, btw  (though perhaps not the most recent edition)...if you need me to look something up, just ask.

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #12
Quote
Quite simply, the waveform plot is a worthless tool for evaluating mastering quality and its use for that purpose should be banished.
I think it would be more accurate to say, subjectively 'bad sounding mastering' can usually be convincingly correlated after the fact with some quality of a waveform or spectral view; but be careful going the other way -- I have seen people tout waveforms with relatively minor, and likely inaudible, amounts of clipping, as proof that the track sounds bad.
Right. That is, the measurement techniques we have tend to be good at telling us when we already know something is wrong, but they are really bad at telling us when something is ok. The false positive (and false negative!) rates can be extremely high. This makes most measurement techniques considerably less useful in their predictive power than people realize. Waveform plots are merely the most egregious offenders here.

Quote
What 'sounds good' is so highly influenced by audio and non-audio factors that I really have to wonder whenever I read rave (or neg) reviews of music these days. Example: since I began ripping DVD-As and 'laserdropping' SACDs, it's been illuminating to look back and see how many of these often highly compressed (though rarely clipped) stereo versions were praised to the skies as the best ever masterings so far...by people who claim to loathe compression when they see it on CD waveforms.  To me, that's gotta be some combination of expectation bias (hi rez = good sound) and actual pleasure with the sound.


Say it, brotha. Did you know that the Yes DVD-A of Fragile was brickwalled more heavily than Gaswirt?

Quote
As for 'ask a mastering engineer', and using Hoffman as a paradigm, given his lifelong aversion to digital (he wouldn't even make a single digital edit to join the two masters of side one of 'Thick as a Brick', claiming it degrades the sound), and some of the bogosity he's spouted about it, I'd hardly consider him a go-to guy on the matter.  I'd go to Katz, who actually accepts that digitally compressed audio can still sound great.


I'm arguing here from a purely informational basis. That is, generally (and you of all people are going to understand the exceptions here), SH is known to avoid or reduce compression, and he is more or less frank about when he makes changes like that to his masters. Ergo, hearing from SH something like "I reduced the drum track compression from 4:1 to 2:1" is about as conclusive an example of a dynamic range increase as you're going to ever get.

I agree that Katz is arguably more trustworthy from an engineering perspective, but that's not really pertinent to the discussion. That doesn't really help your buying decisions any, except on name recognition alone.

Another example: Betcha didn't know that Peter Aczel is a huge fan of Mapleshade Recordings...

Quote
Have you posted your links and thoughts in a thread on Hoffman's forum?  That should be, um, interesting to watch, if so. 
Given that SH.tv seems to be source of the idea of using EAC peak values as an estimator of audio quality, I'm inclined to agree - at least, I'll hear something from people who know what a waveform is.
  • Last Edit: 09 September, 2008, 01:37:11 PM by Axon

  • Lyx
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #13
Quote
Quote
the sound quality of music today is still considerably better than what it was for (most of) the last 60 years.
I think the kind of music that's trashed today would probably have been trashed in a different way had it been made in the past. Many (not most) pop records from the 1960s were fairly lo-fi. However, it's quite easy to find recordings from the late 1950s which sound far better on a decent stereo than typical modern pop recordings. The chance of being able to listen to a pop recording on a decent stereo and actually enjoy it is probably lower now than it's ever been.
And yet..... it's still done. A lot.
       
That's the biggest paradox in all of this that really makes me question how important talk of the loudness war really is. If it really were as bad as everybody says it is, people would not be listening to modern music nearly as much as they actually do. But the teeny boppers are still listening to the Jonas Brothers and Hannah Montana (at least they do in Texas - I dunno what's going down in the UK), and they seem just as thrilled about them - even in the recorded format - as kids from previous generations have been about "their" music. I have yet to see any kind of conclusive evidence that declines in the music business can in any significant way be the result of compromised audio quality.


It is called relativity. They have no comparision. It is easy to be satisfied with eating shit, if you haven't ever tasted anything better. They have no reference! Their "rating" is literarily meaningless. They enjoy it - compared to what? What? More of the same?

Common people aren't capable of making competent and meaningful ratings - they just do, feel and think as they are told to do. They are believers. This is not a rant with the purpose of criticizing those people and requesting a change. No. Those kind of people have always been like that and will always be like that, because they want to be like that. Its comfortable and seemingly riskfree. What i am saying is that the ratings of those people dont matter. They are irrelevant, because they choosed so. What matters are the ratings of people different to those. The mainstream does not invent and set trends, it follows the trends created by others... its only effect is inertia - or in more friendly words: stability.

- Lyx
I am arrogant and I can afford it because I deliver.

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #14
hmm....manually compressing a face....that's a little extreme 


"Gainriding your face"?

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #15
Say it, brotha. Did you know that the Yes DVD-A of Fragile was brickwalled more heavily than Gaswirt?


Yes.  The Gastwirt isn't brickwalled at all, is it?  It only reaches actual 'peak level' once, in one channel.
What's scary is that the latest CD remaster -- on Rhino, credited to Inglot & Hersch -- is even MORE smashed than the DVD-A.

Anyway, I get more jollies from comparing frequency plots these days, than waveforms...the seem to 'tell' me more about what I'm hearing (prolly because as humans we're so sensitive to EQ changes in the midrange)

Check it out.  DVD-A vs Gastwirt:

  • Last Edit: 09 September, 2008, 02:11:27 PM by krabapple

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #16
First of all, check out Bob Katz's comment (and my reply). BAM! As usual, he lays down the law: his vinyl masters do not include the limiting that exists on his CD masters. I really like his idea of peak-to-loudness measurement too.

Yes.  The Gastwirt isn't brickwalled at all, is it?  It only reaches actual 'peak level' once, in one channel.
What's scary is that the latest CD remaster -- on Rhino, credited to Inglot & Hersch -- is even MORE smashed than the DVD-A.

Anyway, I get more jollies from comparing frequency plots these days, than waveforms...the seem to 'tell' me more about what I'm hearing (prolly because as humans we're so sensitive to EQ changes in the midrange)

Check it out.  DVD-A vs Gastwirt:
Ah, yes. That Mona Lisa smiley-face of an eq is not news to me. When I first ripped it, I honestly did think I could hear a lot more detail in the DVD-A version, but after I level-matched everything, the differences are honestly pretty slight. Methinks there may have been some remixing involved - explaining the large channel imbalances at high frequencies, suggesting some creative placement of percussion in the stereo mix.

They would have had to go back to the multitracks anyway for the surround mix, so I'm imagining that they did a brand new downmix for the stereo version to match the 5.1?

Quote
Hoffmans' 'catalog' is relatively small, and largely out of print, so his impact on buying decisions isn't going to be that great...except for rabid audiophiles willing to shell out $100 for a Hoffman 'Aqualung' CD.  And Hoffman's information about various releases has been known to, um, evolve over time...I'd stick with Katz for information if possible.  (It's interesting too that there are some Hoffman CD remasterings that are clipped.)
Right. I certainly trust Katz more than SH on any matter of audio engineering. But like I said.... I'm not looking to mastering engineers for general information about sound quality or beliefs or whatnot, I'm looking for knowledge as to exactly what they did with their records. And for some records (but perhaps an extremely small number), SH has fairly comprehensive info out there on what was done with each mastering project, and that information has been both fairly consistent over time and matches user listening experience.

Fortunately, now that Bob is on the record saying that he sources his vinyl masters without added loudness maximizers, I think this particular issue is moot.
  • Last Edit: 09 September, 2008, 06:16:09 PM by Axon

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #17
Ah, yes. That Mona Lisa smiley-face of an eq is not news to me. When I first ripped it, I honestly did think I could hear a lot more detail in the DVD-A version, but after I level-matched everything, the differences are honestly pretty slight. Methinks there may have been some remixing involved - explaining the large channel imbalances at high frequencies, suggesting some creative placement of percussion in the stereo mix.


No remixing.  That large imbalance -- about 6dB at max -- can be created without it. Remember, this graph doesn't compare channel to channel of the same mix directly -- it's comparing each channel to its counterpart on the Gastwirst version.

That's important to keep in mind. What I showed isn't a direct EQ plot -- it is a plot of *difference* between the DVD-A and an arbitrary reference mastering (Gastwirt's).  And note too that by that criterion, only one channel is 'smiley' compared to the same channel of the Gastwirt;  the other is 'frowny' in the treble region (a half-frown?)

Also: imagine, for example, the same sort of graph where the 'reference' CD was mastered with muddy highs and inadequate bass compared to the original tapes -- a 'flat' remastered CD from the original tapes would look very 'smiley' in a difference comparison to the reference CD.  Yet it wouldn't be 'smiley' compared to the original masters.

This pedantic episode brought to you by me, to show that without knowing what's on the master, and without accounting for what's being measured, it's possible to jump to speculative conclusions. 

Btw, here's successive waveform views of ripped 'Roundabout' remasters -- please don't punch me.

Gastwirt CD


DVD-A stereo track


Rhino CD







Quote
They would have had to go back to the multitracks anyway for the surround mix, so I'm imagining that they did a brand new downmix for the stereo version to match the 5.1?


No, that's rarely done.  The stereo version here, as is usual, is just the original two-track mix, remastered. I wouldn't bother comparing two different MIXES this way, though I suppose it too could be informative.
  • Last Edit: 09 September, 2008, 05:52:24 PM by krabapple

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #18
No remixing.  That large imbalance -- about 6dB at max -- can be created without it. Remember, this graph doesn't compare channel to channel of the same mix directly -- it's comparing each channel to its counterpart on the Gastwirst version.

That's important to keep in mind. What I showed isn't a direct EQ plot -- it is a plot of *difference* between the DVD-A and an arbitrary reference mastering (Gastwirt's).  And note too that by that criterion, only one channel is 'smiley' compared to the same channel of the Gastwirt;  the other is 'frowny' in the treble region (a half-frown?)
You mean, perhaps, a...

Quote
Mona Lisa smiley-face



I've got some handrolled code which sums the spectral content of two different tracks and divides the two together. So again, these graphs are not news to me.

Quote
Also: imagine, for example, the same sort of graph where the 'reference' CD was mastered with muddy highs and inadequate bass compared to the original tapes -- a 'flat' remastered CD from the original tapes would look very 'smiley' in a difference comparison to the reference CD.  Yet it wouldn't be 'smiley' compared to the original masters.

This pedantic episode brought to you by me, to show that without knowing what's on the master, and without accounting for what's being measured, it's possible to jump to speculative conclusions. 
Right - a bunch of different things could cause this. I just figured that the 5.1 mix added an extra layer of plausibility. In fact, at one point I listened to the DVD-A in my hires player over headphones and honestly wondered if there even was a stereo mix on the disc - I might have wired the 2 channel out wrong and got 2 channels of the 5.1 mix by accident. That may have confused my perception of exactly what is on the 2-channel mix.

Regardless, I'm not sure we have concrete evidence one way or another as to what's going on here.

Quote
Btw, here's successive waveform views of ripped 'Roundabout' remasters -- please don't punch me.
I promise I won't gainride you... as long as you also post clips from "Five Per Cent for Nothing", which, as I recall, was one of the clearer instances of limiting on the DVD-A.

Quote
No, that's rarely done.  The stereo version here, as is usual, is just the original two-track mix, remastered. I wouldn't bother comparing two different MIXES this way, though I suppose it too could be informative.
How do you know this?

And yes, this is all horrifically off topic

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #19
Mona Lisa smiley-face

d'oh!


Quote
Quote
No, that's rarely done.  The stereo version here, as is usual, is just the original two-track mix, remastered. I wouldn't bother comparing two different MIXES this way, though I suppose it too could be informative.
How do you know this?

And yes, this is all horrifically off topic


For Fragile , I know it because I know that album like the back of my hand (I'm old enough to remember when "Roundabout" was a hit..on AM radio).  True for many other hi-rez discs I own too. It's hard to do a remix that is EXACTLY faithful to the original, in terms of getting all the parts in there.  The 5.1 channel remix of "South Side of the Sky', for example, has a glaring omission of a bass swell that's intact in the two-channel version.

Elsewhere, because many of these releases date from when ICE magazine was still around, we got 'from the horse's mouth' reports from the remixers, and a remix two-track would have been touted as a news or selling point, as it is in every medium I've ever seen, and I kept track of these things.  In fact offhand, I can't think of any DVD-As or SACDs that feature remixed (not downmixed) dedicated 2-track versions, aside from the controversial Genesis set.  And if the 'remix' is really the downmix of the 5.1 generated on the fly,  that's evident too, on the few discs where that's the case , e.g. ELP Brain Salad Surgery,  Natalie Merchant 'Waterlily', by the absence of a selectable 2-channel DVD-A mix in the directories (nothing 2-channel in the AOB directory either).  You only get two-channel as a downmix. Believe me, this pissed me off when I bought that ELP disc way back when, because the packaging was quite misleading in that regard.  It was later revealed that there was no dedicated 2-channel transfer on that one because the original 2-channel masters could not be found in time. Warner pulled this trick on a few of its early DVD-A, but quickly began including real remastered 2-channel tracks.
  • Last Edit: 09 September, 2008, 06:39:14 PM by krabapple

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #20
Ah, yes. That firsthand knowledge actually dovetails pretty well with my thinking that knowing mastering decisions from the horse's mouth (the mastering engineer) is the most trustworthy method of comparing masterings (besides listening to it yourself). Of course, said engineer can lie, but you've got bigger problems if that's happening.

  • krabapple
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #21
Ah, yes. That firsthand knowledge actually dovetails pretty well with my thinking that knowing mastering decisions from the horse's mouth (the mastering engineer) is the most trustworthy method of comparing masterings (besides listening to it yourself). Of course, said engineer can lie, but you've got bigger problems if that's happening.



Remixing the two-channel would make the production even more costly.  Really, it was never a big part of the DVD-A /SACD push to have remixed two-tracks; the releases were usually touted as including the bestest-ever remaster of your favorite old album along with the shiny new surround mix (though sometimes Sony didn't even give you THAT).  As best I know, every remixed two-track remaster that's on them is also on a CD release as well (Miles Davis, Brubeck, Genesis), I can't think of any that were made JUST for hi-rez.

Here's a list someone posted on HA of some DVD-As that only present downmixes when you select 2-channel. Note that a lot of them come from notorious cheapo disc producer Silverline:

http://www.hydrogenaudio.org/forums/index....st&p=554804
  • Last Edit: 09 September, 2008, 06:41:58 PM by krabapple

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Waveform Plots Considered Harmful
Reply #22
It is called relativity. They have no comparision. It is easy to be satisfied with eating shit, if you haven't ever tasted anything better. They have no reference! Their "rating" is literarily meaningless. They enjoy it - compared to what? What? More of the same?

There are plenty of people my age (mid 20s) and older who pretty much don't notice a thing when they listen to modern masterings. Lots of people listen to both their old CDs and their new CDs, and besides the large volume difference, and a reduced reliance on gated reverb, they do not notice much production difference between the two.

Put another way: if what you are saying is true, Norah Jones would not have a career.

Quote
Common people aren't capable of making competent and meaningful ratings - they just do, feel and think as they are told to do. They are believers. This is not a rant with the purpose of criticizing those people and requesting a change. No. Those kind of people have always been like that and will always be like that, because they want to be like that. Its comfortable and seemingly riskfree. What i am saying is that the ratings of those people dont matter. They are irrelevant, because they choosed so. What matters are the ratings of people different to those. The mainstream does not invent and set trends, it follows the trends created by others... its only effect is inertia - or in more friendly words: stability.


Shut up, old man.

SRSLY: "the music kids are listening to nowdays is too loud and too harsh" has been a complaint since, literally, the 1930s - ever since the advent of amplified microphones and dynamic range compression.

If this can only be justified on relativity and elitism, it's never going to punch the mainstream. What, are we going to pursue the Scientologist strategem, and get all sorts of celebrities to complain about how bad CDs sound nowadays?



Hey, um, we couldn't get a mod to split the DVD-A thread off, could we?

  • Lyx
  • [*][*][*][*][*]
Waveform Plots Considered Harmful
Reply #23
If this can only be justified on relativity and elitism, it's never going to punch the mainstream. What, are we going to pursue the Scientologist strategem, and get all sorts of celebrities to complain about how bad CDs sound nowadays?

I'm simply saying that the "opinion" of the mainstream is just a copy of someone elses opinion. If you get enough hype going on, so that media-outlets tell the mainstream about the problem, it will suddenly as by magic notice it. But: The mainstream on its own, wont complain about that on its own - it does not create, it imitates. It has no own opinion and doesn't honestly want one.

Are you really believing in the "common (non-)sense", "(un-)sensivity" and "(non-)self-confidence" of average joe? Does your average joe seem like someone who notices something unpopular on his own, and then starts to make unusual demands?

Thats how it is currently, regardless of if you like that or not. If you want your "strategy" to work, it needs to include that aspect too. Else you're building on wishful-thinking.

- Lyx

P.S.:Here's a different example. We are compressing the night sky. Yes, we are. By outputting vast amounts of light into the sky, we in most urbanized countries have made almost the entire sky disappear. Only a low amount of handful of stars are still visible. Dozens of sky objects, which in the past were something normal and well known, are now unknown. This is something, which you CAN directly show another person. It's an issue which - similiar to in this community - is decried by lots of enthusiasts. Does your average joe care? Does he even consciously look at above anymore? Do you think the kids born in the last 15 years even have a sufficient comparision anymore? If you tell them about various sky objects, they react as if you're talking about aliens. However, asume the media would create a new nightsky-hype, thus suddenly making average joe be interested in it again. So all those buy googles and stuff - and one of the first enemies they will get to know - the biggest problem nowadays for skygazers - will be light pollution. And suddenly, there would be popular interest in fixing that (there are simple ways to fix it, and it even saves cash). But do you think anything like that will happen out of average joe by himself? Why would someone, who in the past never cared about the topic, suddenly be interested in it, by pure coincidence at the same time as the media report about it?
  • Last Edit: 09 September, 2008, 07:48:43 PM by Lyx
I am arrogant and I can afford it because I deliver.

  • Chromatix
  • [*][*]
Waveform Plots Considered Harmful
Reply #24
*de-lurk*

We are compressing the night sky.

Do you think the kids born in the last 15 years even have a sufficient comparision anymore? If you tell them about various sky objects, they react as if you're talking about aliens.

You might as well be talking about city kids in 1930s Britain.  When they were evacuated in 1939, many of them had never seen the countryside before.  Didn't know what a cow was, never mind that milk came from it.

A lot of them, being from poor backgrounds, found that the beds hurriedly made up for them were vastly more comfortable than what they'd had at home.

I'm fairly sure that the country skies were clearer than the city ones then too - but for a different reason.  Smog.