Lies about vinyl vs digital

[sandyk]

20 minutes ago, PeterSt said:

 

On some aspects maybe. But what that mostly did was taking out dynamic range.

 IMO, her voice sounds much better. I couldn't stand that album in it's original form and would never purchase it.

 The 2000 version in the graph is very obviously compressed, and things have got far worse in many cases since then !!!

[Paul R]

3 hours ago, sandyk said:

Frank

 I have seen many recent recordings that are far worse than this since then.

 

 What many people don't realise is that this excessive compression destroys most of the background ambient information, and with female voices for example, makes them sound far harder sounding, losing any natural sweetness that most males appreciate in a female's voice. It tends to make many female voices sound much the same, losing their individuality.

 

Anything they can do to Britney (sp?)  Spears is gotta be an improvement. Retirement maybe?  

 

[fas42]

15 minutes ago, Paul R said:

 

Anything they can do to Britney (sp?)  Spears is gotta be an improvement. Retirement maybe?  

 

 

The type of limiting that is often done on the final mix, which never actually clips, is not too hard to reverse - done adroitly, it becomes a very reasonable recording to listen to.

 

There was a famous example of extreme compression that was applied to a heavy metal band track that I had a go at once - Metallica?? - and it turned out quite well; there will be an industry doing this one day ...

[PeterSt]

FWIW ...

 

http://www.azwebpages.com/bass/BassCompressionAudioSamples.htm

Compare from Blues Riff in A the Unprocessed Signal with the Heavy Compression. This beautifully shows the dynamic range "utilization" because of compression. Thus, the Unprocessed just has a normal decay; The more the compression, the more the (extra) gaining kicks in and you can just hear the decay go *louder* again (even by means of stepping), in the end not decaying at all.

Of course there is more to listen for, but the eye opener is that already a simple background bass sound (could be backing vocals) lose their being dynamical (soft vs louder humming) because of how compression works out.

 

 

 

 

[sandyk]

3 hours ago, Paul R said:

 

Anything they can do to Britney (sp?)  Spears is gotta be an improvement. Retirement maybe?  

 

 You wouldn't want to hear this one then with IIRC ,the F word too.

Britney Spears - In the Zone 5.1.from DVD-A

 

 

 

[alfe]

11 hours ago, sandyk said:

 Al

 That statement is incorrect. 

 

OK 😔

[davide256]

 

7 hours ago, PeterSt said:

 

Alex, that gif is not an example of compressed music. It's made (or put up) by someone who doesn't understand what compression is or looks like.

The gif shows how at first the digital headroom is not used at all. Then a bit better. And then a bit better again. So the first pic is the worst sounding one (least dynamics). I don't say that the last one will be the best sounding.

 

 

Sure.

 

 

But this is compression and it shows the opposite of that gif, right ? All this needs for the loudness war is to level gain again the red version and there you have it. 

 

That's not true compression... its clipping. If compression is done correctly  the wave form should resemble the original but with a lower peak level and reduced differences

between soft and loud passages. Fire the engineer for incompetent use of technology.

 

Compression is nothing new... most vinyl is compressed because otherwise too much distance between grooves would be required to avoid perturbing a neighboring groove

during master cutting

[The_K-Man]

7 hours ago, PeterSt said:

the first pic is the worst sounding one (least dynamics). I

 

Not really.  Just exercise that lazy wrist and TURN THAT VOLUME KNOB clockwise a little more!  You'll hear the dynamics.

 

I don't know who in Rome decreed that the finished song be dynamically compressed or limited and cranked up to resemble the final two images in that waveform GIF.

 

It's like "Oh no, it's dynamic, the waveform's not filling the whole workspace - OMG, the sky is falling!"

 

Puh-leeez...

 

Learn what REAL MUSIC sounds like!  

 

(vs the dial-tones and leaf blowers that currently occupy the Bill Boards and other music charts)

[The_K-Man]

8 hours ago, fas42 said:

 

Yes. Which is why I never go to 'live' shows; the PA is set up to do as much damage as possible - "You mean, you want me to pay to come in and listen to this awful sound!"

 

 

It will take years ... but uncompressing, and unmixing the master will solve many of the issues with the stuff they're doing now - this sort of processing is being actively investigated; and will keep getting better ...

 

By sorting out I meant otherwise: Reconditioning the public to what a good pop song can sound like again.

 

Going forward, I think loudness(vs peak-)-based metering in recording and post-studios will go a long way towards sane levels in mixes and masters, and we'll be back to something similar to the old VU meter days, where zero is two-thirds up the scale and measures average loudness instead of peaks.

[The_K-Man]

27 minutes ago, davide256 said:

Fire the engineer for incompetent use of technology.

 No.

 

Educate the CLIENT(the recording artists, producers, and record labels) on the penalties of the loudness mentality.

 

Among mastering engineers ther are exceptions(Lord-Alge and Bob Ludwig tend to be hot masterers), but most respectable ones will honor the client's creative wishes: louder, softer, more dynamic, less dynamic, more EQ'd more natural, etc.

[STC]

1 hour ago, The_K-Man said:

 No.

 

Educate the CLIENT(the recording artists, producers, and record labels) on the penalties of the loudness mentality.

 

Among mastering engineers ther are exceptions(Lord-Alge and Bob Ludwig tend to be hot masterers), but most respectable ones will honor the client's creative wishes: louder, softer, more dynamic, less dynamic, more EQ'd more natural, etc.

 

The evidence is “Surprisingly, the results failed to reveal any evidence of the effects of dynamic range compression on subjective preference or perceived depth cues. Perceptual data suggest that listeners are less sensitive than commonly believed to even high levels of compression. As measured in terms of differences in the peak-to-average ratio, compression has little perceptual effect other than increased loudness or clipping effects that only occur at high levels of compression. One explanation for the inconsistency between data and belief might result from the fact that compression is frequently accompanied by additional processing such as equalization and stereo enhancement.”

[The_K-Man]

1 minute ago, STC said:

 

The evidence is “Surprisingly, the results failed to reveal any evidence of the effects of dynamic range compression on subjective preference or perceived depth cues. Perceptual data suggest that listeners are less sensitive than commonly believed to even high levels of compression. As measured in terms of differences in the peak-to-average ratio, compression has little perceptual effect other than increased loudness or clipping effects that only occur at high levels of compression. One explanation for the inconsistency between data and belief might result from the fact that compression is frequently accompanied by additional processing such as equalization and stereo enhancement.”

 

And the point of sharing that is ... ?

[STC]

1 minute ago, The_K-Man said:

 

And the point of sharing that is ... ?

 

And here is another paper. 

 

“The experimental results further demonstrate that caution should be exercised when extrapolating the beliefs of expert listeners to the general public on matters of slight sound quality differences. Without training, these differences may be unperceivable. “

[The_K-Man]

2 minutes ago, STC said:

 

And here is another paper. 

 

“The experimental results further demonstrate that caution should be exercised when extrapolating the beliefs of expert listeners to the general public on matters of slight sound quality differences. Without training, these differences may be unperceivable. “

 

Well, if the general public can't here the differences in this clip:

 

 

Then they are DEAF.

[STC]

2 minutes ago, The_K-Man said:

 

Well, if the general public can't here the differences in this clip:

 

 

Then they are DEAF.

 

Level match the volume and do a poll. In the paper it was also mentioned that how much compression before it becomes apparent. 

 

I am am just the messenger who is representing people who listen music for music. 

[Paul R]

2 hours ago, The_K-Man said:

 

Not really.  Just exercise that lazy wrist and TURN THAT VOLUME KNOB clockwise a little more!  You'll hear the dynamics.

 

I don't know who in Rome decreed that the finished song be dynamically compressed or limited and cranked up to resemble the final two images in that waveform GIF.

 

It's like "Oh no, it's dynamic, the waveform's not filling the whole workspace - OMG, the sky is falling!"

 

Puh-leeez...

 

Learn what REAL MUSIC sounds like!  

 

(vs the dial-tones and leaf blowers that currently occupy the Bill Boards and other music charts)

 

I think telling Peter to “learn what real music sounds like” may be the equivalent of teaching a fish how to.swim. 😁

[The_K-Man]

1 minute ago, Paul R said:

 

I think telling Peter to “learn what real music sounds like” may be the equivalent of teaching a fish how to.swim. 😁

 

It was aimed more at anyone reading it than specifically at Peter.  

[The_K-Man]

4 minutes ago, STC said:

 

Level match the volume and do a poll. In the paper it was also mentioned that how much compression before it becomes apparent. 

 

I am am just the messenger who is representing people who listen music for music. 

 

Ever hear the parable about putting a frog in a pot of water then slowly increasing the heat to a boil?

[STC]

1 minute ago, The_K-Man said:

 

Ever hear the parable about putting a frog in a pot of water then slowly increasing the heat to a boil?

 

But I also heard about much ado about nothing. Dynamic compression when not excessive is not actually a bad thing.

 

IIRC, there was another paper on preference about mildly compressed and uncompressed. 

 

 

[The_K-Man]

12 minutes ago, STC said:

 

But I also heard about much ado about nothing. Dynamic compression when not excessive is not actually a bad thing.

 

IIRC, there was another paper on preference about mildly compressed and uncompressed. 

 

 

 

I see where you're coming from.

 

I'm through with you.

[PeterSt]

1 hour ago, The_K-Man said:

I see where you're coming from.

 

I'm through with you.

 

I suppose it is not easy for many people to hear through "manipulation" and how this leads to deterioration. This is also a lot about allowing yourself to think that a. may sound better with some extra bass while b. sounds better with some less bass (to stick with one dimension only).

 

So in this case for @STC, yes, of course you were the messenger only. But you also always seek for better sound by means of similar manipulation (and I don't care much whether this is with delays - it will be detrimental in my book). So the whole process of DSP is a forbidden thing (again, in my personal book) and things should not be solved that way. It must be inherently good and if it is not that, ditch it.

And "good" does not spring from subjective measure, and thus we need something which can be used objectively. Anything which smells like "remaster" is already a measure of "worse". Sadly it is just so. The exception may come from exactly what this gif shows - too few digital headroom used. And then still it is mighty difficult. It may well be so that we are not able to survive "no compression at all because the sound may be too dynamical". I mean, we always and still have Supertramp's Crime of The Century example. The best "DR" version of it doesn't even use 2/3 of the digital headroom (in digital numbers, like 22000 of 32876) and it is still unplayable because of too soft vs too loud passages. Do what I would like by nature, and you'd have 3dB more of DR and my internal windows will go out. So I suppose such an album firstly requires something like 30 dB of more virtual gain, which is dynamically compressed until it fits in the digital space (I know the sequence can't go like this, but this is how we can envision it, hopefully).

In the end I don't even know how one gets an album (master for digital) so wrong, because IIRC the LP is quite fine (also see the second one below). Notice that the worse is the original (the one but last) as how it ever back came out, which is equal to some Japan version (the 3rd).

 

[The_K-Man]

14 minutes ago, PeterSt said:

 

I suppose it is not easy for many people to hear through "manipulation" and how this leads to deterioration. This is also a lot about allowing yourself to think that a. may sound better with some extra bass while b. sounds better with some less bass (to stick with one dimension only).

 

So in this case for @STC, yes, of course you were the messenger only. But you also always seek for better sound by means of similar manipulation (and I don't care much whether this is with delays - it will be detrimental in my book). So the whole process of DSP is a forbidden thing (again, in my personal book) and things should not be solved that way. It must be inherently good and if it is not that, ditch it.

And "good" does not spring from subjective measure, and thus we need something which can be used objectively. Anything which smells like "remaster" is already a measure of "worse". Sadly it is just so. The exception may come from exactly what this gif shows - too few digital headroom used. And then still it is mighty difficult. It may well be so that we are not able to survive "no compression at all because the sound may be too dynamical". I mean, we always and still have Supertramp's Crime of The Century example. The best "DR" version of it doesn't even use 2/3 of the digital headroom (in digital numbers, like 22000 of 32876) and it is still unplayable because of too soft vs too loud passages. Do what I would like by nature, and you'd have 3dB more of DR and my internal windows will go out. So I suppose such an album firstly requires something like 30 dB of more virtual gain, which is dynamically compressed until it fits in the digital space (I know the sequence can't go like this, but this is how we can envision it, hopefully).

In the end I don't even know how one gets an album (master for digital) so wrong, because IIRC the LP is quite fine (also see the second one below). Notice that the worse is the original (the one but last) as how it ever back came out, which is equal to some Japan version (the 3rd).

 

 

Again, there seems to be an obsession, which you allude to, with always "using all the bits"/dynamic range available. That's not how music in real life operates, even if it can be manipulated that way in a DAW.   Real music does not have a consistent fat average waveform, with peaks all picket-fenced against full-scale in the DAW window.

 

I understand that's the production norm nowadays, but that doesn't make it right.

[PeterSt]

6 minutes ago, The_K-Man said:

That's not how music in real life operates, even if it can be manipulated that way in a DAW.

 

Err, that is the whole point. It can NOT be manipulated that way in a DAW and therefore once we see it has happened, it has happened in illegal fashion. But if one single sample per track would touch the limit of the headroom, I would be fine with it. ... and that too can not be achieved. So there is one solution only: allow for sufficient headroom and indeed don't be obsessed about billions of samples not even getting close to the limit, which already starts with the recording itself.

 

We can also turn things upside down a little ... What do people actually think what happens when we use a digital volume control (hence no pre-amp or other means of analogue volume control) ?

I use it all the time, usually with 16-24dB of digital attenuation (in the 24 bit domain). This makes my Crime of The Century example quite bleak because the 22000 used there (mapped on to 16 bits) would turn out to be 8192 for maximum, hence again more than 6dB worse. Does this bother me ? NEVER. Of course with the Supertramp it starts out with 3dB lost and which never can be gotten back. But if 24dB isn't even the lightest of importance, that extra 3dB will not be devastating either.

 

So we agree on this.

But it is a different subject from me (or STC) suggesting that some albums may require x dB more than the virtual limit, to next compress that down. Try such Supertramp version and see that it isn't playable. Really not.

 

What I refuse to believe is that every album requires this and that THUS it wasn't applied in the old (1983) days. Hence my question what happened to the Supertramp album that it ended up like that. Maybe SandyK's tool was applied a couple of times to it. And btw contrary to what we might think, there is no power as such in the album as well. Kick drums sound like cardboard.

 

[Paul R]

28 minutes ago, PeterSt said:

 

Err, that is the whole point. It can NOT be manipulated that way in a DAW and therefore once we see it has happened, it has happened in illegal fashion. But if one single sample per track would touch the limit of the headroom, I would be fine with it. ... and that too can not be achieved. So there is one solution only: allow for sufficient headroom and indeed don't be obsessed about billions of samples not even getting close to the limit, which already starts with the recording itself.

 

We can also turn things upside down a little ... What do people actually think what happens when we use a digital volume control (hence no pre-amp or other means of analogue volume control) ?

I use it all the time, usually with 16-24dB of digital attenuation (in the 24 bit domain). This makes my Crime of The Century example quite bleak because the 22000 used there (mapped on to 16 bits) would turn out to be 8192 for maximum, hence again more than 6dB worse. Does this bother me ? NEVER. Of course with the Supertramp it starts out with 3dB lost and which never can be gotten back. But if 24dB isn't even the lightest of importance, that extra 3dB will not be devastating either.

 

So we agree on this.

But it is a different subject from me (or STC) suggesting that some albums may require x dB more than the virtual limit, to next compress that down. Try such Supertramp version and see that it isn't playable. Really not.

 

What I refuse to believe is that every album requires this and that THUS it wasn't applied in the old (1983) days. Hence my question what happened to the Supertramp album that it ended up like that. Maybe SandyK's tool was applied a couple of times to it. And btw contrary to what we might think, there is no power as such in the album as well. Kick drums sound like cardboard.

 

Isn’t that the old 80’s gate limited drum thing? Wasn’t it accidentally discovered right about the? Something about leaving a voice mike on over the drum set that was gate limited. 

[esldude]

Actually this uses all the bits. 

 

Same full scale waveform viewed closer up.  See how this full scale sine goes all the way to zero and crosses it on a regular basis. 

 

People think that running nearly full scale means the lower bits don't get used.  That isn't really true.  Now it does mean if the average level is high enough, you could have used less bits and noise wouldn't intrude.  Compression does work that way.  But that is the point.  Even if the recording is pristine, quiet passages get lost while listening say in a noisy car.  Compression lifts some of the quieter passages up so they don't get lost in the car noise or not as lost.  Taken too far it does other things not nice.