Digital vs Vinyl

[levimax]

No music (with a few exceptions) , old or new, has ever been recorded without compression. It is not really practical and it doesn't sound good either so from that perspective some compression is "OK" and "nothing new". That is not the same thing as taking a recording mastered according the original artists choices and removing 90% of the original dynamics which alters the FR and calling that "OK" or "normal" or "better" ..... different yes and differences are always subject to preferences, but it is not the same.

 

[Newman]

@levimax stop saying the DR meter is a useful part of the story. It's highly deceptive and misnamed.

It is the instrument behind the deception of a huge portion of the audiophile community. Supporting it just makes you part of the problem.

Surely you don't want to be the promoter of a fake loudness war? There is a real loudness war and a fake loudness war. The real one is the one that started at a serious level with the Hit Singles from the fifties, like the Motown stuff that was mentioned, Hot Hot Hot and often with clouds of distortion too. If you reckon that was not objectionable, and all the objectionable stuff is on CD, you are pushing the wrong wheelbarrow. Ask Bob Ludwig what he would do if he had to make a hot record today and he had the choice of Motown-era tools or modern tools of audio compression, which would he choose on the basis of sound quality alone. I bet you anything he would go modern.

What happened, that led to the real loudness war, was that some of the kids, who liked all that compressed garbage as kids, grew up and got real hifi and suddenly got standards. But they are conflicted: they are the first generation to like rock'n'roll etc which music inherently starts from a wall of sound -- but they want dynamics too now, because they are audiophiles, y'know. It's easier for all the other genres: tell me about the loudness war in jazz, in classical... (crickets).

The fake loudness war, though, is the one where the DR meter lives. The one where @broad can compress a song from DR14 to DR6 and not be able to tell them apart by ear. Great tool! The one where Ian Shepherd can blend the bass and the DR jumps by 6 despite, he guarantees, zero change in dynamics or dynamic range. Great tool! The one that routinely adds 4 to 8 points to LP regardless of whether the CD is of the same master or a more compressed master, and, therefore, the fake loudness war that gives a vast misleading impression of LP being broadly and generically about 6 points "more dynamic". The one that scores DR8 for the studio master of Random Access Memories, but DR13 for the LP that originated from that same master (guaranteed by the production team). Great tool!

But the most insidious of the many insidious aspects of the DR meter, is the way it has created confirmation bias in a large audiophile community. So, now, many audiophiles say, "what it says, I hear". Well yeah there's a reason for that, guys, and it's not the one you are assuming. We have cases of audiophiles shouting down the line to the engineers that they can definitely hear the DR meter’s extra 5 points of DR on version B over version A, while the engineer is guaranteeing that there is no difference in dynamics, and the audiophiles saying "you're wrong, I can clearly hear it". Man, that is a topic that has gotten seriously out of control. Please don't feed it.

 

[Newman]

a recording mastered according the original artists choices

I might as well tackle this one while I’m at it.

The whole idea that we’ve got recordings that are mastered “according to the original artists choices” is a myth.

The feedback that I get from studio engineers is that the only thing that the musicians want of the engineer is to produce a sound that will sell the most volume. This is nearly universal. Look at that Bob Ludwig video clip a few posts above, and notice how the musicians were exhorting the engineer to produce the sound identical to the Motown sound that was selling the most volume in the day. This is, apparently, completely normal. The engineer is left on his or her own to determine what sound to produce, with the sole mandate that it sells most volume.

So, really, the only thing we are getting with remastered versions is engineers attempting to sell the most volume to a new buyer segment, such as “modern audiophile LP buyer”, with the sound that buyer segment wants. And that is the way it has always been, anyway.

But, “original artist’s intent”? For recordings, it’s a non-thing. A myth.

 

[levimax]

@levimax stop saying the DR meter is a useful part of the story. It's highly deceptive and misnamed.

It is the instrument behind the deception of a huge portion of the audiophile community. Supporting it just makes you part of the problem.

Surely you don't want to be the promoter of a fake loudness war? There is a real loudness war and a fake loudness war. The real one is the one that started at a serious level with the Hit Singles from the fifties, like the Motown stuff that was mentioned, Hot Hot Hot and often with clouds of distortion too. If you reckon that was not objectionable, and all the objectionable stuff is on CD, you are pushing the wrong wheelbarrow. Ask Bob Ludwig what he would do if he had to make a hot record today and he had the choice of Motown-era tools or modern tools of audio compression, which would he choose on the basis of sound quality alone. I bet you anything he would go modern.

What happened, that led to the real loudness war, was that some of the kids, who liked all that compressed garbage as kids, grew up and got real hifi and suddenly got standards. But they are conflicted: they are the first generation to like rock'n'roll etc which music inherently starts from a wall of sound -- but they want dynamics too now, because they are audiophiles, y'know. It's easier for all the other genres: tell me about the loudness war in jazz, in classical... (crickets).

The fake loudness war, though, is the one where the DR meter lives. The one where @broad can compress a song from DR14 to DR6 and not be able to tell them apart by ear. Great tool! The one where Ian Shepherd can blend the bass and the DR jumps by 6 despite, he guarantees, zero change in dynamics or dynamic range. Great tool! The one that routinely adds 4 to 8 points to LP regardless of whether the CD is of the same master or a more compressed master, and, therefore, the fake loudness war that gives a vast misleading impression of LP being broadly and generically about 6 points "more dynamic". The one that scores DR8 for the studio master of Random Access Memories, but DR13 for the LP that originated from that same master (guaranteed by the production team). Great tool!

But the most insidious of the many insidious aspects of the DR meter, is the way it has created confirmation bias in a large audiophile community. So, now, many audiophiles say, "what it says, I hear". Well yeah there's a reason for that, guys, and it's not the one you are assuming. We have cases of audiophiles shouting down the line to the engineers that they can definitely hear the DR meter’s extra 5 points of DR on version B over version A, while the engineer is guaranteeing that there is no difference in dynamics, and the audiophiles saying "you're wrong, I can clearly hear it". Man, that is a topic that has gotten seriously out of control. Please don't feed it.

Before I wrote this I went back and listened (Via Foobar ABX) to some tracks I have that in both "original CD" and "Remastered Digital (usually a hi-res remastered version). I also ran them through DR meter as well as Adobe Audition and rather than DR meter I used LUFS which is "more professional" (LUFS more or less came up with the same answer as the DR Meter). I also ran them through a "LUFS match" and then a FFT analysis. In the case of the tracks I listened to Linda Rhonstadt, Nivana, Led Zeppelin, Lenny Kravitz I could ABX the tracks quite easily (and I do not consider myself golden eared) and in all cases the originals as measured by the DR meter had a DR of ~12 and the remasters a DR ~8. The FFT analysis also showed a similar difference with the remasters showing an increase in the 2 Khz to 10 Khz range compared to the "original" CD's. In all cases the re-masters were noticeably "bright" compared to the originals and for me I preferred the originals.

I have also done similar analysis with LP needles drops.

I am standing by my opinion that on balance the DR Meter is a flawed but useful tool. I find it very analogous to the SINAD that many on this site are so fond of. Of course for some use cases it is "flawed" but the examples sited on the videos and linked posts are extreme examples that in most cases are starting with a modern highly compressed song and doing filtering to show that the DR Meter is way off. In general I find DR for LP's and early CD's to be +/- 1 dB so these claims of 4 to 8 points are way off... yes examples can be "engineered" to show crazy differences but for most use cases like comparing an original CD to a remastered CD the DR meter does a decent job. Another parallel to SINAD is that high DR numbers don't necessarily mean good sound but low numbers always means loud, bright, and compressed.

I do not like the current highly compressed recording style but if modern artists and modern economic pressures demand it then I am fine with it but I do not like heavy compression added to older recordings since that is a lot of what I listen to and to me the compression ruins the music. If it was just louder that would be fine but compression distorts the FR and that is what I don't like... especially on vocals. While flawed anything that helps consumers make objective choices and hopefully moves recording technics back toward "hi-fi" is a positive for me.

 

[b4nt]

So, really, the only thing we are getting with remastered versions is engineers attempting to sell the most volume to a new buyer segment, such as “modern audiophile LP buyer”, with the sound that buyer segment wants. And that is the way it has always been, anyway.

But, “original artist’s intent”? For recordings, it’s a non-thing. A myth.

Add porn, publish a YT video clip. Whatever the poor audio track, success will be granted.

A remastering, or a recent but WTF mix, YT porn, plus promotions, success and sales will be granted.

 

[board]

Before I wrote this I went back and listened (Via Foobar ABX) to some tracks I have that in both "original CD" and "Remastered Digital (usually a hi-res remastered version). [...] In the case of the tracks I listened to Linda Rhonstadt, Nivana, Led Zeppelin, Lenny Kravitz I could ABX the tracks quite easily (and I do not consider myself golden eared) and in all cases the originals as measured by the DR meter had a DR of ~12 and the remasters a DR ~8. The FFT analysis also showed a similar difference with the remasters showing an increase in the 2 Khz to 10 Khz range compared to the "original" CD's. In all cases the re-masters were noticeably "bright" compared to the originals and for me I preferred the originals.

Another parallel to SINAD is that high DR numbers don't necessarily mean good sound but low numbers always means loud, bright, and compressed.

If I understand you correctly, what you're saying here, perhaps without realizing it, is that the remaster has also been EQ'ed to boost the 2 to 10 kHz range. Then, obviously, it should be relatively easy to ABX them, as EQ/frequency response easily produces the most audible changes.
And then if I also understand you correctly, you claim that this boost purely is the result of limiting/dynamic range compression, meaning if someone only uses a limiter (or at least the right limiter) the process will boost that frequency region.
I don't mean to sound antagonistic (rather I'm just trying to have a rational discussion about this), but you haven't provided any proof that this is actually the case. It seems very clear to me that in the cases you point to both limiting and EQ has been applied.
I provided some proof (the charts) that using only limiting doesn't change the frequency response to any noticeable degree, and then I've also done blind tests between the same songs uncompressed and then compressed. You would need to do the same.
So your claim that "low numbers always mean bright" simply can't be considered true, unless you provide evidence for this. It is of course true that low numbers always mean loud - obviously. But I would nevertheless say that if you dynamically compress one of those original CDs you mention I don't think it will sound as "loud" and "in your face" as you might think.
This loudness and "in your face" that is inherent in a lot of modern music I would say is rather a consequence of how the music is recorded, produced, mixed, and EQ'ed, rather than how much limiting is applied.
Also, since I've remastered/EQ'ed quite a lot of music myself I've also seen that if I boost or cut the frequencies in slightly different areas (at the "wrong" frequencies) and at slightly different levels I can make a song sound as if it "presses" on my skull, which I think is what many people think of when they think something sounds "compressed".

I asked the following question of someone else earlier in this debate, so now I'll have to ask you:
Have you tried taking uncompressed songs, e.g. the ones you have already compared to the remasters, and then dynamically compressed them, then lowered the peak volume to match the uncompressed original, and then done an ABX?
That way you can rule out if the frequency response is actually being altered by applying limiting.
If you haven't, please do the test .
I would be happy to send you the songs I used for my testing if you like, as I'm not trying to be antagonistic, but rather I'm trying to clear all this up.

I will, however, add a caveat to my long rant:
My producer/mixer friend told me that to make music excessively loud, some artists add a lot of analogue compression during the recording session, both on individual instruments and the finished mix. It then goes to the mastering engineer, who then applies more analogue compression, then EQ's it, and then finally adds limiting.
I think this might be part of the explanation to why some music sounds excessively loud, although the EQ'ing and mixing in the initial process (before the mastering engineer receives it) is in my opinion probably the main explanation, rather than the amount of compression used. A lot of music is produced and mixed to have excessive energy in the harshness region, and when I EQ/remaster music I almost always lower that region.
Although I don't know anything of the recording process, I would imagine that the song "Natural" by Imagine Dragons is an example of both mixing and EQ'ing to make it sound loud as well as adding a lot of analogue compression throughout the process before finally limiting it to DR6.

EDIT:
By the way: If the Nirvana album you listened to was "Nevermind", did you have a listen to the vinyl version on Back on Black? If not, that's my favourite version, and I think I've listened to them all. I like that version more than even the MFSL edition or the Bernie Grundman cut for ORG, which many hail as the best versions. I find the original recording quite bright and annoying, whereas the Back on Black issue actually sounds like proper music. It's a bit too dark though, so it should be played with a bright cartridge. I used my Audio Technica VM750SH for it. I can send you a recording if you like.

 

[Frgirard]

I might as well tackle this one while I’m at it.

The whole idea that we’ve got recordings that are mastered “according to the original artists choices” is a myth.

The feedback that I get from studio engineers is that the only thing that the musicians want of the engineer is to produce a sound that will sell the most volume. This is nearly universal. Look at that Bob Ludwig video clip a few posts above, and notice how the musicians were exhorting the engineer to produce the sound identical to the Motown sound that was selling the most volume in the day. This is, apparently, completely normal. The engineer is left on his or her own to determine what sound to produce, with the sole mandate that it sells most volume.

So, really, the only thing we are getting with remastered versions is engineers attempting to sell the most volume to a new buyer segment, such as “modern audiophile LP buyer”, with the sound that buyer segment wants. And that is the way it has always been, anyway.

But, “original artist’s intent”? For recordings, it’s a non-thing. A myth.

+1

I doubt Coltrane or Bernard Herrmann left instructions after their death to compress their music during remasterings.

 

[levimax]

If I understand you correctly, what you're saying here, perhaps without realizing it, is that the remaster has also been EQ'ed to boost the 2 to 10 kHz range. Then, obviously, it should be relatively easy to ABX them, as EQ/frequency response easily produces the most audible changes.
And then if I also understand you correctly, you claim that this boost purely is the result of limiting/dynamic range compression, meaning if someone only uses a limiter (or at least the right limiter) the process will boost that frequency region.
I don't mean to sound antagonistic (rather I'm just trying to have a rational discussion about this), but you haven't provided any proof that this is actually the case. It seems very clear to me that in the cases you point to both limiting and EQ has been applied.
I provided some proof (the charts) that using only limiting doesn't change the frequency response to any noticeable degree, and then I've also done blind tests between the same songs uncompressed and then compressed. You would need to do the same.
So your claim that "low numbers always mean bright" simply can't be considered true, unless you provide evidence for this. It is of course true that low numbers always mean loud - obviously. But I would nevertheless say that if you dynamically compress one of those original CDs you mention I don't think it will sound as "loud" and "in your face" as you might think.
This loudness and "in your face" that is inherent in a lot of modern music I would say is rather a consequence of how the music is recorded, produced, mixed, and EQ'ed, rather than how much limiting is applied.
Also, since I've remastered/EQ'ed quite a lot of music myself I've also seen that if I boost or cut the frequencies in slightly different areas (at the "wrong" frequencies) and at slightly different levels I can make a song sound as if it "presses" on my skull, which I think is what many people think of when they think something sounds "compressed".

I asked the following question of someone else earlier in this debate, so now I'll have to ask you:
Have you tried taking uncompressed songs, e.g. the ones you have already compared to the remasters, and then dynamically compressed them, then lowered the peak volume to match the uncompressed original, and then done an ABX?
That way you can rule out if the frequency response is actually being altered by applying limiting.
If you haven't, please do the test .
I would be happy to send you the songs I used for my testing if you like, as I'm not trying to be antagonistic, but rather I'm trying to clear all this up.

I will, however, add a caveat to my long rant:
My producer/mixer friend told me that to make music excessively loud, some artists add a lot of analogue compression during the recording session, both on individual instruments and the finished mix. It then goes to the mastering engineer, who then applies more analogue compression, then EQ's it, and then finally adds limiting.
I think this might be part of the explanation to why some music sounds excessively loud, although the EQ'ing and mixing in the initial process (before the mastering engineer receives it) is in my opinion probably the main explanation, rather than the amount of compression used. A lot of music is produced and mixed to have excessive energy in the harshness region, and when I EQ/remaster music I almost always lower that region.
Although I don't know anything of the recording process, I would imagine that the song "Natural" by Imagine Dragons is an example of both mixing and EQ'ing to make it sound loud as well as adding a lot of analogue compression throughout the process before finally limiting it to DR6.

EDIT:
By the way: If the Nirvana album you listened to was "Nevermind", did you have a listen to the vinyl version on Back on Black? If not, that's my favourite version, and I think I've listened to them all. I like that version more than even the MFSL edition or the Bernie Grundman cut for ORG, which many hail as the best versions. I find the original recording quite bright and annoying, whereas the Back on Black issue actually sounds like proper music. It's a bit too dark though, so it should be played with a bright cartridge. I used my Audio Technica VM750SH for it. I can send you a recording if you like.

I do not know the answers to all your questions, I wish I did as I am very curious as well. I have looked at a good sample (over 50) of classic rock / pop and compared original LP, original CD, and remastered versions. My interest lies in what compressing older music does to it not what new music compressed on purpose sounds like. I have not played around with compressors as I assume I don't have access to the same type of compressor nor do I know how to use them like the professional mixers. I think that is something I will start to look into.

RE: Does compression cause elevated HF or is it caused by EQ added during the remastering process? I wish I knew the answer to this definitively but here is what I have observed as well as an explanation I have seen which makes sense to me.

A. Compression no doubt alters the FR response of a recording. By "bringing up" the quiet parts of a song and "limiting" the loud parts of a song it will change the overall FR because the loud and quiet parts of the music are going to have a different mix of frequencies which will show up in a FFT analysis.

B. I consistently (pretty much always) see elevated 2 Khz to 10 Khz and a slight decrease in the 200 Hz to 500 Hz region on remastered albums compared to the originals. It makes sense that if a compressor reduces peaks (which are commonly found in the "mid-bass / bass region" and increases lower level signals (found in higher frequencies) that compression would indeed alter the FR exactly as I have observed.

I am sure there is no one answer as some remasters have EQ and some don't and compressors and there setting vary but the fact is compression will always change the FR and FR is by far the most audible type of change you can make (especially the broad low Q HF boost present in almost all remasters)

My biggest frustration is why any Hi-Fi enthusiast would be OK with discarding the majority of the dynamic information in an older "popular classic song" which distorts the music in multiple ways. I can see why a car or ear bud listeners may prefer it but not a "Hi-Fi" listener. With streaming primarily playing the latest compressed re-mastered version of older songs I feel like the hobby is being dumbed down and the only way to find high quality recordings to play on my system is hunt down old versions of music or pay up for "audiophile recordings" which are hit or miss at best.

 

[MaxBuck]

There are going to be a lot of pop recordings that have no dynamics. Tell it to the judge.

Actually, the challenge is in finding pop recordings that do have dynamics. And given the way in which many of those audio products are consumed, dynamics has to be considered as a bug rather than a feature.

 

[DimitryZ]

Actually, the challenge is in finding pop recordings that do have dynamics. And given the way in which many of those audio products are consumed, dynamics has to be considered as a bug rather than a feature.

I don't know if Melody Gardot is pop or jazz, but to my ear her recent albums are quite dynamic, but maybe I am wrong.

Try Sunset in the Blue and Currency of Man. I will be curious to know how If You Love Me and Preacherman score on this DR thing.

 

[levimax]

If I understand you correctly, what you're saying here, perhaps without realizing it, is that the remaster has also been EQ'ed to boost the 2 to 10 kHz range. Then, obviously, it should be relatively easy to ABX them, as EQ/frequency response easily produces the most audible changes.
And then if I also understand you correctly, you claim that this boost purely is the result of limiting/dynamic range compression, meaning if someone only uses a limiter (or at least the right limiter) the process will boost that frequency region.
I don't mean to sound antagonistic (rather I'm just trying to have a rational discussion about this), but you haven't provided any proof that this is actually the case. It seems very clear to me that in the cases you point to both limiting and EQ has been applied.
I provided some proof (the charts) that using only limiting doesn't change the frequency response to any noticeable degree, and then I've also done blind tests between the same songs uncompressed and then compressed. You would need to do the same.
So your claim that "low numbers always mean bright" simply can't be considered true, unless you provide evidence for this. It is of course true that low numbers always mean loud - obviously. But I would nevertheless say that if you dynamically compress one of those original CDs you mention I don't think it will sound as "loud" and "in your face" as you might think.
This loudness and "in your face" that is inherent in a lot of modern music I would say is rather a consequence of how the music is recorded, produced, mixed, and EQ'ed, rather than how much limiting is applied.
Also, since I've remastered/EQ'ed quite a lot of music myself I've also seen that if I boost or cut the frequencies in slightly different areas (at the "wrong" frequencies) and at slightly different levels I can make a song sound as if it "presses" on my skull, which I think is what many people think of when they think something sounds "compressed".

I asked the following question of someone else earlier in this debate, so now I'll have to ask you:
Have you tried taking uncompressed songs, e.g. the ones you have already compared to the remasters, and then dynamically compressed them, then lowered the peak volume to match the uncompressed original, and then done an ABX?
That way you can rule out if the frequency response is actually being altered by applying limiting.
If you haven't, please do the test .
I would be happy to send you the songs I used for my testing if you like, as I'm not trying to be antagonistic, but rather I'm trying to clear all this up.

I will, however, add a caveat to my long rant:
My producer/mixer friend told me that to make music excessively loud, some artists add a lot of analogue compression during the recording session, both on individual instruments and the finished mix. It then goes to the mastering engineer, who then applies more analogue compression, then EQ's it, and then finally adds limiting.
I think this might be part of the explanation to why some music sounds excessively loud, although the EQ'ing and mixing in the initial process (before the mastering engineer receives it) is in my opinion probably the main explanation, rather than the amount of compression used. A lot of music is produced and mixed to have excessive energy in the harshness region, and when I EQ/remaster music I almost always lower that region.
Although I don't know anything of the recording process, I would imagine that the song "Natural" by Imagine Dragons is an example of both mixing and EQ'ing to make it sound loud as well as adding a lot of analogue compression throughout the process before finally limiting it to DR6.

EDIT:
By the way: If the Nirvana album you listened to was "Nevermind", did you have a listen to the vinyl version on Back on Black? If not, that's my favourite version, and I think I've listened to them all. I like that version more than even the MFSL edition or the Bernie Grundman cut for ORG, which many hail as the best versions. I find the original recording quite bright and annoying, whereas the Back on Black issue actually sounds like proper music. It's a bit too dark though, so it should be played with a bright cartridge. I used my Audio Technica VM750SH for it. I can send you a recording if you like.

OK I spent some time playing around with compressors in Adobe Audition and here is what I found. I used Lyle Lovett's song "I've got the blues". If you are not familiar with Lyle Lovett his first 4 albums are among the best popular recording for sound quality I have ever heard. Lots of dynamics and great production in every way. It shows what digital music "could have been" if the "loudness wars" has not ruined things. This song has a DR 17 and uses the dynamics as they should be used.... adds life and realism and very pleasant to listen to.

1. As expected I don't know what I am doing with compressors and the results were terrible. I used both "medium compression" and "high compression" presets and while they did their job and compressed the music from DR 17 to DR 9 it sounded horrible. I could ABX them 100% without trying. See below original and compressed wave forms.

2. While I was correct that compression changes FR I was wrong about how. Rather than causing a HF boost it appears that the frequency changes are fairly random and now I am quite sure every recording will behave differently. Apparently when they re-master they also mess with the EQ. See below, the green line is the original and the red line is compressed. In any case compression clearly adds a lot of FR distortion.

My conclusions: Compressing dynamic older music does more damage than I thought. Not only does it change the balance of loud to soft, it messes with the FR and basically ruins the original recording to a greater or lesser degree. After spending time ABX ing different original vs compressed recordings I am starting to get better at it and the changes are now much more obvious to me.... like the saying goes once you hear something you can never un-hear it.

My other conclusion is the DR Meter works OK. If you don't create bogus experiments to "prove" it doesn't work it actually does it's job quite well to at least give an indication of the amount of compression on different versions of the same song. For LP's vs CD I'm sure it is not the best but to compare different digital versions it does provide some useful information.

 

[Newman]

My other conclusion is the DR Meter works OK. If you don't create bogus experiments to "prove" it doesn't work it actually does it's job quite well to at least give an indication of the amount of compression on different versions of the same song.

A tool that sometimes works and sometimes doesn't, is broken.

Claiming that it takes special techniques to fool it, when, in truth, it only takes routine audio editing actions to see it fool itself, is disingenuous.

 

[levimax]

A tool that sometimes works and sometimes doesn't, is broken.

Claiming that it takes special techniques to fool it, when, in truth, it only takes routine audio editing actions to see it fool itself, is disingenuous.

The main objections I see is when the DR Meter is used to compare vinyl to digital which is always going to be an inherently difficult apples to oranges type of comparison if you don't understand the issues. The other objection I see is that the DR meter is used to mislead people which seems to be the norm in the Hi-Fi world... any measurement such as amplifier power, distortion, and SINAD can and have been used to mislead people if they don't understand the measurements so I don't see why the DR meter should be singled out for this. While it has flaws and limitations it was not designed to mislead anyone. Do you know of another tool that reliably and consistently can measure the "relative dynamics" of a recording for both digital and analog recordings?

While not perfect the DR Meter and it's database can be useful if you understand the issues and it is the best/ only source of information we have. Hopefully as time goes on there will be better measurements and information available but if history is our guide I would not expect much... as far as I can see ASR is the most reliable source of information for this hobby .... and people still can't agree what it means.

 

[Newman]

The main objections I see is when the DR Meter is used to compare vinyl to digital which is always going to be an inherently difficult apples to oranges type of comparison if you don't understand the issues.

Then you haven't read my post properly, or been across the issues sufficiently. Not just digital vs vinyl.

The engineer can apply normal compression to an effectively inaudible end result, and DR meter score goes down by 6. What has that to do with digital vs vinyl?

The engineer might want to blend or spread the bass L/R during mixing or mastering to get the sonic effects you want (not just to chow down for vinyl), and the DR meter score is jumping by 2,4,6, in any direction. What has that to do with digital vs vinyl?

...Do you know of another tool that reliably and consistently can measure the "relative dynamics" of a recording for both digital and analog recordings?...

You haven't shown me one that does. I hope you are not holding the TTDR tool in the air while asking me to show you another.

cheers

 

[levimax]

Then you haven't read my post properly, or been across the issues sufficiently. Not just digital vs vinyl.

The engineer can apply normal compression to an effectively inaudible end result, and DR meter score goes down by 6. What has that to do with digital vs vinyl?

The engineer might want to blend or spread the bass L/R during mixing or mastering to get the sonic effects you want (not just to chow down for vinyl), and the DR meter score is jumping by 2,4,6, in any direction. What has that to do with digital vs vinyl?



You haven't shown me one that does. I hope you are not holding the TTDR tool in the air while asking me to show you another.

cheers

If the compression is "inaudible" then why do it? Compression is always going to eliminate information and change the FR which to me is the opposite of "Hi-Fi". If your point is that artistic engineering choices can alter the DR Meter score unpredictably I would agree that is a limitation of the tool but if one is foolish enough to make a make a music purchasing choice based on 2 or 3 DR meter points I agree that is a mistake. On the other hand if an original CD has a DR 15 and one remaster is DR 5 and another one is DR 10 it does provide you with some useful information about the mastering style. For me some information even if flawed is better than no information. I know of no other better tools but I wish I did and I wish more objective information on different mastering's was available.

What I really wish is that I had a choice (besides buying decades old CD's and LP's) to purchase or listen to music mastered in a more dynamic/ Hi-Fi style. Pretty much all current popular music is highly compressed and even all the older music which was originally mastered with a more dynamic and Hi-Fi style has been re-mastered to sound like current compressed music. It is kind of weird that with all the modern technology now available at reasonable prices which is capable of reproducing dynamic Hi-Fi music that the new recordings are made to the lowest common denominator of being loud to stand out.

 

[Robin L]

If the compression is "inaudible" then why do it? Compression is always going to eliminate information and change the FR which to me is the opposite of "Hi-Fi". If your point is that artistic engineering choices can alter the DR Meter score unpredictably I would agree that is a limitation of the tool but if one is foolish enough to make a make a music purchasing choice based on 2 or 3 DR meter points I agree that is a mistake. On the other hand if an original CD has a DR 15 and one remaster is DR 5 and another one is DR 10 it does provide you with some useful information about the mastering style. For me some information even if flawed is better than no information. I know of no other better tools but I wish I did and I wish more objective information on different mastering's was available.

What I really wish is that I had a choice (besides buying decades old CD's and LP's) to purchase or listen to music mastered in a more dynamic/ Hi-Fi style. Pretty much all current popular music is highly compressed and even all the older music which was originally mastered with a more dynamic and Hi-Fi style has been re-mastered to sound like current compressed music. It is kind of weird that with all the modern technology now available at reasonable prices which is capable of reproducing dynamic Hi-Fi music that the new recordings are made to the lowest common denominator of being loud to stand out.

Most people listing to recorded music are not "audiophiles", they want to hear music in less than ideal circumstances. The car being one of the most popular and most problematic. "Audiophiles" are a bunch of "Karens", don't have all that much to do with how most people listen to music. If it's "popular" music, the folks making that music don't have any good reason to care what the "audiophiles" think. The bottom line is the actual concern.

 

[Frgirard]

Vinyl versus CD Amy Whinehouse, At the BBC


https://magicvinyldigital.net/2021/...review-vinyl-lp-cd-digital-files-streaming/3/

 

[board]

OK I spent some time playing around with compressors in Adobe Audition and here is what I found. I used Lyle Lovett's song "I've got the blues". If you are not familiar with Lyle Lovett his first 4 albums are among the best popular recording for sound quality I have ever heard. Lots of dynamics and great production in every way. It shows what digital music "could have been" if the "loudness wars" has not ruined things. This song has a DR 17 and uses the dynamics as they should be used.... adds life and realism and very pleasant to listen to.

1. As expected I don't know what I am doing with compressors and the results were terrible. I used both "medium compression" and "high compression" presets and while they did their job and compressed the music from DR 17 to DR 9 it sounded horrible. I could ABX them 100% without trying. See below original and compressed wave forms.

View attachment 143039

View attachment 143040

2. While I was correct that compression changes FR I was wrong about how. Rather than causing a HF boost it appears that the frequency changes are fairly random and now I am quite sure every recording will behave differently. Apparently when they re-master they also mess with the EQ. See below, the green line is the original and the red line is compressed. In any case compression clearly adds a lot of FR distortion.

View attachment 143041

My conclusions: Compressing dynamic older music does more damage than I thought. Not only does it change the balance of loud to soft, it messes with the FR and basically ruins the original recording to a greater or lesser degree. After spending time ABX ing different original vs compressed recordings I am starting to get better at it and the changes are now much more obvious to me.... like the saying goes once you hear something you can never un-hear it.

My other conclusion is the DR Meter works OK. If you don't create bogus experiments to "prove" it doesn't work it actually does it's job quite well to at least give an indication of the amount of compression on different versions of the same song. For LP's vs CD I'm sure it is not the best but to compare different digital versions it does provide some useful information.

I'm glad to see that you've done some experiments .

I have a couple of suggestions:

1: The charts I have shown were made with the free plugin CurveEQ. I would highly recommend to install it (I use it in Audacity, which is also free, but it can be installed in many other programmes), as the orange line that it draws makes it much easier to see the difference between two songs than in the charts you have shown.
The plugin is very simple and easy to use, and I would be happy to guide you through it if you need any help. I also use the plugin to "measure" frequency response on phono cartridges.

2: If it has your interest I can send you the files I compressed dynamically, and then you can see if you can hear any difference. When I first brought this up earlier in this thread I mentioned that I couldn't hear any difference with a song by the metal band Pallbearer, but after repeating the test with headphones, as opposed to speakers, I passed a blind test with 15/16 correct. When I figured out what to listen to it became easier to spot, but I would however still say the difference was relatively subtle and not terribly detrimental to the sound. On another song it was much easier to hear the difference, but again not something I would call extremely detrimental.
I have no idea how the compressor in Adobe Audition works, but it sounds like it's not very good. I used the one in Wavelab, and although it might be okay I have the impression that it's nothing special - at least I haven't heard of professional mastering engineers using it (although I haven't asked around). Often mastering engineers buy limiters (i.e. hardware) for fortunes, because apparently only the best ones can do limiting that is inaudible or almost inaudible. My producer/mixer friend also told me that a blind test between dynamically compressed and non-compressed music had been presented to professional engineers, and they all failed. If it has your interest I can ask my friend for the paper.
So, it sounds like the compression in Audition doesn't work very well, if it creates such enormous differences, but it doesn't with (certain) other limiters.

3: Can you send me the files you compressed dynamically? Then I can compress them to an equal amount in Wavelab as you did, and then send them back to you, and you can see if they sound different to the ones you compressed in Audition. I would also be happy if you sent me the files that you compressed so I can check them out, especially in CurveEq.

Again, this is not to be antagonistic, but just to figure all this out, and we both seem keen on finding the truth rather than waging war, so I'm happy we can talk about this .
If we exchange files we can use WeTransfer. We can figure this out by private messages.

 

[Newman]

....and if the end result is that you can compress a song (not in any super-tricky, special way) to get DR score down by 6 points, then struggle to tell them apart / clearly prefer one / hear anything obviously detrimental about the compressed version, then, a 6 point lower DR score is no way to even necessarily be suspicious that the compression itself is detrimental to that version of the song. It could be.... or it could be not.

In which case, the TTDR tool is worse than useless: it is actively deceptive and misleading.

cheers
Grant

 

[Pearljam5000]

Is there a TLDR of who won?