[bigshot]

Not only one engineer has continued to insist that the original, non-remastered CDs that I love so much were "not properly mastered". Again, I got banned from more forums by suggesting that if compressing, limiting, and cranking up the originals was "proper" mastering then they could "keep their dynamically squashed crap!" lol But since then I have evolved from more of attack mode to a learning mode. Fast-forward to present.

I haven't read the wall of replies yet, but I'll quickly respond to this one... The earliest CD releases were basically the masters made for LP release transferred straight to CD. They're faithful to how the album originally sounded, but they aren't necessarily more dynamic than later remasters. A good case in point is Led Zeppelin. A big reason I hang onto my original Led Zeppelin LPs is because they are *more* compressed than the later remastered CDs. For music like this, you want that "in your face" sort of sound. The remasterings have improved the sound quality at the expense of balls. In this case, compressed is better.

There are errors on both sides, and there are remasters that nail it and sound better than ever before too. It's because of creative decisions, and that can have something to do with the age of the master too. When an album is newly completed, the band and engineers all get check discs and carefully review everything to make sure it's perfect. The enthusiasm of creating something new makes everyone give their best. A lot of the recent remasters come at the point where the band is in their 70s and are basically just endorsing checks and the engineers are people who are guns for hire with no personal investment into the music they're mastering. That doesn't have anything to do with iPods or lossy audio formats. That is boredom on the part of the people making the remasters. I don't know what can be done about that, because "you can't go home again" sometimes.

There is something to be said for the first release being the best release... and the original format of release being better than music that has been adapted to suit new technology. But that isn't a hard and fast rule. For every legacy title that has been massacred on digital, there's one that has been vastly improved. When they congregate, experienced "rekkid clecktors" spend most of their time discussing this very subject. The best way to know what sounds the best is to ask someone who has heard various releases and can compare them. There's no formula or dogma that will tell you.

 

[TheSonicTruth]

ADMINS:

Please Delete - I don't know WHAT the heck I did!

 

[TheSonicTruth]

I haven't read the wall of replies yet, but I'll quickly respond to this one... The earliest CD releases were basically the masters made for LP release transferred straight to CD. They're faithful to how the album originally sounded, but they aren't necessarily more dynamic than later remasters. A good case in point is Led Zeppelin. A big reason I hang onto my original Led Zeppelin LPs is because they are *more* compressed than the later remastered CDs. For music like this, you want that "in your face" sort of sound. The remasterings have improved the sound quality at the expense of balls. In this case, compressed is better.

There are errors on both sides, and there are remasters that nail it and sound better than ever before too. It's because of creative decisions, and that can have something to do with the age of the master too. When an album is newly completed, the band and engineers all get check discs and carefully review everything to make sure it's perfect. The enthusiasm of creating something new makes everyone give their best. A lot of the recent remasters come at the point where the band is in their 70s and are basically just endorsing checks and the engineers are people who are guns for hire with no personal investment into the music they're mastering. That doesn't have anything to do with iPods or lossy audio formats. That is boredom on the part of the people making the remasters. I don't know what can be done about that, because "you can't go home again" sometimes.

There is something to be said for the first release being the best release... and the original format of release being better than music that has been adapted to suit new technology. But that isn't a hard and fast rule. For every legacy title that has been massacred on digital, there's one that has been vastly improved. When they congregate, experienced "rekkid clecktors" spend most of their time discussing this very subject. The best way to know what sounds the best is to ask someone who has heard various releases and can compare them. There's no formula or dogma that will tell you.

"[The earliest CD releases were basically the masters
made for LP release transferred straight to CD. They're
faithful to how the album originally sounded, but they aren't
necessarily more dynamic than later remasters."

So are the results uploaded on DR Database misleading? Because for this example, Journey's 'Escape', there are at least two remastered versions that return lower DR values than both the vinyl and original CD releases for 'Escape':

Most users upload their DR results from TT DR's Offline Meter.

 

[gregorio]

Notice the location of Zero on the VU meter dial. As can be seen by the spacing of the values on that dial, it is clear the ballistics of the VU meter are weighted toward the are between -3 to +3 VU. That area is given priority between the full 30 or so dB range of that thing. As long as you keep whatever it is you recording, broadcasting, playing back, etc. 'between the 3s', you are guaranteeing consistent level through your equipment, and relatively consistent volumes for your audience. It's a pretty fool-proof measurement tool, although it does not measure peak source content. It also ensured that switching between television stations back in the old analog days you wouldn't blast everyone out of the room surfing past specific channels ...
[2] Recordists, back at the start of the current phase of the loudness war, 'chased zero', even as it migrated from about 2/3 toward the top of the VU meter, all the way to the top of the Peak meter. Increasing doses of limiting and/or compression were applied to keep that meter as close to 0dBFS as possible at all times. Hence, one contributing factor to the loudness war was born. It's been ridiculed by nearly every studio engineer who's read my theory, but Bob Katz, who I have had the pleasure of shaking hands with, seems to suggest metering was a contributing factor, in the followng:

[1] This I'm afraid is very significantly incorrect. In the digital audio age of broadcast, the VU meter was replaced by the QPPM (Quasi-Peak Program Meter), the ballistics of which were very similar to a VU meter (approx 300ms response time) and it therefore read (and missed) peaks much as a VU meter did (hence quasi-peak). It was pretty much the opposite of a "fool proof measurement tool" because it was so easy to fool! You could fool it with compression and EQ for example and it was because it was so easy to fool (and routinely was) that there was so many complaints, and that's why the "loudness measurement" paradigm was developed in the first place and why the CALM Act was passed into law in the US in 2010. Switching channels was a different issue, there was no technical specification for channel broadcast levels as far I'm aware and over many years, some/many channels got louder and louder to out do their competitors, a channel broadcast loudness war if you like. All this was gradually killed by the CALM Act as it came into force during 2012/2013.

[2] Yes, I understand why that idea has been ridiculed and to be honest I'm a bit tempted myself! Are you actually taking about recordists (recording engineers) or mix or mastering engineers? And, Bob Katz is correct in what he is saying but it's not directly related to your "idea". You have to appreciate the point in time and context of what Bob is saying and I don't believe you do.

[1] In section 1, you stated that peak limiting and compression are exactly the same. I disagree. While they are both forms of dynamics compression I feel that they have quite different effects on the sound. The first, peak limiting, is a 10:1 to Infinity:1 ratio used to reign in outlier peaks, or even greater quantities of peaks not so high. Once those 2, 4, or 8dB of peaks are gone, you can apply corresponding gain to bring the rest of the audio up to full scale(or a half-db below it for safety).
[1a] That does not change whatever loudness relationship exists between a lead vocalist and their backing vocalist.
[1b] Limiting, plus makeup gain, just moves everything up higher in level. That lead singer will still sound just as much louder than the backup, just at a slightly lower playback volume on whatever it's being heard back on.
[2] Now compression, applied at lower thresholds than limiting is normally applied, can change how much louder that lead vocalist is compared to their backup.
[2a] It's not the same as controlling those variables at the mix stage, but it can make a difference, for better or worse, depending on how it is applied and the project itself.
[2b] Compression, whether at a 2:1 or 4:1 ratio, 'squeezes' closer together, loudness-wise, everything above a threshold predetermined again, by the engineer, and the needs of the specific project.

1. Yep, that's exactly what a compressor does as well, except for potentially having a lower ratio. BTW, technically a limiter with a 10:1 ratio is not strictly a limiter, it's a high ratio compressor.
1a. Yes it does, if you apply limiting to one or the other.
1b. That's exactly what compression does too.

2. Yep, that's exactly what a limiter will do too.
2a. Compression is always applied at the mix stage! However, in the analogue and less commonly the 16bit recording days, some compression was often ALSO applied during recording, to mitigate noise floor issues of the recording media (tape). And then more was applied during mixing.
2b. Yep, that's exactly the same as a limiter except with a ratio of infinity:1. I'm not sure why you picked specifically 2:1 and 4:1 ratios, what you suggest will occur at any ratio (greater than 1:1). BTW, both compressors and limiters (and pretty much all processors) operate on voltage values, not loudness, careful not to get the two confused!

A compressor and a limiter are essentially the same thing. The difference between you points 1 & 2 are not differences between a compressor and a limiter but differences between what you are feeding them. If you feed a compressor or limiter with the full mix you'll get point 1a and 1b. If you feed a compressor or limiter with part of a mix you'll get point 2.

I get what you are saying about the remastering. But does it have to involve additional dynamics processing on top of what was already applied to the original stereo master, *if* that is what the remaster is sourced from?

I'm not sure I understand. If there's no additional processing to what was already applied to the original master, then how is it a different master (a remaster), it would just be a reissue of the same master, not a remaster?

G

 

[Steve999]

Gregorio’s point: 1, This statement is actually incorrect. What you state about cameras' dynamic range is correct, even with HDR but we don't face the same problem. Even 16bit (CD) provides more dynamic range than we need, essentially from the threshold of audibility to the threshold of pain.

Me: Yes, I think you’re right on that one. I tried to write in fuzzy non-technical language so I couldn’t be proven wrong, but there it is, I was wrong.


Gregorio’s point: 2. That's not really correct either. A symphony orchestra probably has the largest dynamic range, roughly around 80dB or so, and shouldn't cause damage to our ears.

Me: I sort stand of by this one, sort of. I said if we listen at realistic dynamic range and volume, not just dynamic range. Then it depends where we are theoretically sitting in the recording—in the balcony, in the front row, in the orchestra itself—most classical musicians have permanent hearing damage, I do believe. At home, if you have 80 dB of dynamic range, which I easily concede is reproducible by modest but good gear, you are not generally going to listen with the softest sounds at one decibel, since that is normed to be softest sound many people can hear, and would just get obscured and drowned out by ambient noise, so I think you are going to push 85 or 90 dB at a minimum in peak passages, even if you do not hit realistic orchestra volumes. Now I did not know an orchestra had 80db of dynamic range, so I am working from that as a starting point and making the rest of this up as I go along. The main point being, however, that I said volume and dynamic range, not just dynamic range. As I think about it, there is probably room for argument on both sides.

I will look forward to reading what I got wrong this time. : )


Gregorio’s Point: 3. This point however, and the rest of your post, is entirely or mostly correct.

Me: You are kind, sir, and your obsevations are interesting, and your points well-taken.

—Steve

 

[gregorio]

So are the results uploaded on DR Database misleading?

Yes they are very misleading! The TT meter (used to create the DR database measurements) does not work correctly with vinyl, it will give a DR measurement around 4 higher with vinyl than is actually the case.



G

 

[TheSonicTruth]

[1] This I'm afraid is very significantly incorrect. In the digital audio age of broadcast, the VU meter was replaced by the QPPM (Quasi-Peak Program Meter), the ballistics of which were very similar to a VU meter (approx 300ms response time) and it therefore read (and missed) peaks much as a VU meter did (hence quasi-peak). It was pretty much the opposite of a "fool proof measurement tool" because it was so easy to fool! You could fool it with compression and EQ for example and it was because it was so easy to fool (and routinely was) that there was so many complaints, and that's why the "loudness measurement" paradigm was developed in the first place and why the CALM Act was passed into law in the US in 2010. Switching channels was a different issue, there was no technical specification for channel broadcast levels as far I'm aware and over many years, some/many channels got louder and louder to out do their competitors, a channel broadcast loudness war if you like. All this was gradually killed by the CALM Act as it came into force during 2012/2013.

[2] Yes, I understand why that idea has been ridiculed and to be honest I'm a bit tempted myself! Are you actually taking about recordists (recording engineers) or mix or mastering engineers? And, Bob Katz is correct in what he is saying but it's not directly related to your "idea". You have to appreciate the point in time and context of what Bob is saying and I don't believe you do.



1. Yep, that's exactly what a compressor does as well, except for potentially having a lower ratio. BTW, technically a limiter with a 10:1 ratio is not strictly a limiter, it's a high ratio compressor.
1a. Yes it does, if you apply limiting to one or the other.
1b. That's exactly what compression does too.

2. Yep, that's exactly what a limiter will do too.
2a. Compression is always applied at the mix stage! However, in the analogue and less commonly the 16bit recording days, some compression was often ALSO applied during recording, to mitigate noise floor issues of the recording media (tape). And then more was applied during mixing.
2b. Yep, that's exactly the same as a limiter except with a ratio of infinity:1. I'm not sure why you picked specifically 2:1 and 4:1 ratios, what you suggest will occur at any ratio (greater than 1:1). BTW, both compressors and limiters (and pretty much all processors) operate on voltage values, not loudness, careful not to get the two confused!

A compressor and a limiter are essentially the same thing. The difference between you points 1 & 2 are not differences between a compressor and a limiter but differences between what you are feeding them. If you feed a compressor or limiter with the full mix you'll get point 1a and 1b. If you feed a compressor or limiter with part of a mix you'll get point 2.



I'm not sure I understand. If there's no additional processing to what was already applied to the original master, then how is it a different master (a remaster), it would just be a reissue of the same master, not a remaster?

G

RE: Remasters:

Sure it can be a new/different master, even if no dynamics processing is added beyond what was in the legacy. Remastering, if a 2ch stereo master is sourced, can also correct speed issues, L to R channel imbalances, improve the tonal quality with minor EQ corrections. So there's processing besides dynamics that can be used, as necessary.

Who says a remaster *must* always involve dynamics processing? I guess that last question is what has gotten me so worked up over what is sold to the public as "remasters".

 

[TheSonicTruth]

[1] This I'm afraid is very significantly incorrect. In the digital audio age of broadcast, the VU meter was replaced by the QPPM (Quasi-Peak Program Meter), the ballistics of which were very similar to a VU meter (approx 300ms response time) and it therefore read (and missed) peaks much as a VU meter did (hence quasi-peak). It was pretty much the opposite of a "fool proof measurement tool" because it was so easy to fool! You could fool it with compression and EQ for example and it was because it was so easy to fool (and routinely was) that there was so many complaints, and that's why the "loudness measurement" paradigm was developed in the first place and why the CALM Act was passed into law in the US in 2010. Switching channels was a different issue, there was no technical specification for channel broadcast levels as far I'm aware and over many years, some/many channels got louder and louder to out do their competitors, a channel broadcast loudness war if you like. All this was gradually killed by the CALM Act as it came into force during 2012/2013.

[2] Yes, I understand why that idea has been ridiculed and to be honest I'm a bit tempted myself! Are you actually taking about recordists (recording engineers) or mix or mastering engineers? And, Bob Katz is correct in what he is saying but it's not directly related to your "idea". You have to appreciate the point in time and context of what Bob is saying and I don't believe you do.



1. Yep, that's exactly what a compressor does as well, except for potentially having a lower ratio. BTW, technically a limiter with a 10:1 ratio is not strictly a limiter, it's a high ratio compressor.
1a. Yes it does, if you apply limiting to one or the other.
1b. That's exactly what compression does too.

2. Yep, that's exactly what a limiter will do too.
2a. Compression is always applied at the mix stage! However, in the analogue and less commonly the 16bit recording days, some compression was often ALSO applied during recording, to mitigate noise floor issues of the recording media (tape). And then more was applied during mixing.
2b. Yep, that's exactly the same as a limiter except with a ratio of infinity:1. I'm not sure why you picked specifically 2:1 and 4:1 ratios, what you suggest will occur at any ratio (greater than 1:1). BTW, both compressors and limiters (and pretty much all processors) operate on voltage values, not loudness, careful not to get the two confused!

A compressor and a limiter are essentially the same thing. The difference between you points 1 & 2 are not differences between a compressor and a limiter but differences between what you are feeding them. If you feed a compressor or limiter with the full mix you'll get point 1a and 1b. If you feed a compressor or limiter with part of a mix you'll get point 2.



I'm not sure I understand. If there's no additional processing to what was already applied to the original master, then how is it a different master (a remaster), it would just be a reissue of the same master, not a remaster?

G

Assuming we are talking about the final 2-channel here, how can a limiter make a back-up vocalist, subjectively half as loud as the lead vocalist, almost as loud as that lead? I guess what i'm trying to say is that limiters are concerned mainly with peaks of things, of which don't normally include vocals. And as far as 10:1 ratio is concerned, I have read online, and in texts, that that ratio is considered the entry into limiter territory. Apply it at or below the apparent loudness of the backup singer and you'll get a squashed, distorted mess that might actually damage your speakers if you turn it up beyond a whisper.

 

[gregorio]

I sort stand of by this one, sort of. I said if we listen at realistic dynamic range and volume, not just dynamic range. Then it depends where we are theoretically sitting in the recording—in the balcony, in the front row, in the orchestra itself—most classical musicians have permanent hearing damage, I do believe.

Sorry, I'm still going to have to disagree. Yes, it's true that it entirely depends on where you're sitting and you're going to get a much higher dynamic range if you're sitting next to a back desk viola player right in front of the trumpet. However, in practice no one would release such a recording because it would sound completely wrong, the balance would be screwed, the violins would be too quiet and the trumpets would over balance everything. The most extreme practical recording would be in the front row of the audience, where it's possible you could experience up to about 110dB peak levels (although only for very short periods). However, there is the noise floor to consider, the air conditioning and noise/movement of the audience (and musicians), most likely that noise floor is 30dB or higher and therefore the dynamic range would be 80dB. Ideally/Typically though, the recreated audience perspective is several rows back, nearer the middle of the concert hall and there we would be looking at a dynamic range much closer to 60dB (short peaks no higher than about 90dB).

G

 

[Steve999]

I never thought about it that much. And I probably never will again. I believe you are correct. Thanks. : )

Sorry, I'm still going to have to disagree. Yes, it's true that it entirely depends on where you're sitting and you're going to get a much higher dynamic range if you're sitting next to a back desk viola player right in front of the trumpet. However, in practice no one would release such a recording because it would sound completely wrong, the balance would be screwed, the violins would be too quiet and the trumpets would over balance everything. The most extreme practical recording would be in the front row of the audience, where it's possible you could experience up to about 110dB peak levels (although only for very short periods). However, there is the noise floor to consider, the air conditioning and noise/movement of the audience (and musicians), most likely that noise floor is 30dB or higher and therefore the dynamic range would be 80dB. Ideally/Typically though, the recreated audience perspective is several rows back, nearer the middle of the concert hall and there we would be looking at a dynamic range much closer to 60dB (short peaks no higher than about 90dB).

G

 

[bigshot]

"So are the results uploaded on DR Database misleading?.

I don’t trust that particular way of measuring dynamic range. There are too many variables to be able to boil a whole album down to a single number.

 

[RRod]

As for what those other engineers said, they didn't say outright, but seemed more to imply that 16bit "really was not an adequate deliverable", and that DRC needed be used to keep things as much as possible above the 16 bit noise floor.

How often have you noticed the 16-bit noise floor even in classical recordings?

"[The earliest CD releases were basically the masters
made for LP release transferred straight to CD. They're
faithful to how the album originally sounded, but they aren't
necessarily more dynamic than later remasters."

So are the results uploaded on DR Database misleading? Because for this example, Journey's 'Escape', there are at least two remastered versions that return lower DR values than both the vinyl and original CD releases for 'Escape':





Most users upload their DR results from TT DR's Offline Meter.

The DR rating depends on the 2nd highest peak. If I lop-off 6dB from that peak, and the RMS stays (roughly) the same, then the DR will go down by 6. But as we see in my previous example, that doesn't suddenly mean a real audible difference.

 

[TheSonicTruth]

How often have you noticed the 16-bit noise floor even in classical recordings?


The DR rating depends on the 2nd highest peak. If I lop-off 6dB from that peak, and the RMS stays (roughly) the same, then the DR will go down by 6. But as we see in my previous example, that doesn't suddenly mean a real audible difference.

To your first question: Not sure, really! What does the digital noise floor, at any bit-depth, sound like exactly - a softer version of tape hiss? If so then what I think I heard during quieter parts of Berlioz's 'Fantastique' was the air handling ducts in whatever hall that was recorded.

Regarding first vs second-highest peaks, that would seem to make sense. Suppose the very highest 'peak' was really a nasty scratch right through the CD's label side into the read layer sandwiched so moronically 1/10000000 of 1mm right behind it? LOL! So including only the next highest group of peaks probably helps the accuracy of TT DR. (Chuckling to myself imagining a scratch or blemish tricking TT DR into a result of DR47, LMAO! )

 

[RRod]

To your first question: Not sure, really! What does the digital noise floor, at any bit-depth, sound like exactly - a softer version of tape hiss? If so then what I think I heard during quieter parts of Berlioz's 'Fantastique' was the air handling ducts in whatever hall that was recorded.

Well the issue with a truncated noise floor is that it modulates with the signal, so it depends entirely on what the signal is down there. If you've dithered, it sounds like hiss, but for other things it can sound like a weird robot kind of thing. You can always gain something down 90dB or so without dither then gain it back up to test. Or you can take a fade-out and gain it up. I can put up an example later.

 

[gregorio]

Sure it can be a new/different master, even if no dynamics processing is added beyond what was in the legacy. Remastering, if a 2ch stereo master is sourced, can also correct speed issues, L to R channel imbalances, improve the tonal quality with minor EQ corrections. Who says a remaster *must* always involve dynamics processing?

All those corrections can or will involve changes in the dynamics (levels) and therefore dynamics processing is virtually always involved. Secondly is "achieving a mix that approximates the original release as millions of listeners remembered it 20, 40 years ago." How people remember it years ago is relative to how they heard all music years ago, how they hear it today is relative to how they hear music today, which generally is louder! How much louder a remaster should be, comes down to a judgement call, exactly what the original master is, it's genre, what it is trying to achieve musically and what is an appropriate amount of compression.

[1] Assuming we are talking about the final 2-channel here, how can a limiter make a back-up vocalist, subjectively half as loud as the lead vocalist, almost as loud as that lead?
[2] I guess what i'm trying to say is that limiters are concerned mainly with peaks of things, of which don't normally include vocals.
[3] And as far as 10:1 ratio is concerned, I have read online, and in texts, that that ratio is considered the entry into limiter territory.

1. If we're talking about a "final 2-channel here" then a limiter cannot make "a back-up vocalist subjectively half as loud as the lead vocalist, almost as loud as that lead" and neither can a compressor! A compressor cannot do this for exactly the same reason as a limiter can't, because essentially they are exactly the same thing!

2. That's all true BUT it's all true of a compressor as well because they are the same thing. If you're going to say that a compressor can act on the vocals, that would be true but it would be equally of true of a limiter. A limiter is the same as a compressor, the only difference is that a limiter compresses to a limit whereas a compressor doesn't, unless it's ratio is set to infintity:1 with a very short attack time, in which case our compressor is now a limiter!

3. Yes it is entering into limiting territory but it depends on where the threshold is set and the input signal it is fed. I'm not sure you understood what I stated earlier about compressor settings? Let's use your example song again, ff the input signal goes up to say 0dBFS then with a 10:1 ratio and limit set to -40dB the output signal would be -36dB, it won't have actually limited to -40dB and therefore it's not actually a limiter! If however we set it much more sensibly, say -6dB then with a 10:1 ratio our output will indeed be -6dB (or a decimal point or so above), it is now a Limiter! And obviously (hopefully!), if we fed that same -6dB signal into a compressor (with a suitably short attack time and a 10:1 ratio) our compressor would also limit to -6dB and therefore is now a Limiter! A ratio of 10:1 under most usual conditions is enough to be a limiter but to be certain it's ALWAYS a limiter then the ratio has to be infinity:1.

G