[Strangelove424]

someone will always be unhappy about the mastering choices of a specific album. I understand those who somehow hope to get the emotion of a loud and very dynamic experience at home. crank the amp up and imagine being at a live event. I also understand that if I had only that type of albums with how long I spend each day listening to music, my ears today would be absolutely ruined and I'd be one of those guys with an almost permanent headache. or I would play it cool and fail to hear half of the songs because with the dynamic it would be too quiet.
the general loudness war was/is a monstrosity, pretty much everybody agrees. but first, there's a reason why it was used so much, mainly because it worked. let's blame stupid subjectivity instead of the people who found out how to make use of it(which is exactly what they were paid to do all along!!!!!!!). and second, it never ever meant that compression in general was bad or evidence of a sound engineer who doesn't know his job. the polarized view about compression needs to go away even more than the loudness war itself. I don't know what I would do if all my favorite bands started releasing only albums with uncompressed sound. that would be the real nightmare for me.

Loudness and dynamics are not really the same thing. Lullaby music has low dynamics because it is meant to be calming, but it is also not typically thought of as "loud". There are many things that go into making a good album, and many things that go into screwing one up. Lack of dynamics, or loudness, are just some of the ways. I think people need to be a lot more precise about the way they discuss things, and was hoping a thread focused on analysis tools would achieve that. I see a lot of people applying theory and idealized conjecture to mastering, as though all music was under one giant umbrella. I feel like people keep boiling it all down to "Things sucks because they are loud." But it's not that simple. Different genres have been effected in different ways. And quality is effected by other things like frequency balance, or recording. This was never meant to be a complaint thread about mastering, because there are plenty of those to go around. The goal was to help people wanting to spend their time/money on their ideal mastering version - and wanting some verification other than word of mouth to do so. Maybe there should be a rule going forth that all comments must be in context to an actual album. That would throw all the ideologues off, actually having to listen to specific music and comment directly about it.

 

[Husky]

1. No, (...)

Let's conclude that we have some different tastes and feelings about the use of clipping/compression

2. (...) Like you, my mum was "frustrated"!

See above, but let me add that I'm open minded and not that old...

3a. How many more millions of people preferred the vinyl?

Since vinyl is harder to get and limited in playback devices, you can't count like that. But despite the cost and limitations of vinyl, it has become more popular since the raise of the loudness war. I think it is for the reason of sound quality - not because of the format but because of the different mastering (based on what I've analyzed/read/heard).

4. (...) why don't you do something to try and cure your ignorance rather than just keep coming out with nonsense statements? (...) It was already a hot topic of debate amongst engineers when I entered the business in 1991(...)

Still you're unable to proof "nonsense" in my "statements". I have the impression that if something contradicts your personal opinion you classify it as nonsense.
Fine if you admit that the debate is old and not all of the engineers defend the loudness race as you seem to do (by defending the results). Maybe you have seen that I'm not new to the subject either.

5. In effect, yes.

OK, astonishing, let it be your personal experience. I have very different experiences with artists and from what I've read.

6. Well those two exception would be a great place to start!!

Concerning interpretation of music style:
"de custibus non est disputandum", I think it's of no use to further dispute our different taste.

Concerning the use of loudness measures:
There are some good hints in the article why the LRA (EBU 3342) didn't decrease. Additionally I think that the restriction of an upper percentile of 95% may not show some differences relevant to the listening impression with good equipment and environment (some of those peaks and transients could make a difference in my opinion).
In my listening experience, EBU R 128, ReplayGain, DR and the measures in ClippingAnalyzer (Crest-factor, clippings, compression indicator, distribution analysis, histogram, waveform) are well suited to predict/substantiate if I like the sound quality of a track.

You are confusing peak level with dynamic range. Yes, in the first row of an audience you may get 100dB but then it's unlikely the noise floor is lower than about 40dB, so that would be a dynamic range of 60dB (...)

(...) I can't think of any of them that had peak levels of 100dB or more. (...) It seems that the justification for crazy technical requirements has very little to do with real world music making.

You can read about peak levels of 135dB (resulting in 100dB 22m away at the audience) in classical concerts e.g. here:
http://www.noiseandhealth.org/artic...6;issue=68;spage=40;epage=46;aulast=Rodrigues

To make my point clearer: I totally agree that 120dB dynamics on a audio format isn't really needed to play back music at the listener. So there is plenty of headroom that can be used. In earlier times compression (and noise reduction) had some reason in the poor dynamics of the formats. Nowadays there is no need to push the level, no need to have the volume touch 0 dB (I don't mean normalization with this). But it seems some people still love to go there and hate headroom.

The goal was to help people wanting to spend their time/money on their ideal mastering version - and wanting some verification other than word of mouth to do so.

That's exactly the reason why I posted about the ClippingAnalyzer tool (http://www.ber-sd.com/dl_clipping_eng.html). The screenshots were only to show how the results of the tool look like.

 

[luckybaer]

Thanks for posting this.

 

[bigshot]

You can read about peak levels of 135dB (resulting in 100dB 22m away at the audience) in classical concerts e.g. here:
http://www.noiseandhealth.org/artic...6;issue=68;spage=40;epage=46;aulast=Rodrigues

That article says that from the conductor's position, it was about 82dB, which is a normal loud listening volume. Out in the audience it would be even less. And as it has already been explained, peak level isn't the same as dynamic range. You would have to account for at least a 30dB noise floor (more like 40dB with a full audience). That would make the dynamic range in the neighborhood of... surprise surprise... well under 50dB.

 

[RRod]

That article says that from the conductor's position, it was about 82dB, which is a normal loud listening volume. Out in the audience it would be even less. And as it has already been explained, peak level isn't the same as dynamic range. You would have to account for at least a 30dB noise floor (more like 40dB with a full audience). That would make the dynamic range in the neighborhood of... surprise surprise... well under 50dB.

Some else used to make these same arguments. "Well if you put your ear right up to the timpani head you get 140dB!" I've had the privilege of sitting in the back of the viola section right in front of the 'bones for Mahler. No one should want to replicate that experience!

 

[bigshot]

I have a recording of a violin concerto where someone had the bright idea of seating the orchestra in a circle around the microphone- all equidistant. They put the tympani right next to the violin soloist so whenever the drum hits come up, you don't hear anything else. There's a reason some instruments are placed in the rear!

 

[gregorio]

[1] Let's conclude that we have some different tastes and feelings about the use of clipping/compression
[1a] See above, but let me add that I'm open minded and not that old...
[2] But despite the cost and limitations of vinyl, it has become more popular since the raise of the loudness war.
[2a] Still you're unable to proof "nonsense" in my "statements".
[2b] Fine if you admit that the debate is old and not all of the engineers defend the loudness race as you seem to do (by defending the results).
[3] OK, astonishing, let it be your personal experience. I have very different experiences with artists and from what I've read.
[4] In my listening experience, EBU R 128, ReplayGain, DR and the measures in ClippingAnalyzer (Crest-factor, clippings, compression indicator, distribution analysis, histogram, waveform) are well suited to predict/substantiate if I like the sound quality of a track.
[5] You can read about peak levels of 135dB (resulting in 100dB 22m away at the audience) in classical concerts...
[5a] I totally agree that 120dB dynamics on a audio format isn't really needed to play back music at the listener. So there is plenty of headroom that can be used.
[5b] In earlier times compression (and noise reduction) had some reason in the poor dynamics of the formats.

1. Well that's apparently a major change and step forward in your stance. Concluding that it's about "different tastes" is a major change from concluding it's "faulty mastering"!
1a. But that was an analogy of my mother's view on the use of heavy distortion on electric guitars. I'm not accusing you of being old and close minded about guitar distortion, you seem to understand and appreciate why distortion applied to an electric guitar can be musically desirable. What you're apparently incapable of understanding and appreciating is why distortion applied elsewhere might be desirable!

2. Huh? Loudness started becoming a serious issue (a "war") in the late '80's and worsened significantly, to it's present state, during the '90's and early 2000's. What happened to the sales of vinyl during that period?
2a. I don't need to, you're doing that all by yourself. This point #2 is a good example, the correlation between the loudness war and vinyl sales is in fact the exact opposite of what you're claiming/implying. Here's another example:
2b. I am absolutely not defending the loudness war, I've made that clear on several occasions and, I've been fighting the loudness war for about 25 years. My argument is NOT defending the loudness war and your assertion that I am is nonsense! My argument is about how you (and many/most audiophiles) categorise/define the loudness war and the blanket view that heavy compression/limiting is automatically "faulty mastering". That is a nonsense view, not least because sometimes (fairly often) it doesn't actually have anything to do with mastering in the first place!

3. Why "let it be my personal view"? You think that I've never discussed the loudness war or what artists want with any of my colleagues (other pro audio engineers) in the last 25+ years and that I've never read anything about it? Again, nonsense. Actually, your question and the way you phrased it was nonsense, which you would know if you had you any significant experience and were therefore able to contextualise whatever it is that you have read. Typically, artists do not "tell the engineer to brick wall with audible distortions because they feel it was musically better". If you had any significant experience you would know that artists typically tell the mastering engineer that they want their master to be at least as loud as band/artist X, Y and Z because they feel it is musically better. Despite you stating that you've read the article, you seem to have deliberately ignored or misunderstood the most important part, the conclusion, which paraphrasing is: "In the end, it's all about style. .. With some genres, very heavy compression/limiting is probably a good idea. ... Do you want every loud attack modified [distorted] by a compressor/limiter? It might be a good idea in many cases, but it might prove disastrous in others."

4. This statement raises two points: 1. Exactly, that's your experience and your preference. Assuming you're telling the truth, then you cannot like a number of genres; EDM, grunge, hip-hop and in fact pretty much all genres and sub-genres since the early '90's. That's entirely your choice of course but if you're going to assert it's "faulty mastering" then you're doing exactly as my mum did with Hendrix and for exactly the same reason. 2. Neither EBU R 128, ReplayGain, the DR database or ClippingAnalyzer give any direct or accurate measurement of either masting quality, the amount of compression applied, where it's been applied or whether or not it's appropriate and in fact all of these measurement/specification types could easily be passed with pink noise. Obviously we have different tastes, I wouldn't like a track of nothing but pink noise and couldn't appreciate it as a good example of mastering but maybe that's just me?

5. Again, that's peak level, NOT dynamic range. Peak level and dynamic range are only the same if the noise floor is at (or very near) 0dB, which is a physical impossibility in a room with 90 odd musicians and hundreds/thousands of audience members!
5a. What purpose does "plenty of headroom" serve?
5b. But that was only one reason for the use of compression. Around the mid/late '60's there were other, musical reasons for using compression, even moderately heavy (over-driven) compression. Not satisfied with eliminating many/most genres since the '90's on the grounds of compression distortion, should we also eliminate most genres from the '60's onwards as well? I presume not, in which case there must be something seriously faulty in the audiophile rationale of compression/heavy compression being an evil which must be eliminated.

G

 

[gregorio]

[1] I think people need to be a lot more precise about the way they discuss things, and was hoping a thread focused on analysis tools would achieve that.
[2] The goal was to help people wanting to spend their time/money on their ideal mastering version - and wanting some verification other than word of mouth to do so.

1. Yep, precision is one problem. Simple things like loudness and dynamic range only appear simple because we can sense them almost without any conscious effort but in practice they're not simple and measuring them is both difficult and imprecise. Significant time and money has been spent on loudness but the best measurement we currently have is still rather imprecise, it's effectively based on the averaged responses of various groups of test subjects and even then, it only provides this (somewhat imprecise) accuracy under certain circumstances. In the case of compression, probably the most accurate indicator we have is crest factor but it is only an indicator, rather than a measure of compression and it's accuracy in this regard is woeful. Baring in mind a low crest factor "indicates" more compression, it's entirely possible to record something that's heavily compressed, record something else with no compression at all and for the heavily compressed recording to have a higher crest factor.

2. A laudable aim in theory but in practice, impossible. Even with a comprehensive set of professional tools it's impossible, let alone with cheap/simplified consumer tools. For example, as a relatively simple tool to demonstrate and publicise the general effect of the loudness war, the DR database is a good tool but if used to determine "good" from "bad" between individual recordings, then it's a bad and potentially highly misleading tool. So, we are largely dependent on word of mouth and that of course raises not only personal subjective opinion but also understanding, knowledge and experience. The level of misunderstanding and mis-attribution typically displayed by audiophiles is often quite shocking. How for example, can one judge the quality of the mastering? A mediocre end result could be due to a great mastering job performed on a poor mix or equally, it could be due to a poor mastering job performed on a good mix, how can we tell the difference? I've been doing this professionally for many years and sometimes I can't tell, though usually there are clues. Taking everything into account; the compositions, the lyrics, the arrangements, the performances, the recording, the editing, the production and mixing and finally the mastering, there are relatively few great albums. All these elements not only have to come into perfect alignment but must also gel perfectly and having all that at the same time on one album is a very rare occurrence.

G

 

[Husky]

That article says that from the conductor's position, it was about 82dB (...) peak level isn't the same as dynamic range. (...) dynamic range in the neighborhood of (...) well under 50dB.

Obviously you didn't read the article in detail. Peak at conductor's position was max. 120dB:

Now let's take 100dB as peak (could be a little more without C-weighting) and 30dB noise floor (seems reasonable according to this study: http://orbit.dtu.dk/files/9878048/in12_916_pierre.pdf), you would need 70dB dynamics.
But for your argument, let's take <50dB dynamic range. As a consequence, 8bit could be enough for you (48dB SNR)? Hmm, maybe not... (try it if you don't know).
Besides, there is a huge difference in listening to a noise floor resulting from the audience and to a technical noise floor (e.g. quantization error).
Again: modern formats allow more than enough dynamic range, so there is no need to squash dynamics for the format, that's all.

 

[Husky]

(...)the correlation between the loudness war and vinyl sales is in fact the exact opposite of what you're claiming/implying.

It took some time for the consumer to be aware of the loudness war, took some time to react, some time in the industry to release LPs again.
The Wikipedia article in English was created end of 2005, in German in 2008, then vinyl sales started to rise.
I don't know if there is a study about it and I don't pretend to have made my own, but as I wrote I know from forums, articles (e.g. http://www.soundmattersblog.com/vinyl-vs-cd-in-the-loudness-war/), friends that the LP became attractive and preferred over the CD because of better sound quality coming from less hypercompression/clipping.

2b. (...) I've been fighting the loudness war for about 25 years. (...) My argument is about how you (...) categorise/define the loudness war and the blanket view that heavy compression/limiting is automatically "faulty mastering". That is a nonsense view, not least because sometimes (fairly often) it doesn't actually have anything to do with mastering in the first place!

Good to know that you fight the loudness war, you could express that clearer in your statements.
You think it's a nonsense view if me and lots of people (otherwise there wasn't a longtime broad debate) don't like hypercompression and especially audible distortion where it distracts from the music and results in fatiguing, blurred sound, and see it as a fault?
For me personally, clipping that happened by making a track louder (that is it doesn't seem to support the music at relevant parts), is clearly faulty, even if the mastering engineer or anybody else anywhere in the whole process did it by purpose.
And of course these things and bad sound quality often are present way before mastering. I've never ever pretended that only the mastering engineer is to blame!!

(...) Typically, artists do not "tell the engineer to brick wall with audible distortions because they feel it was musically better". If you had any significant experience you would know that artists typically tell the mastering engineer that they want their master to be at least as loud as band/artist X, Y and Z (...)

Well, that's what I've said actually, maybe you should go above and read my post and your previous answers again. But good if we can agree on this small part.

(...)you cannot like a number of genres; EDM, grunge, hip-hop and in fact pretty much all genres and sub-genres since the early '90's. (...)

Wrong. I like this music. Again, distinguish sound quality from genre. A track doesn't have to clip or sound bad just because of its genre!
Take Nirvana as an example. I very much like Nevermind. Especially the MFSL mastering. There is no clipping on that CD, DR 11-13 (original CD 10-13), compression indicator is 0-7%.
Besides, all genres are affected by the loudness war.

(...)I wouldn't like a track of nothing but pink noise and couldn't appreciate it as a good example of mastering but maybe that's just me?

Again you confuse musical content with compression/limiting/clipping. It's ridiculous nonsense if you conclude we'd have to listen to pink noise.
I recommend you to read the explanations and links I provided in order to understand why pink noise is a good reference for measuring hypercompression.

5. Again, that's peak level, NOT dynamic range. (...)

See my answer in my last post above.

I won't spend a lot more time discussing with you. It seems most things have been said and we simply have different opinions and are not easy to convince.

 

[castleofargh]

Obviously you didn't read the article in detail. Peak at conductor's position was max. 120dB:

Now let's take 100dB as peak (could be a little more without C-weighting) and 30dB noise floor (seems reasonable according to this study: http://orbit.dtu.dk/files/9878048/in12_916_pierre.pdf), you would need 70dB dynamics.
But for your argument, let's take <50dB dynamic range. As a consequence, 8bit could be enough for you (48dB SNR)? Hmm, maybe not... (try it if you don't know).
Besides, there is a huge difference in listening to a noise floor resulting from the audience and to a technical noise floor (e.g. quantization error).
Again: modern formats allow more than enough dynamic range, so there is no need to squash dynamics for the format, that's all.

just nitpicking here, but classical music being the one genre where compression is the less applied, if at all, I have a hard time understanding what the loudness war has to do with it.
the other day according to you, vinyls were preferred to CD, now you make your argument on how 50dB isn't enough dynamic. no matter the side of the argument, I spot like a small flaw. ^_^ the medium is not the reason why albums are overly compressed, or compressed at all. it's a non argument and we actually agree on this. except that your conclusion is that we then should stop compressing music, while mine is that if it wasn't the reason in the first place, then it's unrelated and that's it. but let's just stick to the agreed part, it's not the reason why albums are compressed and we could indeed have a lot more dynamic on albums if we wanted to.

how loud a guy can play in a live event doesn't set a standard for how dynamic the album must be IMO. I know a few people do have the gears and the desire to play music at home with 120 or more dB peaks, but it's fair to assume that the average guy listening to a song will not. when making an album you obviously have to account for how loud people will play it back in general. same for the listening conditions that will not necessarily be a quiet room. I'm not for albums stuck within 3dB, but it's unrealistic to argue for albums to have a certain dynamic based on how loud and dynamic some concerts can be as those are not the expected playback conditions.
and about compression in general, when the track with the super loud instrument is attenuated in gain so that other quieter instruments aren't completely drowned by it, the final dynamic is reduced. yet it's done for the listener. when some frequency range in the track is attenuated with EQ so that another instrument can be better perceived, that too might somehow end up reducing the dynamic range if that lowered range happened to get real loud. but again that was done for us listeners. when a voice is compressed because in general it's just more dynamic than most instruments so it might not blend in and become part of the music without compression, or when it's compressed locally simply to make it less sibilant(something very few people enjoy in excess), again that is compression and it's done for us.
as for global compression specifically for the idea of loudness war, if over compressed albums were making a systematic commercial flop, we could come to the consensus that it's just plain bad and we need it gone. but whatever the reason why, this is not what happens. it makes arguing about what's right or not much more complex. there is such a thing as too much of any tool and any effect, but how much is too much can end up being judged the same way we deal with how much salt should go in a meal. for some a little bit is already too much, for others, a all damn lot is what makes the food so good.

Loudness and dynamics are not really the same thing. Lullaby music has low dynamics because it is meant to be calming, but it is also not typically thought of as "loud". There are many things that go into making a good album, and many things that go into screwing one up. Lack of dynamics, or loudness, are just some of the ways. I think people need to be a lot more precise about the way they discuss things, and was hoping a thread focused on analysis tools would achieve that. I see a lot of people applying theory and idealized conjecture to mastering, as though all music was under one giant umbrella. I feel like people keep boiling it all down to "Things sucks because they are loud." But it's not that simple. Different genres have been effected in different ways. And quality is effected by other things like frequency balance, or recording. This was never meant to be a complaint thread about mastering, because there are plenty of those to go around. The goal was to help people wanting to spend their time/money on their ideal mastering version - and wanting some verification other than word of mouth to do so. Maybe there should be a rule going forth that all comments must be in context to an actual album. That would throw all the ideologues off, actually having to listen to specific music and comment directly about it.

well we kind of say the same thing with opposite examples. something made to be played quietly wouldn't be created with huge dynamic, it just wouldn't work. it doesn't mean that very loud requires huge dynamic, but something made for normal to quiet listening will require a fairly small dynamic if we want to simply perceive the all thing. but indeed that's just a limited aspect of what will become a great sound, something judged as technically bad, or just something we don't like.
about tools to check the quality, I agree with Greg.

 

[bigshot]

Again: modern formats allow more than enough dynamic range, so there is no need to squash dynamics for the format, that's all.

Modern formats have nothing to do with dynamic range in recordings. Aesthetic choices and comfort in listening under normal conditions do. A recording that uses a dynamic range of 70dB is going to require getting up and adjusting the volume control all through the recording. It's great to have such overkill on our formats nowadays. Much better than when I started out listening to vinyl. But music is music, and music has requirements for sounding good. That has nothing to do with how loud a close range double fortissimo blast of every instrument in an orchestra all at once is. That doesn't reflect any sort of real world situation. Recordings don't need or use the dynamic range they already have. Overkill is overkill. More isn't better. Too loud and too quiet is uncomfortable to listen to in your living room. Get a dynamic expander and try expanding the peaks and see how you like it. Classical music sounds perfect with a dynamic range of around 50dB.

 

[Husky]

(...)the other day according to you, vinyls were preferred to CD, now you make your argument on how 50dB isn't enough dynamic. no matter the side of the argument, I spot like a small flaw. ^_^ (...)

There is no flaw at all. Vinyl can handle 70dB. The argument is: if there is a CD with a lot of distortion from hypercompression and there is a LP based on a master with much less of it, then the LP is preferred over the CD, despite the fact that the CD format has superior sound quality. With other words, the higher noise floor on the LP isn't as harmful to the listening as hypercompression on a CD is (and believe me, I don't like crackle).

The technical possibilities for great sound quality are better than ever, but the music industry managed to create sound quality worse than ever (OK, that's provocative and exaggerated).
Let me add, that bad sound quality isn't something new at all (and that the better resolution of HD formats doesn't make any audible difference to the CD format at playback in my view).

(...) when making an album you obviously have to account for how loud people will play it back in general. same for the listening conditions that will not necessarily be a quiet room. (...) I'm not for albums stuck within 3dB, but it's unrealistic to argue for albums to have a certain dynamic based on how loud and dynamic some concerts can be as those are not the expected playback conditions.

I don't deny that. And I agree that this isn't the reason for hypercompression.
On the other side, I think it's a pity if playback conditions like in the car or public transport with cheap headphones are targeted. I would prefer to let the consumer decide.

(...) as for global compression specifically for the idea of loudness war, if over compressed albums were making a systematic commercial flop, we could come to the consensus that it's just plain bad and we need it gone. but whatever the reason why, this is not what happens. (...)

There is a simple reason: the consumer doesn't have the choice and/or information. So we don't know if over compressed (and clipping) albums sell in spite of it. Because there is no equivalently available (and promoted) alternative version.

To be clearer: assumed there is a new album and I like the music (defined as I described earlier, i.e. the musical content in terms of melody/singer/arrangement etc., irrespective of the level of compression/limiting). Then I have the alternative not to spend money on it (doesn't mean not to listen to it). So I think many people spend the money in spite of the sound quality. And many don't know if and how it could sound better. Another alternative could be to look for a similar artist with similar music. But first, similar is not the same. Second, as hypercompression is so common, it's nearly impossible to find an alternative artist with better sound quality.

It would be great if the industry let the consumer decide.
As a starting point, the consumer could be offered even a rough mix before hypercompression was applied. To further minimize costs, It could be offered online (for download). And for a higher price than the normal version. I'd be willing to pay and I'm surely not the only one. I think the digital economy could give some good opportunity for such new sales channels, better matching consumer preferences and possibly increasing profits at the end.

 

[Husky]

(...)Classical music sounds perfect with a dynamic range of around 50dB.

Make a test. Take a good, dynamic recording from CD (normalized to 0dBFS), add noise peaking at -50dBFS (I think this is very conservative), listen to it in comparison with the original. The noise will be very noticeable in pianissimo (and distracting compared to the original) - normal hearing assumed. You may as well test 8bit PCM (48dB dynamic range). But maybe we're talking at crossed purposes.

 

[Glmoneydawg]

Just for the record, that's a George Carlin quote... Credit where credit is due.

RIP George.

Might be one of only a few quotes from Mr Carlin clean enough for this forum...truly funny guy.