Turned out that radio stations hated that massively squashed stuff from the '90s and '00s because it did not leave them any scope to apply their own radio-specific compression.
What causes listening "fatigue"?
- Thread starter critofur
- Start date
Turned out that radio stations hated that massively squashed stuff from the '90s and '00s because it did not leave them any scope to apply their own radio-specific compression.
This kind of compression scheme has been standard for dance oriented pop production since 80's, but I agree that he is the one who sophisticated and utilized it for mainstream pop rock music. Since dance oriented music has become mainstream, mainstream engineers simply followed that trend. This trend is also related to Loudness War, indeed, but it's not his fault. This is not his invention.
I have never had a chance to hear his premastered mix, but it seems that his mix is not too much compressed, very average, in my impression. I have his Waves Plugin suites, and their presets look pretty aggressive at first, but I find they are very sane. Most of the mixing and mastering engineers including CLA are against loudness war, because it simply sounds bad.
Who really has pushed the loudness war is mostly ignorant record companies A&R and the artists who want to make their music sounds more competitive on the radio.
Let me try a bit of "arm-chair" perceptual guesswork. (OK, maybe a bit more of that considering my recent speculations.)
With music compressed and played loud, you have two vectors for fatigue. First, there is the loudness per se. Second, there is something that might be called "density". In classical music, there is an ebb and flow of almost all the music elements, with quiet parts between the loud parts. With the awful pulsating dance music, maybe no rest for your ears at any point before the blessed end.
B.
With music compressed and played loud, you have two vectors for fatigue. First, there is the loudness per se. Second, there is something that might be called "density". In classical music, there is an ebb and flow of almost all the music elements, with quiet parts between the loud parts. With the awful pulsating dance music, maybe no rest for your ears at any point before the blessed end.
B.
Green Day, American Idiot, mixed by CLA, dynamic range average 5dB max 6dB min 4dB.
@bentoronto you're spot on there.
@bentoronto you're spot on there.
I've been through a few of CLA mixes for other bands/artists and they are all pretty similar to this while the mastering engineer Ted Jensen around the same time did stuff with much greater dynamics. Ted also did the Eagles Hotel California, he's been at it for a while.
PS from wiki:
In 2008, loud mastering practices received mainstream media attention with the release of Metallica's Death Magnetic album. The CD version of the album has a high average loudness that pushes peaks beyond the point of digital clipping, causing distortion. This was reported by customers and music industry professionals, and covered in multiple international publications, including Rolling Stone,[11] The Wall Street Journal,[12] BBC Radio,[13] Wired,[14] and The Guardian.[15] Ted Jensen, a mastering engineer involved in the Death Magnetic recordings, criticized the approach employed during the production process.
PS from wiki:
In 2008, loud mastering practices received mainstream media attention with the release of Metallica's Death Magnetic album. The CD version of the album has a high average loudness that pushes peaks beyond the point of digital clipping, causing distortion. This was reported by customers and music industry professionals, and covered in multiple international publications, including Rolling Stone,[11] The Wall Street Journal,[12] BBC Radio,[13] Wired,[14] and The Guardian.[15] Ted Jensen, a mastering engineer involved in the Death Magnetic recordings, criticized the approach employed during the production process.
Last edited:
I agree about density, and the comparison with classical music illustrates an important factor in musical expression; the use of dynamics as an expressive aspect.
Mozart stated that the silences are as important as the sounds.
I checked up the record I mentioned earlier that was produced to give more of a live sound and it was very high. Funny thing is that they made an edited version of one of the songs and it is completely different, sounds much more "radio friendly", it would be interesting to see how that one stacks up.
I have worked with Ted Jensen a few times. A trustful mastering guy, very professional.
I'm curious why I do feel CLA mix is not over compressed, so I checked a few records here. Many of them are too much compressed obviously, but even so, they sound absolutely fine for me. DR is about 6-9dB, which is normal for pop music today.
Anyway, we can check DR chart at loudness-war.info. For example, Bruce Springsteen's 2012 album, Wrecking Ball, the dynamic range of CLA track is the same as the other mixing engineers on the same album. Also you can look at the vinyl version's DR of the same album, they are around 11dB, which is pretty healthy, so we can see that over compression was obviously done at digital mastering stage. We can't achieve 5dB dynamic range with analog equipments anyway, and CLA does not use digital compression at his 2 bus as far as I know. I think he uses Focusrite Red comp.
Bruce Springsteen 11dB
Album details - Dynamic Range Database
American Idiot (Edition Studio Masters) 9-10dB
Album details - Dynamic Range Database
Last edited:
Kind of related thing just popped into my mind, does difference in distortion between a multiway speaker drivers affect fatigue? for example a steep lowpass on a woofer attenuates harmonics of said woofers upper register, then comes in the tweeter with possibly a bit too low crossover point having much more harmonics generated and audible at the lower register making the whole system distortion vary? Anybody have thoughts over this? are the distortion figures attenuated with the driver response since the distortion generating driver is after the electronic crossover making this non issue? 🙂 What if one of the drivers had worse distortion graph in general than the other driver(s)? do speaker designers ever consider this distortion "uniformnes" or is it just minimizing the system distortion in general? thanks.
ps anyone have checked if different kind on technology (woofers/tweeter/compression drivers/ribbons) have dirrerent distribution of 2nd and 3rd harmonics that could make a sudden jump in system distortion distribution near crossover?
ps anyone have checked if different kind on technology (woofers/tweeter/compression drivers/ribbons) have dirrerent distribution of 2nd and 3rd harmonics that could make a sudden jump in system distortion distribution near crossover?
Last edited:
The distortion products from any given driver are a failure of that driver at the driven frequency, and this produces artefacts at a higher frequency which a steep Xover will not attenuate, it is mainly mechanical in nature.
This applies particularly to metal cone midranges and woofers which make a hell of a racket above their pass band, these excited by the passband energy.
Regardless of the steepness of the attenuation due to a Xover, if the tone is in the middle of the pass band, and it is a poor driver, it will distort that tone; the steep Xover has no effect on it, but serves only to prevent the driver giving a deliberate O/P outside its good, and hence used range.
In the design process the aim is to get the best from each driver, curtailing the passbands to get that, and distortion products being one consideration in choosing the operating band. Running tweeters too low is a problem which often produces horrible distortion.
This applies particularly to metal cone midranges and woofers which make a hell of a racket above their pass band, these excited by the passband energy.
Regardless of the steepness of the attenuation due to a Xover, if the tone is in the middle of the pass band, and it is a poor driver, it will distort that tone; the steep Xover has no effect on it, but serves only to prevent the driver giving a deliberate O/P outside its good, and hence used range.
In the design process the aim is to get the best from each driver, curtailing the passbands to get that, and distortion products being one consideration in choosing the operating band. Running tweeters too low is a problem which often produces horrible distortion.
I hope that you guys realize that you are talking about two different things simultaneously. There is "fatigue" from the musical source and there is "fatigue" from the playback system. The two things are completely different and shouldn't be mixed together.
I only care about "fatigue" from the playback since I'll select the music source that I don't find fatiguing. Source choice is a personal matter, playback fatigue is a design issue.
Right.
Might be worth considering what shortcomings of sources can be ameliorated by what users can do in playback? "Expansion" comes to mind... however unlikely to help.*
The smart folks who re-process Caruso recordings must have clean-up tools for sources? And on the opposite hand, RCA vinyl records had distortion added to compensate for cartridge tracking error (yuck, bad idea) which is distorting the source to clean up the playback gear.
There is some overlap in the kinds of crap introduced by sources and by playback. So times when hard to discriminate. While records are produced with intentional sonic garbage added (anything with an electric guitar, eh), Rice-Kellogg drivers kindly add mostly musical harmonic distortion which is better digested.
B.
*Hello old timers, remember expanders using incandescent bulbs?
Might be worth considering what shortcomings of sources can be ameliorated by what users can do in playback? "Expansion" comes to mind... however unlikely to help.*
The smart folks who re-process Caruso recordings must have clean-up tools for sources? And on the opposite hand, RCA vinyl records had distortion added to compensate for cartridge tracking error (yuck, bad idea) which is distorting the source to clean up the playback gear.
There is some overlap in the kinds of crap introduced by sources and by playback. So times when hard to discriminate. While records are produced with intentional sonic garbage added (anything with an electric guitar, eh), Rice-Kellogg drivers kindly add mostly musical harmonic distortion which is better digested.
B.
*Hello old timers, remember expanders using incandescent bulbs?
Last edited:
Yea, not to me there isn't. The art is the art.
"Cleaning up" a recording back to its more "original state" is a fine thing to do.
Interesting. I have heard a lot of speakers, my own included with metal drivers.... and I dont find them tiring.
So maybe theres a limit, to where you cant hear such distortion you mention?
Hi tmuikku, my take is that the lower range of a driver's passband is pretty much always where the weakness lies.
I mean, it's where excursion is maximum, it's where BL is losing linearity.
So, like you say, asking a tweeter to dig too low will suck.
But this is where a steep HPF xover to the tweeter can be a benefit.
You can take the tweeter lower when steeper, with less excursion worries, than when trying the same xover freq with a shallower xover..
Only problem then is the higher-order xover that is needed for steep....excess phase wrap, screwed up group delay, ....iow and simply said, blown timing.
And tis why linear phase xovers rule, almightily so, amen so be it, forever and ever 😉 😀
The resonances in the metal cones were pointed out to me in discussion with a professional designer, whilst we were contemplating a design using the Heil.
We googled, and I think it was Scanspeak anodised cones which showed severe breakup in the mid, above the pass band.