Why are masters so different?
Oct 23, 2016 at 11:33 AM Post #76 of 132
  Let me ask this again: how do you think your stuff would sound if somebody took that recording and ran it through the same brick wall processing a second time?
 
A. Better
B. Worse
C. No change

 
I don't get the point of the question. If it were something I had mastered, I would expect the answer to generally be "B" or possibly "C" under certain circumstances. If the answer were "A", then I wouldn't have done my job properly in the first place and if the answer were "C" then I might not have done my job as well as I could have, depending on the requirements of the client. I could say this about almost any aspect of the piece though, any of the processing, compositional elements, arrangement or orchestration, which is why I don't get the point of the question?
 
G
 
Oct 23, 2016 at 11:33 AM Post #77 of 132
Please provide the studies and statistical evidence that supports this. YouTube subscribers is not a valid metric because it doesn't address lurking variables.


Do I really have to convince you people usually don't run around with 500$ (even the ones who obviously had the money for it) headphones and they find the bundled ones good already?
tongue_smile.gif

 
Anyways, this is way off topic. I know I don't mind little dynamic range and even clipping is ok (*gasp*) in some cases and I know a lot of people are like me in this regard (and a lot of people aren't like me, obviously).
 
Actually, you could try and see yourself how brickwalling changes the sound if it wasn't "clean sounding" (for a lack of better term) to begin with:
 
https://drive.google.com/open?id=0B1RKepESfvrWajBVZENyeDYyc3c       (brickwalled and volume matched)
https://drive.google.com/open?id=0B1RKepESfvrWVnEtNVhLdWJTQWM   (limiter removed and gain adjusted accordingly)
 
 

One of those is supposed to be horrendous compared to the other however if you listen to them you will find that they sound pretty much the same. I'm not even sure if I could tell them apart.
 
Oct 23, 2016 at 11:39 AM Post #78 of 132
Do I really have to convince you people usually don't run around with 500$ (even the ones who obviously had the money for it) headphones and they find the bundled ones good already? :tongue_smile:

Anyways, this is way off topic. I know I don't mind little dynamic range and even clipping is ok (*gasp*) in some cases and I know a lot of people are like me in this regard (and a lot of people aren't like me, obviously).

Actually, you could try and see yourself how brickwalling changes the sound if it wasn't "clean sounding" (for a lack of better term) to begin with:

https://drive.google.com/open?id=0B1RKepESfvrWajBVZENyeDYyc3c       (brickwalled and volume matched)
https://drive.google.com/open?id=0B1RKepESfvrWVnEtNVhLdWJTQWM   (limiter removed and gain adjusted accordingly)


One of those is supposed to be horrendous compared to the other however if you listen to them you will find that they sound pretty much the same. I'm not even sure if I could tell them apart.


This does not mean they prefer one over the other, simply that they don't care. This is not a definitive answer.
 
Oct 23, 2016 at 11:52 AM Post #80 of 132
   
I don't get the point of the question. If it were something I had mastered, I would expect the answer to generally be "B" or possibly "C" under certain circumstances. If the answer were "A", then I wouldn't have done my job properly in the first place and if the answer were "C" then I might not have done my job as well as I could have, depending on the requirements of the client. I could say this about almost any aspect of the piece though, any of the processing, compositional elements, arrangement or orchestration, which is why I don't get the point of the question?
 
G

The point is: If your mix gets airplay, that perfectly brick-walled final mix you did will be re-processed through an aggressive processor with it's own version of final peak control.  If mastering did less, the damage to the final on-air product would be confined to the final broadcast processor. As it is, you'll essentially run your stuff through the brick wall twice when it's aired. 
 
Tell that to your clients.  Or invest in a broadcast processor and check your mastering through it.  
 
Oct 23, 2016 at 12:43 PM Post #82 of 132
  Tell that to your clients.

 
What, you think I don't know that or tell that to my clients? What a low opinion you have of me, without even knowing me! Notice I said that under certain circumstances the goal would be "C", this would be one of those circumstances. Commonly a separate master is made for radio play, partly due to radio broadcast processing and partly due to timing, especially with EDM.
 
G
 
Oct 23, 2016 at 1:38 PM Post #83 of 132
   
What, you think I don't know that or tell that to my clients? What a low opinion you have of me, without even knowing me! Notice I said that under certain circumstances the goal would be "C", this would be one of those circumstances. Commonly a separate master is made for radio play, partly due to radio broadcast processing and partly due to timing, especially with EDM.
 
G

If you're creating a less-processed version for broadcast, that's great (I suspect not typical of the industry).
 
Then you're aware of the different processing requirements of FM vs HD Radio.  And you're aware that HD Radio hasn't penetrated the market significantly.  So, you'd be mastering for FM.
 
What FM broadcast processor(s) do you monitor through?  Presets? And they'd be feeding a low-power FM transmitter, then a receiver, then your monitor system.  Any specific gear you use to make that custom broadcast mix?
 
Oct 24, 2016 at 10:17 AM Post #84 of 132
  What FM broadcast processor(s) do you monitor through?
 
If you're creating a less-processed version for broadcast, that's great (I suspect not typical of the industry).

 
I don't have an FM broadcast monitor, as I don't master radio edits, I refer clients who need a radio edit. My b-chain includes Dolby processors, as my main work is TV/film, although my past is in music recording and production. I'm a rather reluctant music mastering engineer and don't actively seek/advertise for mastering clients.
 
Certainly some of my recordings have had a specific radio broadcast master created, mostly the major label stuff. Not so many clients are bothered about radio edits these days though, they're more interested in what it'll sound like on Youtube or iTunes. Just to be clear, because I find your use of the term confusing, when you say "less processed", you don't actually mean less processed, just less final limited/compressed?
 
G
 
Oct 24, 2016 at 12:31 PM Post #85 of 132
   
I don't have an FM broadcast monitor, as I don't master radio edits, I refer clients who need a radio edit. My b-chain includes Dolby processors, as my main work is TV/film, although my past is in music recording and production. I'm a rather reluctant music mastering engineer and don't actively seek/advertise for mastering clients.
 
Certainly some of my recordings have had a specific radio broadcast master created, mostly the major label stuff. Not so many clients are bothered about radio edits these days though, they're more interested in what it'll sound like on Youtube or iTunes. Just to be clear, because I find your use of the term confusing, when you say "less processed", you don't actually mean less processed, just less final limited/compressed?
 
G

Thanks for your frank and honest reply.  I appreciate it.
 
The concept of a special radio master necessitated by the additional on-air processing is a new one to me.  I checked with a friend who is chief engineer of a major group of stations here, he said all their music comes from a corporate server, but has not been able to confirm there are any specially processed broadcast versions.  Still working on confirmation of that.  Not doubting what you say, I'm trying to determine if the practice is sporadic, common, or industry standard.  
 
Because I'm a broadcast guy I lump all forms of processing together.  Each individual processing element contributes something.  If you take one part of the chain away it affects the total.  I acknowledge that the final limiter is the biggest offender, but distortion is also a function of the total release time of all gain-variable elements individual release times added together, so when I say "less processed", I am not specifically meaning the final limiter, but other aggressive stages too.   BTW, clipping is a completely valid form of processing, and can even be done inaudibly...with great care.  It's part of "processing" too. 
 
I would welcome the opportunity to at least audition such a radio broadcast master.  That's why I asked my CE friend, if he can locate a definitive radio version, I'll be able to compare it to the general release version.  It would be great to have several samples, because one tune won't tell much about industry practice.  It's likely a futile pursuit anyway, but I'll at least ask for info.  
 
I suspect I'd find that, as you say, broadcast masters are rare.  Other delivery systems like YouTube, downloads, even HD Radio are so completely different from broadcast in how they handle maximum signals.  Analog broadcast has pre-emphasis at the transmitter and de-emphasis at the receiver (both AM and FM), and that's not a small thing to deal with.  The FM curve has a difference in gain of 17.5dB at 15kHz relative to 500Hz, defined by the time constant of 75uS, but with rapid roll-off in the 15kHz to 17kHz area to protect the pilot (analog TV sound is essentially the same).  AM uses the same time constant, but more band-limited.  An FM receiver provides exactly complimentary de-emphasis, but an AM receiver's de-emphasis is a bit wild and uncontrolled, very receiver dependant.  The AM curve was worked out with a survey of average receivers balanced against occupied RF bandwidth.  Doubtful anybody worries about what music sounds like on AM anymore, I just mention it so readers might see there's a difference.
 
By contrast, all digital distribution channels, streaming, download, disc, digital TV and HD Radio are essentially flat response, full output level across the entire pass band.  That makes FM VERY different from anything else, and the job of mastering for broadcast laudable, but challenging at best.  For that reason I suspect very few broadcast versions are created.  
 
And that means, when we hear brick-walled EDM on the air, we've got two rather aggressive processing chains in series...at least, likely, most of the time.  
 
Just so you understand where I'm coming from, not blowing any horns...I'm an old audio/broadcast guy, career spanning more than 45 years, much of it dedicated to audio and audio processing, including processor design and development, and adjusting on-air processing...years of time spent listening to all sorts of dynamics processing.  Oddly, within the last 10 years, my musical taste did expand to include EDM!  I'm shocked too, but hey, some of that stuff is pretty good and well made.  I was able to engage in an email exchange with one of my favorite EDM artists a few years ago regarding releasing his stuff in other formats, like high-def audio and 5.1 surround.  His stuff had recently become pretty badly brick-walled, and frankly the result drove me away from listening to it.  We bantered about higher quality releases, I tried to define what that meant.  He eventually took part of my suggestion and released a few things in "High Quality Audio" (his term), 24/96 to anyone else.  What came out was just as same brick-walled mess he'd released in standard form, probably just his masters up-sampled.  His stuff is great music, innovative, melodic, etc., but so distorted, with such a low crest factor I find those newer releases unlistenable, and favor his earlier, less processed work much more listenable.  Disappointing, but that's how it goes.  Nothing else I listen to has that kind of processing on it.  I hear myself saying "These kids today...", but...well, I'm old. 
 
Oct 24, 2016 at 9:52 PM Post #86 of 132
  ...
 
Compression does remove a lot of emotional impact, however.  From a standpoint of impact and listener involvement, it never does any good, in any genre.  
 
...

 
Lol, I call BS here. It's an eargasm running some vocal stems through an LA-2A opto-compressor, or a drum track via a Fairchild. Likewise running the stereo out on an SSL-G master buss. Without these tools I would venture that any track would have no emotional impact whatsoever. These devices define modern music and it's rather silly to make that statement above.
 
Oct 25, 2016 at 12:06 AM Post #87 of 132
   
Lol, I call BS here. It's an eargasm running some vocal stems through an LA-2A opto-compressor, or a drum track via a Fairchild. Likewise running the stereo out on an SSL-G master buss. Without these tools I would venture that any track would have no emotional impact whatsoever. These devices define modern music and it's rather silly to make that statement above.

If you've been reading the thread...
 
Per-track compression: generally good
Overall compression, brick-wall processing...well, that's what we've been debating.   
 
An LA-2 is perhaps one of the nicest, kindest compressors ever made. Lovely device.  Even the plug-in simulations are good.  And I'm not griping about a Fairchild on a drum track (nice that you even have one!).  An 1176, as a peak responding FET box, generally in my tool belt as well. 
 
Just not at all what I'm complaining about...but whatever...
 
Oct 25, 2016 at 3:30 AM Post #88 of 132
  Per-track compression: generally good
Overall compression, brick-wall processing...well, that's what we've been debating.

 
That's maybe been part of the problem in our discussion. You said "compression" and the post to which you originally responded said "compression", but you've only been debating in terms of just one aspect of compression, the brick-wall limiting/compression at the final/near final step of mastering. There appears to be a mis-conception of a dynamic mix being created in production, which is then crushed to death during mastering by smashing a limiter as hard as possible. I'm not saying that never happens but in many cases extreme compression has already been applied at the per track and/or sub-group level, usually considerably more than is added in mastering! This is very genre dependent however, classical music for example usually has no compression before the mastering stage, EDM on the other hand, usually tons, while older rock genres probably no more than moderate compression and so on. Many/Most of the genres/sub-genres which evolved in the '90's and later employ and depend on heavy compression (in the production phase). I've mastered EDM tracks where no more than a couple or so dB of compression can be added (during mastering) before unacceptable audible distortion occurs because so much compression has already been applied at various stages of production and that amount of compression during production is necessary to create the feel which defines that genre/sub-genre! It's not at all uncommon (not just with EDM but most other modern and even some sub-genres originating in the '80's) to find a number of individual tracks which have 10-16dB of compression applied, then another 6dB or so at the sub-group level and even the vocals, often the most acoustic sounding element in the mix, having 8dB or so of compression. I'm not saying that compression during mastering is not an issue, I'm saying that it's only part of the picture. The loudness war is not just about inappropriate/excessive mastering compression, it's about the very nature of the modern pop genres themselves!
 
RE radio broadcast. Certainly up to the mid '90's or so, a radio edit was pretty routine/standard practice, I think that's very uncommon today. In fact, during content creation there's very little, if any, consideration of analogue broadcast these days and not just analogue radio but TV and film as well. Certainly pretty much all TV audio deliverables are spec'ed for digital and if any analogue broadcast occurs, it's up to the broadcast processors to deal with. We, as content creators, are no longer concerned with analogue broadcast. The only real monitoring concern we have these days when creating TV content is the fold-down from 5.1 to stereo and the Dolby limiters built into everything.
 
G
 
Oct 25, 2016 at 12:30 PM Post #89 of 132
   
That's maybe been part of the problem in our discussion. You said "compression" and the post to which you originally responded said "compression", but you've only been debating in terms of just one aspect of compression, the brick-wall limiting/compression at the final/near final step of mastering. There appears to be a mis-conception of a dynamic mix being created in production, which is then crushed to death during mastering by smashing a limiter as hard as possible. I'm not saying that never happens but in many cases extreme compression has already been applied at the per track and/or sub-group level, usually considerably more than is added in mastering! This is very genre dependent however, classical music for example usually has no compression before the mastering stage, EDM on the other hand, usually tons, while older rock genres probably no more than moderate compression and so on. Many/Most of the genres/sub-genres which evolved in the '90's and later employ and depend on heavy compression (in the production phase). I've mastered EDM tracks where no more than a couple or so dB of compression can be added (during mastering) before unacceptable audible distortion occurs because so much compression has already been applied at various stages of production and that amount of compression during production is necessary to create the feel which defines that genre/sub-genre! It's not at all uncommon (not just with EDM but most other modern and even some sub-genres originating in the '80's) to find a number of individual tracks which have 10-16dB of compression applied, then another 6dB or so at the sub-group level and even the vocals, often the most acoustic sounding element in the mix, having 8dB or so of compression. I'm not saying that compression during mastering is not an issue, I'm saying that it's only part of the picture. The loudness war is not just about inappropriate/excessive mastering compression, it's about the very nature of the modern pop genres themselves!
 
RE radio broadcast. Certainly up to the mid '90's or so, a radio edit was pretty routine/standard practice, I think that's very uncommon today. In fact, during content creation there's very little, if any, consideration of analogue broadcast these days and not just analogue radio but TV and film as well. Certainly pretty much all TV audio deliverables are spec'ed for digital and if any analogue broadcast occurs, it's up to the broadcast processors to deal with. We, as content creators, are no longer concerned with analogue broadcast. The only real monitoring concern we have these days when creating TV content is the fold-down from 5.1 to stereo and the Dolby limiters built into everything.
 
G

Regarding paragraph 1: My fault for not discerning that the thread was full of people in the recording business.  My comments were targeted at the listener.  I'm not only talking about brick-walling, but the cumulative effects of all of it.  It's additive.  Yes, track/instrument compression can be radical and not nearly so apparent in the mix, as is appropriate.  And yes, it's all genre related.  And the end result is all that matters to the listener.  When the mix has been processed for loudness and tightly peak-controlled, there is very little peak energy left (no headroom for it anyway).  This will be overly simplistic and probably shot at yet again, but...peak energy provides impact and excitement.  Lack of it greats a flat wall of dynamics with no variation.  No variation is constant and unchanging stimulus, and that translates to lack of excitement and emotion.  My opinion, again, for decades of doing it.  But thanks for the explanation.
 
Regarding paragraph 2: As to no consideration for analog TV...well, in the USA there is almost none anyway, so that's appropriate.  And worrying about 5.1 fold-down would be the biggie apart from levels.  No analog film to speak of either, so we're good there.  But HD Radio has not made significant inroads at all.  The receiver base is tiny, almost insignificant.  Music on radio is still nearly all analog FM, and that has, for the last 30 years, been a heavily processed media outlet.  And while we'd like to think that traditional FM radio has collapsed into a non-factor relative to on-line media, statistics don't support that at all. There is still a lot of music listened to on FM.   So if material is expected to be broadcast, and for any genre of significant popularity it should be, then understanding and reacting to what is being done on-air on top of what's being done in mastering would definitely result in superior sound delivered to the listener.  That's not happening at all.  The result is easy to understand: the general release master is over-baked, the FM version is then burnt to a crisp.  But listeners have no options, other than to take it or leave it, so they generally take it, skewing the understanding of what they would actually prefer.  And now we have a generation of listeners raised on low-rate .mp3 and smashed within a micron of it's life EDM.  Just impresses me as something wrong there.
 
It was my impression that the old "radio edit" was a musical edit, often shorter, of the main release.  The one's I'm aware of are no different in final processing, which makes sense because of the distribution channel of label to station in the 1990s.  In my own checking on the radio-master question with a large radio group, so far I find zero evidence that it's being done in current music.  So, my comments regarding double-brick-wall processing stand.  The only opportunity to change the situation is upstream in production, and from what I can tell, there's no interest in doing that.
 
Oct 25, 2016 at 12:39 PM Post #90 of 132
   
Lol, I call BS here.

 
I'm not an expert on music theory... but I was under the impression that classical composers used (what we call) dynamic range for this very purpose... to create a sense of emotion or drama. I don't know what said composers actually call this, but the idea is the same. 
 
I have no horse in this race, so feel free to ignore this particular opinion. 
 

Users who are viewing this thread

Back
Top