another article for comment [Archive] - SAWStudio / SAC User Group

Mulder

07-10-2005, 05:02 AM

If you didn't like that last one, you probably also will not enjoy...

http://www.cprec.com/html/loud_vs__punchy.html

Same rules as before, guys:
Any and all comments are welcome, especially those that help me improve the work. If there's a hole in it, I WANT to know, so don't be shy (as if we were shrinking violets here :D

I can think of some holes..

98% of the householdloudspeakers music is beeing played on is bogus, these units will distort at the slightest dynamic peak. Those little units with 'subwoofers' (eugh) that are designed to sit near computermonitors often produce more distortion than original sound (THDfigures over 100% ;) ) so masteringhouses tend to take that in consideration when -for instance- mastering pedopop, children often don't own -thank thou lord- sound installations that can reproduce their music in dynamic detail.

I'm mixing metal, and learned from my FOHwork that a human ear can produce pretty much distortion on its own when soundpressure reaches a very high SPL. So 'distortion' is a funtion of loudness for the human brain, so I'm trying to have the guitartracks in the recording sound the loudest in the mix without sacrificing the other instruments.
It's pretty difficult to get a proper balance from this pov, but it gives indeed a very loud impression. Problem with that is that a good distorted guitar (mesa triple rect) doesn't have that much dynamics itself, beeing fully saturated. Its the job of the drums to put the dynamics back in, but only so far it just puts it's head above the guitars, same for the (screamed) vocals. Especially the latter gives a good effect, if it sounds like the voice is screaming it's rectum open to get above the wall of guitarsound the total impression is one of 'full throttle'. After that I really don't put much limiting or compression on the total mix, maybe just a bit for some peaks & stuff. But for me to get a loud impression means to have the distorted instrument prominent in the mix since your ears tell you that distortion is the same as high SPL.

Also a funny trick to do is to mount 'shakers' in your seat, those contactwoofers. Thing is, you also listen with your guts. If a sound is loud enough it will rattle your colon and that also sends a message to your brain (and some other parts nearby I guess). Adding a shaking sub to the back of your seat will give you the impression of loudness while mixing on cans, very fun indeed.

Cary B. Cornett

07-11-2005, 08:40 AM

You made some interesting comments:

98% of the householdloudspeakers music is beeing played on is bogus, these units will distort at the slightest dynamic peak.

That has been true for one reason or another for as long as I can remember. Even so, I can remember listening to and enjoying many records as a child growing up that I now know were made without the benefit of the powerul mastering processes that we have available today. IMO, the present trends in mastering were not driven by the needs of average listeners, but by the demands of record company and broadcast execs (demands that were just as bogus as the speakers most people hear music on).

... masteringhouses tend to take that in consideration when -for instance- mastering pedopop, children often don't own -thank thou lord- sound installations that can reproduce their music in dynamic detail.

There is no doubt that there is benefit that can be derived from the advanced processing tools that we now have. My concern is that these tools are often misused to the point that much of the artistic value and long-term listenability of the music is destroyed. It is desirable to mix and master music in a way that will allow it to still be enjoyed with inferior systems, but this can be accomplished without removing all of the dynamic subtlety and impact from the music.

I'm mixing metal, and learned from my FOHwork that a human ear can produce pretty much distortion on its own when soundpressure reaches a very high SPL. So 'distortion' is a funtion of loudness for the human brain,

This is a good and useful point.

... I'm trying to have the guitartracks in the recording sound the loudest in the mix without sacrificing the other instruments.
But for me to get a loud impression means to have the distorted instrument prominent in the mix since your ears tell you that distortion is the same as high SPL.

"Prominent" is a matter of perception. A distorted sound naturally sounds more prominent than an undistorted sound at the same level, which makes it able to SOUND more prominent without having to use up as much headroom in the mix as "rounder" less distorted sounds would need to sound equally prominent. (A fact I mentioned in another article.)

Thanks for your comments. I think some of what you called my attention to will be useful in a future article.

mikebuzz

07-11-2005, 09:57 AM

Nice article Cory however this has been beaten to death everywhere I have been PWS etc. I don't see a solution to the problem in fact the HOTTEST mastering guys right now are the ones that crush it the most.

I do like the way it is explained very understandable .

LAter
Buzz

PS: Mulder you should have your hearing tested years of FOH for Metal at the levels you have expirienced can cause hearing loss ( my Dad used to work for Lockheed , back then they did'nt know Jets would make you go deaf he lost most of his hearing in one ear)

:eek:

Bert

07-11-2005, 10:05 AM

Cary B. Cornett

07-11-2005, 10:28 AM

Nice article Cory however this has been beaten to death everywhere I have been PWS etc.

I didn't think that I was blazing any new trails with the article, as I have seen many comments on this issue in the past. I am operating on the assumption that a lot of new/budding DIY artist/engineer/producers have not yet been exposed to these concepts, and one of the purposes of my website is to provide as much clearly understandable help for this rapidly growing group of people as I can.

I don't see a solution to the problem in fact the HOTTEST mastering guys right now are the ones that crush it the most.

The "hottest" mastering guys right now are also, for the most part, the most experienced. They got that way by staying in business for a long time. They stayed in business for a long time by responding to the demands of their clients. The top mastering guys are not, then, the problem (they were among the first to complain about it!), so it is certainly not my intent to attack them. IMO, the place to attack the problem is at where the demand is coming from. Now, I can't change the thinking of the "big label" guys, but they do not buy the majority of the mastering time. The more "new guys" can be informed early on of the problem, the easier it will become to provide market pressure to solve the problem.

I am also writing for the benefit of many who, for whatever reason, will decide to do their own mastering. If they can be helped to understand the consequences of advanced audio butchery, they may be encouraged to do better work (by not OVERDOING the processing). I don't expect to change the world, but I do hope to positively influence it just a teensy bit!

I do like the way it is explained very understandable . Thank you. Understandability is one of my top goals.

Cary B. Cornett

07-11-2005, 10:33 AM

Since I am not a native English speaker I'm always curious how understandable things are for me. I really like the language you use to describe things - fun to read.

(I have the "Mixing Engineers Handbook" from Bobby Oswinsky which seems to be the most popular mixing book at amazon. While I find it OK I don't think it's better than your articles).

The wider the variety of people who can understand what I write, the happier I am about it. I also like things to be fun, because fun stuff is easier (and more pleasant) to remember.

Your favorable comparison of my efforts to some very successful published work is high praise from my POV.

Thank you.

AudioAstronomer

07-11-2005, 11:00 AM

Good article, but the subject has been beaten to death. The language was a little simple for my tastes at times, I guess at times it traded simplicity for accuracy. A few instances of hyperbole that made me cringe a wee bit, but overall I guess you cant really do THAT much on such a subject that's been (forgive the pun), smashed to death.

mikebuzz

07-11-2005, 11:05 AM

Cory I was'nt attacking the mastering guys I realize thier situation and feel for them , most of them hate to do what they have to ( from what I have read , I don't know any personnally so it's just what I gleam from the net and articles like yours )

I find myself now Crushing things so they compare to what is out there I really like BIG dynamics but when A/B ing it becomes really obvious . They CRUSH the SHIITE out of this stuff !!!!

I really like bobs levelizer because it keeps a bit of dynamics in the track but helps it get THERE ( wherever that is ??? )

Keep up the good work your explainations are just what is needed for people learning this craft.

Later
Buzz

PS: I listened to a song in Saw the other day from a commercially released album and there was 3db of dynamics ?????? MAN thats crushed :eek:

PSS: thanks for the feedback on that mix the other day VERY useful info :D

Cary B. Cornett

07-13-2005, 07:02 AM

Good article, but the subject has been beaten to death.

That is certainly true within the experienced professional community. I am operating on the thesis that by far the majority of "new" engineers neither can afford nor find opportunity for proper professional training, and that informing this new and LARGE crowd as soon as possible on this issue can help lessen the lemming-like rush to "crush". The more we can get this issue out there, however seemingly fundamental or overused, the better our chances, IMO, of stemming the tide of SHRED.

The language was a little simple for my tastes at times, I guess at times it traded simplicity for accuracy.

That was not a tradeoff I intended to make, so I am interested in specifics to help me improve the work. Please elucidate.

A few instances of hyperbole that made me cringe a wee bit, but overall I guess you cant really do THAT much on such a subject that's been (forgive the pun), smashed to death.

By "hyperbole" do we mean "extreme of emphasis" or "compromise of accuracy in favor of impact"? If I become aware of something that could be improved (within my particular goals for the article), I definitely want to get it right. Your pun is, sadly, very much apropo.

Naturally Digital

07-13-2005, 07:56 AM

Hi Cary,

That's a fun one. I think your coverage of the history of it all makes for a good review.

I must admit that the title had me waiting for more tips on achieving "punch" and if I were writing the article, I might title it "What is Loud?" or something similar. Just my two cents on that one. ;)

A few minor points...

I would re-phrase "Eventually the term “Muzak” became something not very nice." to "Eventually the term Muzak inherited a negative connotation." or similar.

Also, "There began to be companies that manufactured better and more sophisticated loudness weapons for radio stations to use." may flow better if phrased: "Companies began to manufacture...".

I suggest breaking this up into two sentances: "Now every mastering engineer had the exact SAME peak limit beyond which they could NOT go, because otherwise they would run out of numbers, and smashing into that limit would sound like “SPLAT!” (it sounded like something else, too...)."

Cary, this paragraph "Now, I will grant you that the Radio Guys do not understand this principle, so naturally they expect the recordings they get to be HOT. It is probably a good idea to give these guys special CD’s that are mastered they way they demand, but if you want to have good, dynamic, PUNCHY recordings your radio promo discs should be the ONLY ones that are “aggressively “ mastered. You may have to cater to the ignorance of radio execs, but you do not have to force the same stupidity on your paying customers, the people who actually BUY your recordings." goes counter to the advice of Bob Katz in his book where he talks about 'the myth of radio ready'... I'm not an expert but you may want to get a second opinion on this. Perhaps Steve B. could chime in with his suggestion on this. Personally, I don't produce louder versions for the radio.

Finally, "Bob Katz of Digital Domain mastering goes to some length, both on his website and in his book “Mastering Audio”, to explain this idea." Might be: "Bob Katz of Digital Domain mastering covers this idea at great length, both on his website and in his book “Mastering Audio”. I dunno... just a 'flow' thing for me. :)
Hope this helps.

Leadfoot

07-13-2005, 08:41 AM

Pedro Itriago

07-13-2005, 08:53 AM

Cary, I have to confess something. I didn't go to your new article because I thought it wasn't yours but a link to another thread in who-knows-where about this week's tired subject.

But then David's post made me realize it was another piece you'd written, so I went right away to read it.

Let me tell you, to me you have a very unique way to describe things. All those things I already knew were looked in a way that gave me a new perspective of them and not necesarily in audio. Pieces & bits of info's that I had completely unrelated in my mind (maybe they degraded to that point in there) got re-connected and re-processed.

I really, really enjoyed that one. Thanks for sharing

Cary B. Cornett

07-13-2005, 10:36 AM

David,

Yes, your comments are helpful, and will probably result in some "tweaks" to the latest article.

I must admit that the title had me waiting for more tips on achieving "punch" and if I were writing the article, I might title it "What is Loud?" or something similar.

It is sometimes difficult to know how much to cover in one article, and when there should be a division into two or more articles. In fact, I am considering doing a follow-on article to be titled "Loud AND Punchy or Can We Have Both?" with the idea of providing some better definition, particularly of the term "punch", and perhaps offering some thoughts of preserving or even Creating it. I agree with you completely that these are important ideas worthy of coverage.

In fact, I already wrote Bob Katz to see if my understanding of the term "punch" was what he intended in using it. His initial answer implied that I am at least on the right track, if not quite "there" yet. In fact, his reply began with
"Well, ultimately this does boil down to the question of how to define something in words that can only be heard. It is a Chinese Puzzle!"

I may just go ahead and make my best "stab" at the definition, possibly subject to later correction when I find errors in that attempt. The part about how to preserve, better yet, to ACHIEVE punch in a mix is more difficult. Part of this is because I am not yet an accomplished master at the art of mixing (I may, perhaps, have achieved "journeyman" status ;), and in particular I am still refining my compression methods. I would hate, in my ignorance, to steer anyone down a completely wrong path in this regard. Of course, I may end up relying on the crutch of "later correction" there, too.

I am still ruminating on your stylistic suggestions, as they are worth consideration.

Your one major "catch" in terms of accuracy will be addressed in another post.

Thanks!

TotalSonic

07-13-2005, 10:50 AM

I disagree with this one. Radio stations already heavily peak limit and compress the heck out of the signal, if you give them a track that is squashed and normalized it actually has the reverse effect and will sound small and distorted. If anything, I would give the station a less limited and compressed version of a mix, that will translate to a more dynamic sound after all the processing they do to it. I know this for a fact cause I had a couple of my mixes played on a major station here in Chicago, and the first one I sent them was heavily peak limited and sounded great on everything I played it on, but did not come across too good on air. The second time I got it right by doing the opposite.

Tony

Tony makes a very good point here. Clipped wav forms especially tend to not play well with things like Optimods and various other broadcast processors. In general more dynamic recordings actually come out sounding better over radio stations (an easy way to hear this is to just compare the overall fidelity of an oldies or classic rock station to a contemporary top40, "urban" or "modern rock" station).

It's been posted this link here before but a must read on the subject is Frank Foti & Rober Orban's (the inventors of the Optimod) article on "What Hapens To My Recording When It's Played on the Radio" at
http://www.omniaaudio.com/tech/mastering.htm

btw - Cary - I think your article overall is very well done.

Best regards,
Steve Berson

Cary B. Cornett

07-13-2005, 10:50 AM

Tony,

You and David both addressed the same important point, and it is indeed and excellent catch by both of you.

I have long been painfully aware of the aggressiveness of processing in radio from my days as a broadcast engineer, although only recently have I gained a better understanding of how careful one needs to be about the effects of mastering on the subsequent radio processing. I had suspected trouble for some time, but did not fully understand it.

if you give them a track that is squashed and normalized it actually has the reverse effect and will sound small and distorted. If anything, I would give the station a less limited and compressed version of a mix, that will translate to a more dynamic sound after all the processing they do to it.

Just a couple of days ago I got into the exellently detailed explanation of exactly why/how this is and description of a truly definitive test that several engineers collaborated on as published in Bob Katz' book "Mastering Audio". Unfortunately, I had already posted my latest article, and before you and David called my attention to it had forgotten that I had made that error. I will definitely post a correction within the next day or so.

I know this for a fact cause I had a couple of my mixes played on a major station here in Chicago, and the first one I sent them was heavily peak limited and sounded great on everything I played it on, but did not come across too good on air. The second time I got it right by doing the opposite.

What you just described rather closely parallels part of the test I just mentioned. In fact, I wish that I could get a CD with both versions of your mix on it and them played "on air" or at least through the processing chain with all of the radio execs in my market made to sit down and listen to the result. In fact, I wish that a CD of the "radio processed" results could be made to play for as many budding engineers as possible to perhaps check their lemming-like rush into the sea of SHRED.

Thanks for your comments.

Cary B. Cornett

07-13-2005, 12:06 PM

In response to some thoughtful comments, I have made and posted revisions to the article
"Loud vs. Punchy" (http://www.chinesepuzzlerecording.com/html/loud_vs__punchy.html)

You are invited to see whether the issues raised have been properly addressed.

Thanks

Carl G.

07-13-2005, 12:42 PM

I disagree with this one. Radio stations already heavily peak limit and compress the heck out of the signal, if you give them a track that is squashed and normalized it actually has the reverse effect and will sound small and distorted. If anything, I would give the station a less limited and compressed version of a mix, that will translate to a more dynamic sound after all the processing they do to it. I know this for a fact cause I had a couple of my mixes played on a major station here in Chicago, and the first one I sent them was heavily peak limited and sounded great on everything I played it on, but did not come across too good on air. The second time I got it right by doing the opposite.

Tony
Tony... it totally depends upon how hot the station loaded your CD into their computer... or if playing CD direct, how hot the guy on the board is pushing their limiters. From station to station, of course, it also depends upon the way their multiband compressor is set up.... lots of other factors too.

But, like you, I do find that generally the less processed the better - unless you know you're dealing with a station that has little processing.

UpTilDawn

07-13-2005, 10:15 PM

So out of curiosity...
I've noticed that past cds I've given to my local public broadcasting (a mostly jazz station) that had very little compression or limiting sounded too soft in general (compared to the average commercial cd they aired) and I could really hear the station compressors clamp down hard on uncontrolled peaks in my program material.

Since then, I've begun to use the Levelizer's peak limiter on the master out at anywhere from 80-73%, plus the normalizer set to 98-99% (along with individual track compression, mostly to control peaks) and now don't appear to be fighting the station's setup.

Can somebody give me an idea what I'm probably dealing with at the station end since my experience appears to be oposite the advice I'm seeing here?

Thanks for your help,
DanT

Carl G.

07-13-2005, 10:58 PM

So out of curiosity...
I've noticed that past cds I've given to my local public broadcasting (a mostly jazz station) that had very little compression or limiting sounded too soft in general (compared to the average commercial cd they aired)DanT
That's why you'd probably want to use a little RMS compression on your material.

and I could really hear the station compressors clamp down hard on uncontrolled peaks in my program material.DanT
It wasn't probably so much that the station compressors clamped down on your uncontrolled peaks.... it's simply that the ratio of your peaks to average body of material is probably too great... pushing the body of your material way down. Gently taming your peaks with the Levelizer will set you on a higher average output. than normalizing you peaks.

Have a great time!
Carl

Leadfoot

07-14-2005, 04:27 AM

I think Carl is absolutely correct on that. I was gonna say the same thing. What I was talking about is somewhat of a different nature, for one, the music I'm referring to is commercial hard/pop rock that is by nature usually mixed/mastered as loud as possible to begin with. I don't think you'd have the same problem with a jazz mix that could potentially be rms maximized (more than what you did) and still have a lot of dynamic in it.

Tony

Cary B. Cornett

07-14-2005, 08:15 AM

I've noticed that past cds I've given to my local public broadcasting (a mostly jazz station) that had very little compression or limiting sounded too soft in general (compared to the average commercial cd they aired) and I could really hear the station compressors clamp down hard on uncontrolled peaks in my program material.

Since then, I've begun to use the Levelizer's peak limiter on the master out at anywhere from 80-73%, plus the normalizer set to 98-99% (along with individual track compression, mostly to control peaks) and now don't appear to be fighting the station's setup.

If you are hearing an improvement in what happens on the air, you probably did the right thing.

Can somebody give me an idea what I'm probably dealing with at the station end since my experience appears to be oposite the advice I'm seeing here?

I don't know how much this still holds true, but at one time public radio stations tended to be less processed than their commercial cousins. They might be using similar processors, but the processors would be set more conservatively. I would guess that most public radio stations figure that their listeners tune in for their content rather than for how loud they are on the dial. The way the processing behaves is determined largely by how it is adjusted, and there are a LOT of different adjustments that can be made.

As to "conflicting with advice", the professionals complaining about excessively aggressive processing in mastering are NOT saying "don't process at all". It is more a matter of going too far with an otherwise useful and beneficial process. If the bit of limiting you describe is the only "mastering" you did, just pulling down occasional short peaks, I would say you were being quite conservative compared to common rock/pop mastering fashions. I cannot speak to what is "normal" for jazz...

jeromee

07-14-2005, 08:26 AM

Hey Guys

I noticed the same problem with my CD on a small station here in Va. Shortly after that I bought Bob K's book and everything started making sence (although I still catch myself pumping up the levels too much:( ) My question would be, If I were using the sonoris meter and my music is kind of a bluegrassy/rock sound, should I meter with the k14 and still keep my peaks landing on 0 or is it alright to let them hit +8. This probably sounds like a goofball question but I am trying to learn this meter reading stuff:D
Thanks
J

PieterS

07-14-2005, 09:22 AM

I would say, focus on the sound first. The best way to sound good on the radio is to make the music sound good on itself. Choose a peak to average ratio that fits your music and use the meter as a guide.

UpTilDawn

07-14-2005, 09:47 AM

I'm assuming the music sounds good before I send it out! :D

Only having pursued recording professionally for a few short years (if that can be a reason), I haven't delved into the use of compression/limiting/gating.... all those dynamic processing instruments in any great depth until more recently.

This is one of the most curious areas of the recording process that I've ever tried to understand. I don't have a great knowledge of the math, but I grasp the basic principle behind the tools and the words drescribing their function.

I think that rather than change the course of Carey's post entirely, I'll post my follow-up questions to a new thread.

Thanks for the replies.... this leaves many more unanswered questions for me.
DanT

Cary B. Cornett

07-14-2005, 11:21 AM

If I were using the sonoris meter and my music is kind of a bluegrassy/rock sound, should I meter with the k14 and still keep my peaks landing on 0 or is it alright to let them hit +8. This probably sounds like a goofball question but I am trying to learn this meter reading stuff:D
Thanks
J

If I have understood his writings correctly so far, Bob K intends his metering standard, not as a Replacement for judging levels by ear, but rather as an aid to that process. I do know that he is trying to get people in "the industry" (including "DIY independents") to adopt properly calibrated monitoring systems as possible. After enough listening with a properly set up and calibrated system, you will develop a particular "comfort zone" for perceived loudness. Using that developed perception along with the calibrated monitor level control and K metering will make it possible to select the "processing level" that you want and then consistently hit that target by setting your processing so that the volume "sounds right".

So, if I have understood correctly, the use of K metering can help you some by itself, but not as much as when you set up your monitoring system according to Bob Katz's recommendations. Imagine being able to read the tachometer on the sports car you are driving, but not being able to see how fast the landmarks are passing by...