I also think this is a great thread. And yes Mike we agree. I'm just not sure what it is we are agreeing on. I think I understand what you posed to start this all off but I don't think I ever gave it much thought before. At least not in this context. Any rate good thread.

What IS exactly right?  Is there such a thing?

the best place to have the highs and high mids is as close as possible to the average mix of the major songs in that genre

How to do that:

After you've gotten the sib/cymb/air balanced, apply a temporary stand-in for what will be your mastering compressor and do an A/B Mash with the major songs of the genre, adjusting your EQ until you have high end chaos and can no longer tell which cymbals and sibilance are yours and which are theirs. When you turn the other song off, your mix will be very close to the standard, and will not cause the listener to reach for the EQ.

That's it.  Of course there's no ultimate perfect, but that's perfect for that song.

MasterInTheMix


Why should the vocal sibilance be loudest? Because if the listener can't understand the words, they'll be less involved in the song, usually a lot less. Now, that doesn't mean that they should stand out 30dB over every other HF sound. They should be brought down to within 10 or 15 dB of the vocal air with a de-esser or multiband compressor (LP-64 MB). That leaves a fairly narrow window between the sibilance and the vocal air for the cymbals.

Why should the cymbals be next? Because they're fun to hear AND they let you get away with a lot more vocal air without sounding ridiculous. It tricks the ear and makes the hot vocal air sound plausible. If you put cymbals over the sibilance, you can't hear the words. If you put them under the vocal air, you run the risk of the vocal sounding artificially bright even though it's exactly as bright as its cohorts.

Why is the vocal air next? Two reasons. 1. It's exciting rather than dull, compelling you to listen. 2. Along with the air you also get all the non sibilant articulations up to where they can be understood. M's L's and W's can actually be heard rather than guessed at by the sequence of vowels (like the "Peanuts" teacher).

I very much like this. I must remember it. Some great concepts here that can be applied to more than just vocals and cymbals, but anything with high components (and ultimately everything). It's important to decide the relevance of each part of the frequency spectrum of each instrument in a song and what feeling it gives the listener. EQ is just so amazingly important in mixing. If I could only have one tool, that would probably be it. I love eq.

I am going to have to disagree on the idea that there is some sort of hierarchy of levels or formula to mixing, even within a specific genre. 
 
O.K. so I'm a little behind the times, but I just got my first MP3 player so I can listen to music in the gym.  I am listening to professional, world class acts recorded in multi-million dollar studios and I am amazed at the difference in not just the loudness of the different tracks, but the use of the sonic spectrum.  Some tracks are brilliant and some, by comparrison, are dull.  Some are bass heavy and some mid-rangey, but all of them are great mixes on their own.  I have never listened to any of these albums and thought the mix was lacking anything, but when you hear one song from one album followed by one song from another, the differences really stand out.
 
I believe the bottom line is this: If you can make a mix that does not distract the listener from being able to enjoy THE MUSIC, you have made a good mix.

John

ba_midi

John


I also think this is a great thread. And yes Mike we agree. I'm just not sure what it is we are agreeing on. I think I understand what you posed to start this all off but I don't think I ever gave it much thought before. At least not in this context. Any rate good thread.

Who's Mike in this case?

Ah, wrong name sorry Billy. No excuse on my part just simply wrong.

OH it's ok John :)  I just wasn't sure if *I* was missing something or misread something, etc.
 
Or I thought, just for a minute, I had drank your vodka instead of mine lol :)
(j/k)
 
 

johnnyV


 

That's a pour example, Those albums have been encoded to MP3 by god only knows who!
Go back and listen to the originals.
MP3 files are notoriously all over the map with loudness. Some are terribly distorted, some sound like they were transferred from a cassette or with a Sears o Matic turntable system.
 
.................................................................................................................................................................................................................
 
I strongly disagree.  I ripped all of the mp3s myself from the original CDs.  They were all done exactly the same way on the same equipment.  Still not convinced?  I have done work with actual audio tracks taken from the original CDs and brought in to Sonar as wav files.  The results are the same: Big differences in volume, highs, lows...the sonic spectrum varies greatly from album to album.  To even begin to think that you could take different artists, different studios, different equipment, different producers, different EVERYTHING and end up with all mixes sounding relatively the same is ridiculous.
 
 

Philip


Hot topic ... not much to add as I'm learning exceedingly.

Totally agree that some sizzle (9kHz+) helps most mixes if not too hissy/essy ... and done at the track level as Evan and others cited.
(thank you!)

Mastering: Warming the mids and tightening the lows with boosts at 47 (sharp) and 113 for hopeful kick and bass pomp).

Of course every song demands completely different EQing and multi-comps and tape vs tube depending on genre, style, preferences, audience, time-of-day, moods, blah blah!

Perhaps equally important (for me of late) is:

1) applying Haas effects and compression, especially vox thickening techniques, vox chorus ... etc.  Remember it is vox recording that makes or breaks everything.

2) and exactly the same thing for dist guitars.  JMO and 'feeling' of late.

3) I Find a vox or guitar sample-loop that I like and emulate every parameter and feeling of it.

At risk of sounding like a broken record - I always take issue when someone states absolute values such as "Boost ast 113Hz".  I just don't think that's a wise approach in general.  Unless someone is using the EXACT same setup in detail, all the time, then maybe a specific frequency works.
 
But that's hardly ever the case, really.  Even with the same setup -- at least in 'rooms', there will be some variance somewhere.
 
The reason I take a stand on this in particular is I see, all too often, this being said to many who may be new to the whole area of recording.  SO they read something in a forum and take is as an absolute/law.   I think it's just not the best thing for newbies, in particular, to hear.
 
 

jimmyman


  When I think of Lows and Highs the term dynamics comes to mind. One example is
say for example the the mix is sounding very nice then all of a sudden "something"
say in the high end is "too snappy" at a given point and time/place in the song.
At another place in the song/mix things may be "just right" with that same setting,
I call it the "mix" because that is where the "end result" is apparent.

So then I say O.K the problem is in a given track so I go there and do something.
Problem solved? No, because sacrificing something in a track by eq multi band comp
or whatever (or the the main mix) becomes a (have to fix) thing. That is why
I speak of (dynamics) in a given area of the song/mix as a whole.

I think that (or at least in my case) in order to have the mix (the individual tracks
as a whole unit) sound alive and nice and well rounded is that sometimes it takes a
lot of thinking about what and where to do something.

Yeah, Jimmy, I agree dynamics play an important role as well -- I think we might all agree there.
 
The thing about this thread is that I was thinking more along the lines of the resulting personality of a mix.
Let's take a human as example:
One can wear new underwear, new socks, shiny shoes, nice pants (or skirt), nice tops, etc...  and all those elements "collectively" will be the final result we see as the fashion statement, so to speak.
 
In a mix, we take a number of individual elements and treat them various ways.  The composite of those elements become "the mix."  
 
We also often hear of and speak of certain frequencies that almost *have* to be dealt with (low end rumble is one example), but there seems to be much less discussion - and reference to - the 'high' end in general.
 
How many times have you heard or read that it is wise to roll off/cut below 40-50Hz to get rid of the rumble down there.   Many times, right?
 
And, how often do we hear someone say "roll off/cut above xxKHz to get rid of the brittleness" (put any word you wish in place of brittleness).    My point being it seems it's just less discussed for some reason.
 
YET, when I listen to the records I really like, it does seem as if there's some "commonality" in the upper range that makes it sound more like a record and/or makes it sound more polished, so to speak.
 
When I started this thread, it was with the hope of getting some views on the 'overall' commonality approach - or general approach for the composite (ie., "mix") rather than drilling down to the individual elements.
 
I think we all agree that each element needs some type of treatment on its own, for sure.  But the composite mix takes on a personality of its own, I would assert.
 
 

And, how often do we hear someone say "roll off/cut above xxKHz to get rid of the brittleness" (put any word you wish in place of brittleness). My point being it seems it's just less discussed for some reason.

Then the question should be should it?  I don't know I normally don't as a rule. Though I have more lately then in the past. If I do its a very small amout and a gentle curve. If its above what I can hear I don't worry about it. I start not hearing much at about 15 kHz. It rolls off at a nice curve from there.

I am not too sure about "personality" except as it relates to a producer. They are usually tasked with trying to create a coherent recognizable sound. 

The notion that songs have "personality" is sort of obvious so I don't think that is what you mean.
This relates to instrumentation as well as form.  A trio or whatever.  Type of music and so on.

It also possible for a mixer to lend themselves to the overall sound and thus add something that would not be ther otherwise. However some would not look at this notion as a good thing.
Many would prefer a mixer to be non noticeable. I am talking about a person not a device BTW.

Again this a cool thread you started with a very cool start off question. 
 

John

And, how often do we hear someone say "roll off/cut above xxKHz to get rid of the brittleness" (put any word you wish in place of brittleness). My point being it seems it's just less discussed for some reason.

Then the question should be should it?  I don't know I normally don't as a rule. Though I have more lately then in the past. If I do its a very small amout and a gentle curve. If its above what I can hear I don't worry about it. I start not hearing much at about 15 kHz. It rolls off at a nice curve from there.

I am not too sure about "personality" except as it relates to a producer. They are usually tasked with trying to create a coherent recognizable sound. 

The notion that songs have "personality" is sort of obvious so I don't think that is what you mean.
This relates to instrumentation as well as form.  A trio or whatever.  Type of music and so on.

It also possible for a mixer to lend themselves to the overall sound and thus add something that would not be ther otherwise. However some would not look at this notion as a good thing.
Many would prefer a mixer to be non noticeable. I am talking about a person not a device BTW.

Again this a cool thread you started with a very cool start off question. 

Ok, I'll try to answer my own quest, a little ;)
 
It does seem to me that so many of the records I like have a very limited upper range.   On a spectrum analyzer I can still see > 10KHz signals on many records, it seems that the "meat" starts being attentuated somewhere above 6KHz.
 
I don't want to be too specific, so I'm just talking "ranges" here.
 
There's one track that has this large amount of hihat/cymbal use that I really like -- and I mean a LOT in this one track I'm referring to -- and just listening would make one think there's a lot of energy up in the 7 to about 12KHz range.   Yet under analysis, there is a big slope downward from about 5.5K or so.
 
The result however is both pleasing and a bit surprising on that track.   I was so envious as to how the producer was able to get all that cymbal-y stuff without creating a teeth-knashing sensation to the listener.
 
Yes, there was a great deal of (well used) compression, and other production techniques -- but the shear "end result/composite/mix _personality_") was totally impressive to me.
 
And in that regard is what I mean by "personality". 
 
I know due to the type of music I tend to make these days, that I have a general range of 6-8K, sometimes 11-12K that I seem to always "treat" (manage, etc).   Part of that is the use of soft synth/sampler drums; part is the inherent audio engine of hosts/DAWs, etc;  but in general, the "personality" changes when there's too much of those ranges in a non-acoustical mix, and in a dramatic way.
Frankly I'm starting to think I _should_ lean much more toward less "high end" overall, since that is what seems to be the trend in (good) recordings as well as not-so-good ones.
 
If one eliminates the unavoidable compression built in to MP3s and such, and if one discounts the "Radio" compression effect -- every record/track does have a personality (which may or may not be helped by these forms of compression).
 
OR - maybe I'm the only one at the moment who has the slightest clue what I mean by "personality" of a mix LOL.  I hope that's not the case ;)
 
 

guitardog247

Philip

Remember it is vox recording that makes or breaks everything.

+1

Except for instrumentals :P

Middleman


+2

Be there or be square.

Ah, but do you mean a Square Wave? 

For me the lows have always been a pain in the low-end, with my low-end gear. Then to listen to on my low-end car stereo, and low-end bookshelf stereo, and low-end PC speakers, and then to ear buds on mp3 player.

Like, is it muddy? Oh, every track I play by everyone is muddy in the low end, if played on low-endish equipment.

Butch- Sorry if I assumed you were talking downloaded MP3's. Very rare that people actually rip CD's anymore so hope you take my apologies. But that said some ripping software will mess with normalization so beware how you rip.

 I myself have ripped my entire collection using Wave Lab and they are not much different in level or EQ. They are all loaded into my computer and we play them through the home system.  What I am talking about is they are close enough in sound as to never having to run for the volume knob or turn down the bass. The opposite is true of downloaded MP3. As a sound man and reluctant DJ I loath using MP3. They hand you a CD they made from downloads and every song is completely different. The most common problem is to much bass! Dance groups drive me batty!
So my observation is that Professionally released  POP music is pretty well pre adjusted to playback without the end user having to change the EQ or level of there car or home system.
But then I have no NEW music or any death metal in my collection so what you are talking about might be a recent development.

Back to the OP topic.
I agree regarding cookie cutter EQ settings. And I think the whole process from Mike to CD has to be done with the best quality equipment and acoustics possible to get that magic sound. I can't afford it so my sound has a given kind of raw edge to it. I don't try and pretty it up because that's phony to me.
I'm a big fan of live albums because they always PLAY the songs better. Sometimes the sound is not as good as the studio version but it's more fun to listen to.
My formula remains the same no matter what format it goes to.
1. Good room
2. Good performance
3. Clean, pure sound capture
4. Very little processing
5. Use boost 11 on + 24db   ( just kidding)

6. The right drugs help too....