final mix before mastering

tone_aot

Owner of ToneJonez.com
Hello! I've been seriously thinking about getting more professional with my music. This is my dream. I wanna start getting my music mastered. I was wondering what should I NOT do in the final mix. I'm so used to hearing mastered commercial mixes that when i do the final mix i try to get the same sound.
 
there are general things that shuldn't be done (no hard limiting and cranking up the mix volume, for example). However, it's really a good idea to ask the person who will master your music, precisely what he expects....
 
yeah, I would suggest for one, keep a good copy of what you've done to the track already for the mastering engineer, it'll help them better whatever a lot more.

1 thing I know for sure, do NOT compress too much, and like said, watch the volume levels. They gonna do more things to it & will need the headroom. They'll also compress it even farther. On top of that, radio stations also compress AGAIN. Anyways, their job is to polish what you did & make it sound better. I'm no mastering engineer, so I don't know exactly what they do, but I know it'll be in the lines somewhat of what we do @ the studio already (I think).
 
I wrote something about this a while back:

Mixing

First of all, get away from the track for at least two days if you can. Avoid listening to anything you've done; listen to stuff you like by other people.

Ok, once you're ready to start mixing, try this method:

Bring up just the vocal track till it's at a comfortable loudness level. Now, slowly start bringing up the other tracks, one at a time. As you bring up each track, stay focused on the sound of the vocal track. If you bring up a track that seems to make the voice sound a little softer, shut the fader off for that track and go on to the next track, always listening to the vocal track as your reference, killing any track that makes the voice sound softer.

After you've gone thru all the tracks, you'll probably have 3 to 5 tracks shut off. Those are your problem tracks. You'll hafta eq, pan, compress, or reverb them, or leave them off altogether.

But listen closely to what you currently have going. Do you "REALLY" need "all" of those 5 missing tracks in the mix? Does the mix sound stronger without them? On the tracks you absolutely hafta have, put them where they don't cover the vocals, using any of the tools mentioned above. Then take about an hour break, get out of the studio, and come back later and listen to what you have. If you're satisfied, shut everything down, leaving all your settings in place.

The biggest problem most new people have when they wear too many hats is understanding and accepting what's really important to the song.

Stripping the song down to the bare essentials and building up from there is not easy when you've invested a lot of time in arranging a particular part. But sometimes you just hafta stand back far enough to look at the song objectively and honestly decide whether that part is really helping the song, or is it just your ego talking?

------------------------------------------------
My friend, Mixerman, said:

"This has been posted many times in several different places. Seeing as I will be bringing these steps up from time to time, and there is a constant influx of new people, I will post it here as a reference.

1. Mixing is an attitude.
2. If the song sucks, the mix is irrelevant.
3. Working the room, keeping people happy and relaxed is half of mixing successfully.
4. Putting everything proportional in a mix is going to make a shitty mix.
5. Gear are tools in a mix that make life either easier or more difficult, they are not what makes a mix good or bad.
6. A mix can be great and not have great sound.
7. If nothing about the mix annoys someone in the room, the mix is often times not done.
8. Mixing can not be taught, it can only be learned.
9. The overall vibe of the track is much more important than any individual element.
10. Just because it was recorded doesn't mean it needs to be in the mix.
11. Be aggressive.

What can I say? My steps are kind of like a Marshall amp. They go to 11."

------------------------------------------------
And I once posted this:

The Dreaded Mixdown

This is where many new recordists fall down. It's one of the hardest things to get right, but there are a few things you can do to help get your mixes closer to where they should be, right from the start. (MixerMan, who gets paid big bucks to do this, will hopefully jump into this thread at some point.) It requires a different mindset from tracking and arranging. It also requires that you not be married or in love with any one part in the song.

Tip 1. Get as far away from the song as possible before you try mixing it. Don't try to do a mix right after a tracking session. Your ears are fried, and you're too close to the song right now. Objectivity is the word to remember. Wait a few days or even a week or more, if you have that luxury.

Yes, some people can do a good mix right away, but that usually takes years to acquire that skill. If you haven't been doing mixes for many years, you ain't one of those people, so wait.

Tip 2. Mix low. Yes, cranking it sounds cool, but it will also introduce more room reflections and give you a warped picture of the sound. Crank it when you think you've got the mix nailed, but keep it low for as long as possible.

Tip 3. Listen to the song, not the tracks. The biggest mistake new mixers make is soloing each track and making it sound full and rich by itself, then they wonder why the whole thing sounds bloated and muddy. There are several methods that work to construct a good mix. You can start by bringing all the faders up, with the pan pots centered, and all effects turned off, or you can decide what the key element in the song is (the vocal, for example), and start working from that. Different engineers use different methods.

Tip 4. Build a box - a small stage in your mind. Imagine a stage. You control where the player appears on that stage. Panning lets you control left to right placement, volume and reverb lets you control front to back, and eq lets you control the frequency blend (low to high).

Tip 5. Resolving conflicts in the mix is the single biggest problem facing a mixer. You'll always find several tracks competing for attention in the same frequency range. The kick competes with the bass. The bass competes with the low guitars. The guitars may be competing with the vocals. The keyboards are all over the place. It becomes an even bigger problem for most people when they solo a track and work to make that instrument sound as big as possible. Bad move. All the instruments hafta work together and a particular instrument has to sound good with ALL the other instruments.

For the good of the song, some of the bottom end on the bass or the guitars may have to be eliminated. Yes, the instrument may not sound good when it's soloed, but it will blend in better when you listen to all the tracks. It's up to you to decide which instruments need to be shaved, but if you concentrate on the song first, it will start to become more and more obvious what needs fixing.

Tip 6. Take frequent breaks and get away from the music for a few minutes. Rest your ears. If you're doing it right, it's the most demanding part of the whole recording process. You are literally listening to ALL the instruments at the same time, following them all at once, and it's easy to burn out. Wanna see an engineer really blow up? Try talking to other people in the control room while he's trying to work on a final mix.

There's a lot more, but we'll save it for another day, or wait to let others weigh in on this most difficult of all subjects.

Harvey Gerst

----------------------------------------------------------------------------------
And Mixerman added:

"Ahhhh.. my favorite subject. I could speak for hours and hours on mixing. Harvey's tips are great. Definitely valuable to the beginning mixer.

What can I add? Well let's start with the fundamentals of what you're working with. It's a lot to digest, particularly with Harvey's list, and it should probably have it's own header, but I'll put it here anyway.

Barring 5.1, you only have 2 speakers to work with. But we live in a 3 dimensional world. So we're basically creating an illusion so that a mix sounds 3 dimensional. Let's call this a spatial illusion

When mixing there are 5 planes of spatial illusion. Level, panning, frequency, spatial perception, and contrast. These five planes are all used to create space in a mix.

Front to back: (Level)
Level gives an element of a mix it's own space. Compression on individual channels helps keep the level so that it doesn't disappear in the mix. A loud instrument will appear forward, or towards the front. A quiet instrument will appear to be back or further away.

Left to right: (Panning)
Panning allows you to give an element of the mix it's own space. For instance putting a guitar part hard right keeps it from washing out the vocal.

Up and down: (Frequency)
Frequency is the use of EQ to boost or cut frequencies that either muddy or clear the mix up. For instance 250Hz-700Hz are fairly muddy frequencies, and if you have too many instruments using this frequency range the mix could be muddy. Everything in an arrangement or mix should have it's own unique fundamental frequency space.

Far and near: (Spatial Perception)
Spatial perception is the use of reverbs, chambers, plates, delays, far mic placement, etc.. to create the illusion of space in the mix. An instrument with allot of reverb can sound like it is placed in a large hall. An instrument or a vocal with a long delay, can sound like it's in the alps. An instrument that's completely dry, will sound like it's in a small carpeted room, right next to you.

Sparse to dense: (Contrast)
Arrangement is the use of muting, and altering the recorded arrangement to create space where it is needed to accent the more dense parts. The use of density to contrast sparse is great for creating the illusion of dynamics in a mix, within minimal dynamic range. The use of a limited dynamic range makes for better listening in more casual environments, where there tends to be external noise.

All 5 of these planes work together to create the illusion of space in a mix. One is no more important than any other in general, although one or two of the planes could prove to be more useful in a given mix. Not all are a requirement for a great mix either. For example, your mix should to be able to break down to mono, and still be a great mix."
 
I can tell you one thing I heard Tone and that is NOT to use any mastering effects in your mix at all...................
 
Excellent suggestions Harvey!

A couple of things to keep in mind as far as mastering:

You don't really know how a song is going to fit on an album until you have all of the tracks mixed. So don't do anything that's going to negatively affect how the ME does his job. This includes trying to set the final level of any of the tracks by over limiting/compression, overall EQ, and even fades.

Keep the final tracks at the highest resolution possible. For example if you're mixing at 48K/24 bit, keep the mix at 48k/24 bit, don't mix down to an audio CD.

Try to address issues in the mix that are difficult to address after everything has been mixed down. For example levels and EQ of individual tracks, amount of compression on individual tracks, placement (front and back) of the tracks, effects, etc.

Hope this helps ...
 
Mixerman via Harvey Gerst said:
Sparse to dense: (Contrast)
Arrangement is the use of muting, and altering the recorded arrangement to create space where it is needed to accent the more dense parts. The use of density to contrast sparse is great for creating the illusion of dynamics in a mix, within minimal dynamic range. The use of a limited dynamic range makes for better listening in more casual environments, where there tends to be external noise.
Well, it's very refreshing to have someone else on this forum whi is not only familiar with, but uses the multi-dimensional perspective to mixing. :) Seems like every time I bring it up, I'm met with the Internet forum version of blank stares ;).

Everybody has a slightly differing take or interpretation on just what the exact number of dimensions "should" be and exactly which dimension is which. Those semantic difference are not extremely very important, IMHO; the important thing is the idea of using the idea of at least a 4-dimensional soundsatge (or 5, depending upon how you want to divide the definitions) to give yourself a complete picture of the blank canvas you have to work with.

I'd like to add a little to the discussion of the "density" dimension, as mixerman calls it. While a rose is a rose, I prefer to call this dimension either "drama" or "movement". Thinking of it that way allows one to consider more than just density (which is very important to this dimension, of course), but also such creative elements as pan and level automation, mix arrangement to fit the song arrangement, etc. It's the use of density and space, volume and quiet, and dynamic focusing on different arrangement elements to creating a moving dynamic to the mix rather than just a static "blueprint".

Alsi, I find it interesting that mixerman divides the front/back dimention into two dimensions, front/back and far/near. The way I visualize it it, it's all one dimension, the one of linear distance from the listener. The fact that one is *mostly* acheived via gain levels and the other is *mostly* acheived via spacial effects such as reverbs to my mind is aactually a limiting perspective for two reasons. Not only because those spacial jurisdictions can overlap, but also because those processes can be (and are) often combined for the mazimum resolution in that dimension.

Again, not saying that mixerman is wrong to divide the two. Honestly, whatever visualation works best for the engineer is what's best for the engineer to use, and there are differences between visualiations. I'm just using this as an example of how visualizations can differ and why.

And it's important to understand that some of the "dimensions" are virtual sonic dimensions and not real physical dimensions. This can be confusing to some rookies. For example, the dimension of frequency is not really an up and down dimension in the physical sense. Changing frequency balance in and of itself will not cause a guitar to appear to rise up off the stage or drop through the false floor. It's just that the "up/down" metaphor is an easy one to use since one usually considers the audio spectrum in terms of ranging from low frequencies to high frequencies, and being able to go up and down the frequency spectrum. But it has nothing diretctly to do with what we call "up and down" in the physical world.

It's a similar situation with density or drama. This sonic "dimension" has no relation to any of the for tangible physical dimensions. While it does tend to play itself out in the dimension of time, it is not time itself. It again is just a virtual dimension, or property of music and music mixing.

G.
 
There's an interesting comment from Brian May on the "Classic Albums - A Night at the Opera" DVD. To paraphrase, the arrangement (and by extension the mix) needs to evolve to help the song tell its story. Cloned choruses prevent that evolution so cloning choruses is something else to avoid.
 
WOW! Thanks for all the help guys. I'll definetely use all the tips given here. I even had the great Harvey Gerst respond to my thread. This is cool! Thanks i appreciate the help. Is mastering really high for a whole cd, 12-18 tracks? I've fiddled around with mastering plugins but i want the real thing.
 
Mastering isn't really all that expensive for what it can do for your finished product. As far as making a 12 - 18 song album, I wouldn't, if I were you. I'd do this instead:

Choose the 12 best songs out of the 18 and make a quick CD, in the order you want. Get away from it for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 11 songs on it.

Get away from that for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 10 songs on it. That's what you send to be mastered.
 
Harvey Gerst said:
Mastering isn't really all that expensive for what it can do for your finished product. As far as making a 12 - 18 song album, I wouldn't, if I were you. I'd do this instead:

Choose the 12 best songs out of the 18 and make a quick CD, in the order you want. Get away from it for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 11 songs on it.

Get away from that for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 10 songs on it. That's what you send to be mastered.


Love your Knowledge Man!
 
I guess I should post some things I've done to at least let people hear some of the different stuff I've recorded. I tried to convert these to MP3's, but I hated how that changed the balance in the mix. So, here are some WAV files that are big, but honest representations of 3 songs I did about 10 years ago.

I had 3 Tascam DA-38's and the Soundtracs Topaz board. I think the most expensive mic I used was a Shure SM81 (that I picked up used for $100). The reverbs were probably a Lexicon Alex or a Roland SE50 (or both). Probably ran the whole thing thru a Waves L1.

They're all pretty big WAV files so if you're on dialup, forget it.

Soul Providers
Great big soul sound, with Jeff Barnes sitting in, adding two sax parts. I think I'd move Jeff's solo more to the center if I ever redid this song.

No Exit
Softer pop sound with Jeff again on sax, and Kathy Kersey doing all the backup vocals. Joanne (the vocalist from Soul Providers) is plaing congos.

Kirby Kelley
One of the best slide guitarists around, just jamming on Crossroads with some of his friends. Two great guitarists, so I just panned them wide - Kirby's on the right.

Hopefully, these will download okay.
 
Last edited:
Harvey Gerst said:
Mastering isn't really all that expensive for what it can do for your finished product. As far as making a 12 - 18 song album, I wouldn't, if I were you. I'd do this instead:

Choose the 12 best songs out of the 18 and make a quick CD, in the order you want. Get away from it for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 11 songs on it.

Get away from that for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 10 songs on it. That's what you send to be mastered.
One of the best posts on preparing for mastering I've ever seen. Thanks, Harvey.
 
Harvey Gerst said:
Mastering isn't really all that expensive for what it can do for your finished product. As far as making a 12 - 18 song album, I wouldn't, if I were you. I'd do this instead:

Choose the 12 best songs out of the 18 and make a quick CD, in the order you want. Get away from it for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 11 songs on it.

Get away from that for a week or more and listen to the CD again. Pick out the weakest song and make a new CD with the remaining 10 songs on it. That's what you send to be mastered.

Harvey,

Why 10 songs? Does this number have anything to do with licensing/publishing?
 
masteringhouse said:
Harvey,

Why 10 songs? Does this number have anything to do with licensing/publishing?
Nope, but the chances of ANYBODY putting out a CD with even 10 "good" songs are slim to none. Ten songs works out to somewhere between 35 to 45 minutes which is a nice length for listening.

If ALL of the 18 the songs really are great, he's still got 8 left over for a second CD.

Basically, it just helps him do a little self-editing, so that the final product is as strong as possible.
 
masteringhouse said:
Harvey,

Why 10 songs? Does this number have anything to do with licensing/publishing?

I'd add this:

In my opinion, the long-playing vinyl record imposed the optimum time constraint on an album. I have many CDs which use all the available time (70-80 minutes), and I suffer listening fatigue before they're done. No matter how good the music, I'm ready for something else after 35 minutes.
 
AGCurry said:
I'd add this:

In my opinion, the long-playing vinyl record imposed the optimum time constraint on an album. I have many CDs which use all the available time (70-80 minutes), and I suffer listening fatigue before they're done. No matter how good the music, I'm ready for something else after 35 minutes.

There's a Punk band that I work with that has songs that are about 2 minutes in length (average). They even have one song that was 4 seconds long and barely made red book standard for a track. I don't know that the 10 song rule applies to them so well as a standard, but Harvey's point is a very good one in general. Leave people always wanting more as an artist, not like they've gorged on a feast and are about to throw-up.
 
Thanks guys! Good info. That's a great process for me to make my cd a good one. Yall's advice is greatly appreciated. :D
 
Back
Top