..:: TAREKITH'S GUIDE TO MASTERING ::..

Tarekith · Post by **Tarekith** » Mon Sep 10, 2007 11:22 pm

I'm writing this because I have been seeing the same questions asked more and more these days, "How do I master my song?" or "Can anyone recommend some good mastering plug ins?" I guess this is a natural thing, as more and more people are learning the ins and outs of their home studios, they eventually start to finish songs and realize that they don't sound as good, or as 'polished' as what they hear when they buy a CD or download tunes online.

So, to start with, who am I and why should you read this? I'm the owner of Inner Portal Studios, a Seattle based facility specializing in mixdowns and mastering. Just some background so people don't think I'm just talking without any experience as I write this.

So, what is mastering (technically it's called pre-mastering BTW)? Surprisingly it can be many things:

Making all the songs on a CD sound cohesive.
Prepping the song so that the playback level is not too quiet, and more importantly these days, not too loud.
The final quality control for projects going to a replication house.
An experienced, fresh set of ears to help achieve the overall balance of frequencies in a song.
Just a few examples.

I think the one thing most people want to achieve when it comes to mastering their songs is two-fold though: Getting the overall level of the song right, and getting a good balance of frequencies in the song as well (i.e., making sure it's not too bright or too bassy). The easiest way to achieve this is to let someone with the right gear and the lots of experience handle it. But, if you're reading this, then chances are you really want to do this on your own. So that's not an option. How then, do you go about mastering your own work? What tools do you need to get the job done?

I think it goes without saying that two truths will always remain true when it comes to music production:

More expensive gear will almost always give you better results. A $1000 plug in will likely sound clearer and smoother than a free plug in. There's always exceptions, I'm just saying in general.
Experience and trained ears will always get you better results than just good gear. That is, an experienced mastering engineer will be able to get very, very good results no matter what they use.
To me, this means one thing. Becoming good at mastering (or anything) will take practice and lots of it, and having good quality tools at hand will help as well. Lacking these two things, there's one more way to approach the situation, and that's trial and error, which to me is a form of practice as well. So where do we start?

The first thing you need to look at is your listening environment. Having good speakers is only one small part of knowing you are hearing everything in your music, especially with any sort of accuracy. Far more important is the way that the sound from your speakers interacts with the room you're in. The best speakers in the world can sound like dog poo if your room is negatively influencing what you hear. I won't go too much into acoustics here, but if you're working in a room with no acoustic treatment (i.e., diffusors, bass traps, etc) you're already starting off with things not in your favor. Given that this sort of stuff can be expensive to address, I'll assume that most of you haven't, which means you need a plan B.

The best way to get around this limitation then, is to make sure that you listen to your song on as many speakers and playback locations as possible. For the longest time my car was my second test location, as I by far listen to more music there than anywhere else. But don't stop there, try your song on your iPod, your home stereo, your kitchen radio, your friends' stereos (they hate this by the 5th or 6th month BTW), etc. Learn how to translate what you hear in your home studio, into what it sounds like elsewhere. Pay attention to things like the bass. If it sounds good in your studio, but you keep noticing it's too bassy elsewhere, then you know you need to compensate for that at home. So you mix and master with the bass sounding weak at home, so that when you test it elsewhere, it sounds good.

It's not that simple though, you need to learn this for the whole frequency spread, the mids, the highs, the low mids, etc. Eventually you'll slowly start to hear what the deficiencies are in your studio, and how to compensate for them. This takes a LONG time. It's not something you do in a day, over a weekend, or even in a couple months. You need to train your ears to always listen for the overall balance of what you are hearing, and then test that against how it sounds in your studio. What's worse, getting better speakers, a new soundcard, or even finally springing for some acoustic treatment all changes this balance and you need to start all over. Without the proper room and gear, this is the only way you'll know what's truly going on in your music: trial and error. So it's not ideal, but my point is that it can be done with enough practice and perseverance.

"Yeah yeah", I hear you muttering, "but how do I master my songs in the first place?"

To put it bluntly, you don't.

What?

Let me say this again, in case you skipped over it. If you are writing and releasing your own songs, there is no reason to “master” them per se.

Everything you need to do to make a song sound good can be done in the mixdown, and this is where you should focus 100% of all your learning and attention if you ask me. There's only one exception, and that's getting the levels more inline with today's standards, and I'll come back to that later.

A good-sounding song doesn't need anything done to it by a mastering engineer. It already sounds good as is. This is what you should strive for. Putting things like multi-band compression, EQ, aural exciters, sonic maximizers and such over your mixed down song is the WRONG way to approach it. Those tools were created to give mastering engineers more flexibility when they didn't the luxury of going back and fixing the individual elements in a song. They were forced to work on a single stereo file of the song, and couldn't adjust anything in the mix. Thus tools like these were created for those RARE instances they needed to adjust something beyond what simple EQ or compression might fix. You have the luxury to go back into your DAW and adjust the problems right at the source, so use it!

If you listen to your completed song and think it needs these tools, then stop and go back and fix your mixdown.

- If the song is too bass heavy, then go back and turn down the bassline and kick, or add some EQ to just those parts.

-If it sounds too mono and centered, start panning some instruments until you get a wider sound-stage. Add some chorus or stereo delay to a part to make it seem wider.

- If it sounds flat and dull, take off all the effects and EQ you added previously and start over. I'd say 90% of the time this “dull” phenomenon is due to people over using tools they don't understand.

Contrary to all the popular magazines, you don't need compression on every single track, especially for electronic music. I maybe use 1 or 2 compressors total in one of my songs, if that. The best way to get a rich-sounding song is to not overdo the effects, and to try and get a good mixdown without using anything but your volume faders first. It's also important to realize that this stuff takes lots and lots of practice (years and years), so even if you do all the above and then some, it's just not going to happen overnight when you're just starting out. Keep experimenting at home in your free time, and eventually you'll get the hang of it. You'd never expect to be as good as Jimi Hendrix on the guitar in a week, and good sounding productions are the exact same. Patience!

The point of all this is simple, if you're doing your own songs and no one else is going to work on them, spend your effort getting it sounding good while doing the mixdown. Burn copies of that and live with them for a week while you play it on as many systems as possible. Always strive to get the mixdown sounding good first. That should be exactly how you want your song to sound. There are no magical tools that will make this better in mastering. By far, you have way more options, and more transparent options, when addressing this while mixing.

If your mixdown is balanced the way you like, then the only thing you're going to notice while playing back your new song everywhere is that it's quieter than everything else out there. This is fine when you're testing the mixdown, just turn up the playback device to compensate.

DO NOT WORRY ABOUT HOW LOUD THE SONG IS UNTIL YOU ARE COMPLETELY DONE WITH IT.

Doing so is only distracting you from what's important, and trust me, it's so simple to fix you don't need to worry about it yet. When you're doing your mixdown, make sure than the master levels in your DAW never go above -6dBFS, and just turn up the playback device to compensate. Here's a quick guide I wrote on setting levels and why you want to leave the master channel alone, for those that want more info:

http://tarekith.com/assets/DigitalLevels.html

It's mainly geared towards Ableton Live DJs, but the principles are the same for all computer-based musicians.

So, you're happy with the way your song sounds. You've listened to it for a couple weeks on multiple systems (yeah right) and it's perfect the way it is, you wouldn't change anything. Well, except for the fact that it's just too quiet still, right? All you need to do now is use a limiter to gently raise the overall volume. The key word here is gently. Far too many songs these days are over-limited purely for the sake of 'apparent' volume, especially dance music. Google "loudness wars" if you really want to read more about it.

There are two ways of applying this limiting: by using a limiter on the master track in your DAW and then rendering the final file, or by rendering the file and then applying limiting in another app like Wavelab, Soundforge, Peak, or WaveEditor. The results should be the same no matter what you use. It mainly comes down to your own working preferences, and if you have enough CPU power left to add a good limiter in your DAW. If you're going to use another app for mastering, render your mixdown as a 24bit file (you should always use 24bit anyway) and make sure Live's normalize function is off when you do.

Anyway, get your best limiter, set the release to Auto, the main output to -0.2dBFS, and then lower the threshold until the very highest peaks of your song are only being limited by about 3dB's. You'll see this on the gain reduction meter. And I mean only the very highest peaks of the song. The GR meter should just barely flicker up to 3dB. If your mixdown was well done and balanced, I'd bet anything that you're pretty close to the ballpark you need to be in. It'll sound pretty competitive with most music out there, in terms of volume. Maybe a touch quieter, but that other music is too loud anyway, right? This is where experience really comes in though. Sometimes you can go more or less than 3dB, but you have to know what to listen for, and when you're truly doing more harm than good. When in doubt, err on the side of caution and use less limiting.

But what if you don't have a limiter? Well, luckily most apps come with a freebie limiter or a limiter preset for one of their compressors, since a limiter is nothing but a compressor with an infinite ratio (or as close to inifinity as we can get digitally). Live for example has a limiter preset for Compressor II (just change the output to -0.2dBFS before you start). While in general I think there's no such thing as a preset in mastering, if you truly don't know what you're doing this is by far the safest option. And since people will ask, my favorite limiters are the UAD Precision Limiter and especially the Voxengo Elephant 3.

Ok, now you've got the overall spectral balance right, and you've sorted the levels, what's next? The very last thing in preparing your song is to apply dither. Dither is used when converting 24bit (or higher) files to 16bit. So if you're only working with 16bit files, don't worry about it. If you are using a 24bit file for your mixdown, then you want to insert the dither plug in AFTER the limiter, it's always the very last plug in you apply. Always. And you only ever want to apply it once. Since Live did not include a dithering plug in prior to version 7 (tsk tsk Ableton!), you'll need to use a third party one (I like UV22Hr in Wavelab myself). If you don't have a dithering plug in, it's not essential, it just helps to make the 16bit file have a bit more depth (no pun intended). I bet most people would never notice it if you didn't use one, so don't stress too much over this. If you want some examples of what dither can do, here's some I made:

http://tarekith.com/assets/ditherexamples.zip

Basically, you can hear how the dithered version trails off more smoothly than the truncated version, albeit at the expense of some added noise. These examples were boosted A LOT to make the effect more audible, normally this stuff is extremely quiet, like -94dB or so. It's subtle, like I said. This is just a more audible example of it for the sake of understanding. Again, Google or visit wikipedia for more detailed info than I can go into here.

Anyway, that's more or less it. Render your song as a 16bit wav file, burn it to CD, and you're done! I hope I've made it clear that there's really no reason whatsoever to use a complicated mastering chain, or fancy multi-band tools, on your own songs. You will achieve FAR, FAR better sounding results if you do as little as possible to your song outside of your main DAW, and instead focus on getting everything right there first. It's not as complicated as people make it when you're prepping your own material, it's only when people like me or true mastering engineers have to deal with other people's music that it can get complicated and we need to resort to these tools.

I hoped this helped some of you, and if you still want to have someone else master your songs, you can, of course, contact me.

scottwallis · Post by **scottwallis** » Mon Sep 10, 2007 11:31 pm

Tarekith, that was an excellent read.

Thank you very much

funky shit · Post by **funky shit** » Mon Sep 10, 2007 11:49 pm

what a read!!

Thanks!
shit load of info there!

again many thanks

COSM · Post by **COSM** » Mon Sep 10, 2007 11:53 pm

Interesting Read - Thankyou

Tarekith · Post by **Tarekith** » Mon Sep 10, 2007 11:54 pm

Oops, fixed the dither link, sorry.

stjohn · Post by **stjohn** » Tue Sep 11, 2007 12:28 am

great stuff thanks!!!

another one is i sometimes find that making everything 200-300 Hz mono can tidy up the bass regions... but im usually wokring with techno and other bass heavy stuff...

aburgener · Post by **aburgener** » Tue Sep 11, 2007 12:35 am

nice - this should be a sticky... or saved somewhere

thanks for the info

Tarekith · Post by **Tarekith** » Tue Sep 11, 2007 12:38 am

stjohn wrote:great stuff thanks!!!

another one is i sometimes find that making everything 200-300 Hz mono can tidy up the bass regions... but im usually wokring with techno and other bass heavy stuff...

That's a mixing tip, not a mastering one, you learn well!

aqua_tek · Post by **aqua_tek** » Tue Sep 11, 2007 1:50 am

great read sir! Quite enjoyable.

Here's a question for you, and I know it's very subjective... but you thoughts on a plugin like Ozone? I like the fact that it has a Limiter, Multi-pressor (which I dont necessarily intend to use, but is a nice thing to have available), EQ, AND dithering, at a very modest price. Does the "you get what you pay for" message apply here?

I keep going between wanting to get Ozone and a UAD card. Big difference in price, I know. But I'd be willing to pay for quality

beats me · Post by **beats me** » Tue Sep 11, 2007 2:09 am

Most informative post I've read on here in a long time. And I actually understand it. Thanks for sharing your experience and knowledge.

Can we set up some kind of script that when anybody types "mastering" into the topic line on the board it just redirects them here? I would like that.

Tarekith · Post by **Tarekith** » Tue Sep 11, 2007 2:26 am

aqua_tek wrote: Here's a question for you, and I know it's very subjective... but you thoughts on a plugin like Ozone? I like the fact that it has a Limiter, Multi-pressor (which I dont necessarily intend to use, but is a nice thing to have available), EQ, AND dithering, at a very modest price. Does the "you get what you pay for" message apply here?

I keep going between wanting to get Ozone and a UAD card. Big difference in price, I know. But I'd be willing to pay for quality

Well.... Ok, I'll be honest and completely un-PC, but I really don't like Ozone. It's got a decent of tools for the price, but for whatever reason I can't quantify why I never liked the sound of it. It's just me, I've heard some people get great results out of it.

But yeah, you really can't go wrong with the UAD IMO. I'd take the Pultec and Precision Limiter over anything else any day of the week, by far my favorite mastering tools. SOOOOOO smooth and transparent, I'm getting misty eyed thinking about it

I had to sell my card when I went 100% laptop, and it was one of the hardest choices of my studio life. I definitely want to get the new Xpander asap. The UAD stuff is IMVHO some of the best sounding plug ins out, well worth the money. I'm a big fan.

Tarekith · Post by **Tarekith** » Tue Sep 11, 2007 2:27 am

Oops, forgot to mention, downside of the UAD is still no dithering included. I'm sure there's some freebie ones out there, I just haven't had to look into it too much myself.

pulsoc · Post by **pulsoc** » Tue Sep 11, 2007 3:07 am

Why should I waste my time reading this long-ass post when everyone KNOWS mastering means sticking a brickwall limiter on your chain and cranking the input until your rms is -.01db?

Tarekith · Post by **Tarekith** » Tue Sep 11, 2007 3:18 am

dude, why you gotta be giving away all the good secrets?

lunabass · Post by **lunabass** » Tue Sep 11, 2007 7:18 am

Yep, excellent read Tarekith. Gave me a fuzzy feeling inside as i realised I already do NEARLY everything you have written here...so I feel I'm on the right track although I know I've still got a long way to go. i really appreciate your effort in putting all this up.

I do tend to limit more than you with my largest peaks hitting the 5db mark with some of my trax but I guess this is personal taste and for me is dependent on the track

One question I have though, Why dont you like using a multiband comp in conjunction or as part of the limiting process? Personally I've found that a touch of multiband before the limiter gives me a more transparent comp/limiting result...just wondering what your thinking is with regards to that.

cheers