You've finally nailed the EP. The days you spent watching YouTube tutorials feels worth it on reflection, as the efforts of your mix-down seep confidently through the speakers. There was some guesswork, sure; but the thrill of completion more than makes amends. All that's required now—and you have been eschewing thoughts on the subject—is to master the thing. You understand that it's an important step—vital even—but wonder really if the outcome will justify the cost? Would a simple application of limiting and EQ at home not suffice? And how loud should recordings be anyway?

Stefan Betke, AKA Pole, has been addressing questions such as these since 1996. Hailing from Düsseldorf, but now residing in Berlin, Betke initially stumbled into the profession simply as a means to make a living, taking a position at Berlin's fabled Dubplates & Mastering. In 2000 he opened Scape Mastering and has since applied the finishing touches to productions from artists like Thomas Fehlmann, Richie Hawtin, T. Raumschmiere, Mike Huckaby, The Mole and Delano Smith.

Under his production alias Pole, meanwhile, he continues to mine the deepest depths of dub and reduced techno—explorations that have produced seminal album projects such as the late '00s number series and 2007's Steingarten. Ahead of his mastering masterclass at this week's AKG Scholarship Of Sound, we sat down with Betke to try and better understand what goes on behind the mastering studio doors, and what producers need to consider before taking that crucial final step.

The fundamentals

Pre-cut vinyl

If someone was to say to you very simply "what is mastering?" how would you respond?

The easiest way to explain what mastering is? A mastering engineer optimizes a stereo file mixed by the producer, musician or band, focusing on different formats like CD, online release, vinyl or a commercial on TV. There are different needs to make the sound of a recording sound best in a certain context. You can't say what sounds good for vinyl is good for CD or what's good on CD is good on the radio automatically. You have to make a decision: What is the track made for and how can you optimize it for a certain use?

Where the product is eventually going to be heard.

Exactly. This is the technical side of mastering. I always like to divide the process of mastering into two different parts. There is the technical side which has to do with physical aspects, like frequencies, volume and what is needed to make a track become, for example, cut-able for vinyl. If you have a lot of bass problems it might become difficult to cut this track for vinyl; if you have too many high frequencies it might cause distortion in the cut and so on.

The second part in mastering is the musical element, which means you have to understand the idea of the artist and the concept of the album and then mastering becomes like a last instrument to the whole music—you have to decide if "the snare is loud enough" or "the snare is not loud enough." The engineer becomes a third ear in the process of releasing music. Since the engineer is not involved in the composition, their ears are more objective. They have a different perspective, and maybe a clearer view to it than if you had been sitting on the album for three months and mixing everything day-by-day.

Would you say there's a big difference in how a recording would sound out of different mastering houses?

It would sound totally different.

Do you think the average listener would be able to distinguish between them?

Maybe if the listener would know the un-mastered version. On the purely technical side it will probably be the same. For example, if you visit different studios to get your 12-inch cut and the vocals on your tracks are very sharp and not de-essed, every engineer will probably try to correct this to optimize the vocals on the album to make them sound better. This is a technical decision that the mastering engineer has to do. But at the same time if you go to my studio or to another one, the taste of the mastering engineer is always different and the understanding of the music is different, so there would probably be one person who wants to make it really loud and reduce the dynamics whereas I maybe would like to keep the dynamics or the bass, etc. These musical decisions are made based on how the engineer understands the idea of the music they're working on.

Artist relations

How much input would an artist have into this ordinarily?

When I do mastering I discuss every decision with the artist, except if an artist wants me to decide on my own. But, in general, if an artist decides to come to me and they know my work they usually leave me alone. If I don't know the artist and haven't heard his music before then I really try to find out who he is and try to talk to him and understand where he comes from and what ideas he has. Is the music made for a club, released on vinyl? Or is it made for multiple functions like music that works in the car and on the radio as well as in the club? I really try to find out who the producer is and what he wants before I start working.

And would you usually just have an email exchange with them?

Yes I will, or sometimes I just call... I always give the artist a listening copy before I do the final cut or CD, so that they can listen to the master on their home stereo which they know best. When they come to my studio it's nice but they don't know the room, they don't know the speakers, so it's a little bit difficult for them to make a decision. So this is an important part and it's one of the tips I can give for both sides: It's important that the artist is not shy and starts communicating to the mastering engineer and asks any question he has.

There are no stupid questions when it comes to "what will happen to my music?" It is the artist's music not the engineers so it's really important—and if you feel insecure with your mix, please ask before the final mix-down is happening because you can't change it afterwards sometimes. On the other side, as a mastering engineer, it is important that you talk to the artist to understand what his music is really about. In my opinion—others might see it differently—this is the first step to become a good mastering engineer: understanding and communication.

Would you feel comfortable fulfilling artists' stylistic wishes if you don't agree with them? For example, if they want it very loud and heavily compressed.

I mean, in the very end, it's not my opinion that counts. It's the artist's opinion that counts. It's his music and as a mastering engineer I have to be respectful. I can give tips—that's my function. For example, you have a track that you want to master and the artist says, "I want this to sound like a big dubstep production and it should rock the house." But there's no bass and it's not well-produced and it's relatively quiet, then I might have to say "I'm sorry, but this might not end up sounding like a big dubstep track because you didn't produce it in that way." Then the artist has a choice: Either to live with this compromise or go back to his studio and change it. And I'll give him/her production tips.

Do producers often come to you and, as a result of your feedback, go back to the studio?

Sometimes, but not that often. Very often people say, "thanks for the tips, I'll do it for the next track." That's another tip I can give: if you have a mastering engineer that you are happy with then try to stay with him. The more he works for you, the more he understands your music. At the same time you as an artist learn more about what's needed for him to make things sound really good. The mastering engineer and the mix-down engineer and the producer, they grow with every track they work on together.

The mix-down

Tube-Tech Compressor CL 1B

I guess this is a good opportunity to ask what would you advise, on a core level, producers get right in their mix-down?

I have to say one sentence first. If you start producing, make sure you that you have a room in which you can trust the sound that you hear. Listen to a lot of records from artists that you know and like and try to understand the room. If you feel comfortable in this room and if you think you hear the bass clearly and the high end clearly then start your mix-down. Follow the idea that everything that sounds good is probably in the right position in the mix-down. If the snare hits your ears and you make it louder and then it really hurts you, you can be sure that the snare is not well EQd and is not balanced right in reference to all the other instruments in the mix-down. So this is the first step: listen to it and decide, does it sound good or better, does it sound like I want it to? The balance of instruments is important.

How much do you think the quality of things such as EQ plug-ins that you use during a mix-down will make a difference to the overall quality of the recording?

It's not really a question of high class equipment to get a good result in your mix; it's a question of you listening to what you do. You can use a cheap compressor—this might add some noise to your system because it doesn't use the highest quality components, but it is a compressor and if you reach the goal that you want to go for then it works. It's always good to have better equipment, but this will grow in time anyway. When I started mixing for my first Pole album I had a Behringer compressor and I had one cheap EQ; I had a 16 channel Mackie mixer and I made my first three records on this mixing board. It doesn't really matter.

If someone was to say to you, "I'm going to try and master my tracks at home," what would you advise them? Would you simply tell them not to bother?

I had this discussion with a young artist from London just recently or basically with his manager, because his manager said "I am releasing many online releases of this artist as promotion. It becomes really expensive to go to mastering every time." So he asked me if I could explain to him a little bit how to master his music himself in his studio. I said: "That doesn't really work because he would need a lot of experience," but what I suggested was that he should learn how to do better mixes, because the better the mix is, the less you need mastering.

So in theory there could be a perfect mix-down and you would not need mastering, or not that much. The artist or the producer in his studio should not think about "How could I become a good mastering engineer?" He should learn better how to become a good mixing engineer. Then it solves the same problem. Don't waste your time learning a different profession; mastering is not mixing, it's a different profession with different skills, different equipment and different knowledge.

Mastering professionally

VMS 70 - Georg Neumann Gmbh cutting lathe

If someone was interested in becoming a mastering engineer, what would you advise them to do?

The easiest way to do it is to start in a mastering studio, try and get an internship or something like that and learn it there. It's much quicker if you have someone who can tell you things and who can explain what's necessary on the physical side, and what's important to know about all the gear... Otherwise you do a lot of tests at home and you can never be 100% sure if that's the right way or not.

Do you think it's possible for someone to have a natural aptitude for mastering or is it something that is purely taught?

You need to be a little bit talented I guess. A mastering engineer should have really good ears. You can train your ears. For example, when you study an instrument or when you study tone engineering you have to learn an instrument at the same time because you really learn about the frequencies much better when you play an instrument yourself. Anyone who has good ears and is really interested in music could become a mastering engineer, at least in theory.

Would you say it is a pretty big learning curve?

Yes, it is a pretty big learning curve, especially when you want to learn all the different formats like CD mastering, vinyl cutting, mastering for TV, cinema—all these formats need different knowledge, which means it needs a different process of mastering. To be honest, the learning curve never really stops.

In process

Under the microscope: vinyl grooves

If I was coming to you with my project, before I've even sent over the files, what would you ask of me?

I think I would ask you to send me some music first. But after receiving it I would start to listen to your mix-down and would like to ask you to do the same. Just to make sure you are happy with the result. I want to hear from you something like: "everything is kind of good, but I'm not sure about the kick drum..." So we come back to this balance thing again. You should make sure that the mix-down is as good as you can do it and you are happy with it.

On the technical side it's important that you mix it down in the highest quality that you have. So if your DAW supplies 24-bit, and 44.1 khz [sample rate] then please use it instead of using 16-bit. You can use 96 khz but it's more important that the bit rate is high than the sample rate is high. Sample rate eats a lot of space. If you record it in 44.1 khz this would be absolutely fine especially because we would convert it down to 44.1 for a CD anyways. In my opinion it is absolutely enough to use 44.1 khz, but 24-bit is a must I think.

Why is that?

It's just that the dynamic range is much higher and the noise ratio is much better if you record on 24-bit. To use really simple words, it plays back more information frequency-wise than a 16-bit file would do, but this is now on a really simple explanation level.

Just to give you more to work with basically...

Yes, so this is really important, and what also is important is to leave a little bit of headroom in the mix-down.

How much are we talking?

Peak level I would say you can hit the 0db some times. RMS or average I would record between -2db, -1db. Even -3 is good as that gives us enough headroom to work with the music.

I know this is a very broad question, but could you talk us through the typical process that you would undertake?

Where to start? [laughs] Let me try to explain a simple and normal mastering procedure. If I get tracks sent over, I first listen to the album, or the EP, to every track that is delivered in the running order it will be released at the end. After listening to it I get an idea about what the music is about, what it's made for. If I have any questions about it at this point I contact the artists or the label and discuss these questions with them and then the next step is that I start to develop an idea of what I would like to do with it. Do I want to use digital equipment mainly or do I want to use a combined version of analog and digital gear? And if I use a combined version, do I want to go first through digital and then through analogue or the other way round?

So let's say I made a decision and I would like to combine digital and analogue gear and I would like to go through analogue first. Then I might run it through, for example, a tube EQ just to add some warmth to it or to work on the high end a tiny bit, and then the low end a little bit to get a first little change in the balance of frequencies if this is needed. That could be a start. I never do it the same way, but this is just one example. And then I might, if it's a very dynamic recording, just put it through a limiter after that. But just very softly—not heavy limiting, just very, very soft limiting to catch a little bit of the peaks and to get better dynamic control. It's not a drastic change; it's just like a fingertip.

Top: Urei model 1178 analog limiter
Bottom: Weiss (Gambit series) DS1 digital de-esser / compressor / limiter

And the decision over whether you go digital or analogue: Is this just something that you get a sense for over time?

Yes, that's right. As an example, if you work on pure digital productions, someone who made his music in the computer and the hi-hat sounds very digital, very sharp, then it might be necessary to run it through a tube compressor or tube EQ just to add a little bit of warmth to it. Sometimes it is already enough to run the music through the EQ even in bypass mode.

To continue in the process... I could go through a digital EQ that works with a very narrow bandwidth—a very, very narrow bandwidth—then I boost the band that I would like to work with, and sweep through the whole frequency range from 50 Hz up to 20 KHz step-by-step and see where I can find resonances or bad sounding elements in it to filter them out. That is one method that I use, it's called subtractive EQing. You search for the problematic tone and you drop the band and it takes it out by -2, -4db or whatever's necessary.

So you might need to use a very narrow Q and then drop by up to 4dB?

You drop it as much as it's needed actually. Of course you have to be careful with that. You should not really do a drop of -24db at 160 Hz if you are not really sure. You might lose some energy on the kick drum. So it's always a compromise. You have to go as far as needed but you should always be aware of where the limits are.

Are there such problems in every track you work with?

Nearly in every track, it's just normal.

Is it just a collection of frequencies that creates a problem area?

It can sometimes be really problematic, but not always. The better the mix-down and the experience the producer has the smaller the problems. But I barely hear any track which doesn't have elements that could be optimized.

Are there any other tools aside from EQs and compressors that are key to the process?

It's mainly EQ, de-essing and compression or limiting—more limiting than compression, maybe. Compression is something that should be used more on every single instrument in the mix-down already so I prefer to do just some limiting on the whole track.

Is the process between mastering for different mediums very different?

Very different. Especially mastering for vinyl, it's a different procedure. Cutting vinyl keeps some physical aspects in addition to the mastering problems itself. On the digital side you can reproduce everything—it's just a code of numbers, everything can be reproduced on CD. On vinyl, you have the problem that these phenomena cause problems on cutting and reproducing music onto the dubplate. A phase shift in bass frequencies for example would cause the needle to jump out of the groove. Extreme amplitudes in certain high frequencies may increase the heat of the cutting head. Distortion in the reproduction of these high frequencies might occur. What sounds good on the digital side might then sound distorted on vinyl.

Pole

Do you think doing what you do has influenced your own music?

Yes, there's a good element in it and a very bad element in it. When I started making music without being a mastering engineer I had more dirt in my music. It was not so much "sound designed." I think the biggest criticism of my productions nowadays is that they sound too good. This is the reason why I changed the production style for my next album... I'm not going to finish every track up to the final mix-down. I'm just recording the instruments and I will ask a friend of mine to mix it down for me, with me sitting next to him. Usually I work two days on a bass drum and two days on a guitar and then two weeks later I've finished with the mix-down of one track. And now I would like do one track per day, or a day-and-a-half.

And what about the flipside: what's the good thing that you take away from it?

The good thing is that my bass sounds really good! [laughs] I think I know how to make a bass sound good. It's also good for my live performances. When I play a concert it's easier for me to adjust the PA in a way that fits my music, and most of the time in-house sound engineers allow me to adjust the sound on my own because they know I'm a mastering engineer. I just played in Tokyo a few days ago at club Unit and when I came in they were like, "OK, this is all yours."