Rashad, what is mastering?
Mastering is also the strictly technical process of producing a master. Which can be a lacquer for vinyl production, or a CD or whatever medium, but there is a finished product at the end of the day, which is the master from which duplication is launched.But from an artistic point of view mastering also is approaching someone's music in a state which should be as close as possible to the vision or intention of the artist, and evaluate how much of this aim is achieved and give it the necessary help. The changes applied can be very subtle or sometimes very drastic.
Nowadays, many times the music comes direct out of the stage of composing, which is often strongly related to the mixing process already, or composition and mixing has been merged into one thing, and then the first outer-world process is the mastering. So there is a heap of projections: if music sounds good people blame the mastering, if music does not sound good people blame the mastering. The importance or significance of the mastering process has been upgraded a lot because it has to deliver everything which was originally related to other steps in the production process. Things like having an outside producer or somebody else to take care of details in regards to the big picture.
The initiative for making music has been shifted to the production side and so did the perception of the process. The colour of the composition, musical inspiration and decisions, all these aspects have been shrunken a little bit perceptional in favour of the production and that's also why mastering is seen now as having such a musical impact.
So what happens is: people come to me with their music, we talk about the format, because from the start on you have to consider what format do you treat the music for, and as a first step I arrange the pieces in the final playing order.Then the next step is basically listening to the music and finding out what intention is prominent in the music, what is the basic... um.. sculpture or basic message or basic groove or whatever it is, that carries the piece, and to judge a) how much of that intention is achieved and b) how much of my personal feeling for how it should be is achieved?
Of course the artists intention is prior, but very often mastering is about confidence and also people not wanting to be in charge anymore for what the result will turn out to be. People want to trust.Technically speaking it is a very little thing. Like with all audio treatment you refer to frequencies, to amplitudes, and to phase relations, and these are the three parameters you can work on. For that purpose you utilise equalisers, filters, compressors... and that's it.
So the set-up is like a left-right set-up physically, but the former left channel carries the mid signal, and the former right channel carries the side signal. Decoding signals like this allows for a very precise control over the stereo image during the mastering process.I think the most important tool probably is the parametric equaliser so I have a couple of bands of parametric equalisation here. Followed typically by a tube compressor, then followed up by another filter section which is a passive filter designed for the low range, and two active bands for the higher mid range and the high range. Then there is another stage of transistor based compression.
Optionally another stage of EQ is available which I classically use very narrowly and then there is the direction mixer which decodes the signal back to left-right and gives the opportunity to make final corrections in the stereo field, followed by a brick wall limiter. And after that there is the analogue to digital conversion. So my signal chain in this room is exclusively analogue, framed by the DA / AD converters.
They all have a pretty decisive character and I think that's for me mostly the point. When judging audio art one of the prior parameters for achievement is decisiveness.So I don't have a strong feeling for 'this sounds right or this sounds wrong' - or this feeling is always bound to: does it sound decisive or does it sound like someone was twiddling around and achieved something and was just happy? Does it sounds like somebody made a decision to have it sound like exactly this? This is for me a very, very important quality.
I tend to say that the science of the machines I have chosen is, that they all have a very clear vision of what they should aim for, what they should help for, what they should add to the signal. And I do appreciate that they are all high end designs, so they can be rather invisible, as well as being able to add a lot of character.That's why there is such a lot of EQs and compressors in the signal chain, because all of them are used very decently and it sums up.
I forgot to mention the other type of tool I came to like and use very much in a mastering application, which is tube distortion, which I also use in mid/side coding. The input of the distortion is fed from a graphic EQ so I can emphasize and roll off signals which I don't want to be included in the distortion. Sometimes the distortion is rather brutal but mixed in at very very low levels. And sometimes I only apply it to side signal, sometimes only to the mid signal.I did not expect that distortion could be so useful, I thought is was more a production tool but is has become a very central piece for mastering for me.
But you can apply the parameter of sound quality to distortion as well. There is low quality distortion and there is high quality distortion and what I can provide here is rather high quality distortion. Here I can add harmonics, in a way that is impossible to achieve with any kind of equaliser or filter alone, because they can only enhance what is already there.I am not in favour of trying to add colour with an equaliser; most equaliser designs are not really fit for that. Sometimes you need to roll off a lot to get rid of unwanted signals and the mix is totally in the mud and you don't have nice or interesting high frequencies and there is no sense in boosting these parts of the spectrum. So it might sometimes be better to replace it by something else. Adding distortion also can help in definition of spatial parameters, helps broaden the stereo image in a subtle way.
I was eager to have at least a software limiter after the conversion so I don't have to run the converters too hot to achieve nowadays zero margin levels, and I found one plug-in which kind of does the job ... and I was really eager to find something which is great- it doesn't seem to exist actually.This is not really digital versus analogue topic! In the digital realm there is high end stuff which is in some regards superior to analogue stuff, especially when it comes to the very crucial topic of limiting. Take the Weiss DS1 (digital mastering) compressor for example! I didn't encounter any plug-in yet that comes close to the quality of the DS1.
Hmmm.... I don't like to use plug-ins, maybe they just don't have the right interfaces.[A note, from late October 2008: Rashad finally found a compressor plug in that he likes interface and sound-wise enough to use it frequently R.H.]
With my hardware I don't ever have to look at it. And this is important for me, I really tend to stop looking, two inches before my eyes, when working with sound. I don't want to see the timeline of the piece, and most plug-ins show some sound related graphical information, which gives you a certain kind of precognition, some kind of pre-expectation towards what is gonna happen next or how the dynamic range has changed.I am fine with seeing a little bit of gain reduction, some blinking lights. It doesn't even have to be precise, because I wanna hear if the result is right, I don't wanna be tricked into looking and seeing like Whaaa! 2dB of gain reduction, that's gonna be good!
Most plug-ins provide too much visual feedback for my taste, which keeps me personally away from mapping the sound to my body, and to my ears - and that's how I master like: that I scrutinise the sound how it hits me. Not necessarily brutally, but how it addresses my body, and if I have to focus on a computer screen and operate with a mouse that is something else, its just a different world.
interfaceI have to look and I will look at every parameter if I have a graphical response that shows me that I've notched away 2.73dB or 2.89dB. I will take this information in, whereas when I work with analogue or hardware gear I just don't care at all about this kind of information, I don't get that anal about the values! With pIug-ins I tend to really go far beyond the decimal point and that's just not the right setting, not the right state of mind for me.
Same happens with customers, here people sometimes for the first time see their album layout on a computer screen as waveforms, piece after piece, and just from the way it looks to them, they might like it or might not like it, and they judge the proportion of the piece from a graphic perspective, they see different levels, they see that one piece has maybe a much thicker representation than the other one, and suddenly they get the feeling that it might be a weak piece, or definitely too low in volume. Even though before seeing it they felt it had the right prominence.They come to the mastering place or the mixing studio and I transfer all tracks from their HD recorder into the computer and they suddenly look at their music and start losing trust in the tension of the music and in the narration. They hear it a second time, they see the Boom! coming and suddenly they are really unsure - they ask me: how long is the piece again? And all that considerations which have not been a topic for composing suddenly become very important.
This all comes with using computers and this also happens in the mastering stage if you work on a computer screen. You are not independent of these things. Well, I am most certainly influenced by that, and that's why I like to work with hardware. You have the narration in the room from your speakers, and that is it. That's what your are judging.
So I do the musical mastering, and afterwards I take care of the technical restrictions of the target medium in a second step. The difference between the CD mastering and the vinyl mastering is that you don't have any limitations on most parameters of the sound expect for the final peak volume on CD, and you have much more physical restrictions which do translate to sonic restrictions, when transferring sound to vinyl. So that's why its a kind of two stage process. In the first step the considerations are, does it feel right? Is the narrative emphasis where it is supposed to be... stuff like this, and then in the second step the question is, will I get this on vinyl like that? No... but, yeah... but no... but yeah, sometimes...
vinylSometimes you have to do musical restrictions to the pieces when mastering explicitly for vinyl. Like you might have to lower the prominence of critical parts of the vocals if you want to make a very loud yet undistorted cut.
The question is: how precise will the needle be able to follow the geometry provided by the groove? The accelerations and the movements of the playback needle can get rather complex, especially if any kind of phase information is present, which leads to highly complex groove shapes that are engraved into the record.
I have been cutting several thousand of vinyls and I really have to think hard about when ever I had to cripple a stereo bass signal beyond musical recognition because it wasn't translatable to vinyl. That might be three cases, in all that years.People should not bother about that, they should make the music the way they feel it should sound like!!!!
99% of problems related to phase can be corrected in the mastering room effortlessly without making a musical disaster out of the piece. With most phase adjustments that need to be done for vinyl, you can nicely live, and I never experienced somebody who was really shocked or totally unhappy with the result. Of course in the high days of IDM there were a couple of pieces that had 180 degrees out of phase full level 50 Hz sine waves and such, and that gives you problems...But that is so much the exception that it is not in balance that every producer takes this so strongly into account which just keeps him away from making music. People should focus on sculpting colours the way they feel it should be, and not think about the post-production.
People should not make music for post-production.
Instead I try to educate my customers to think the other way round: Scrutinise every signal for consistency, check for what disturbs it, and try to remove that, and not primarily check the signal for what's too little...I always think negative. I know this is much less fun actually, but the results will be much more consistent and also louder. People try to achieve loudness by saturating media and that's just the wrong way, its the other way round! Saturation can be done at the very very end. If you saturate your medium from step one on, you will have music which will have a constant high level but will not sound loud.
The basic mistake is that people compress or limit without a musical vision.They compress or limit because they feel it should be louder, and they do not do proper AB comparisons ... that's the thing you have to learn - making correct AB comparisons. Its absolutely not easy and it takes several years of practice.
When it comes to limiting, whenever you have the chance to go to a mastering room, don't do any limiting at all! Sometimes it helps on a single track to get rid of certain spikes, but don't do it on the sum. Also be careful or at least think twice before using compression in the sum.
limitingDon't use maximisers. Never ever. Except if you need something to sound as loud as possible on TV. Maximisers are just not a tool to use in a musical process. A mastering engineer will not have the chance to do anything with the music anymore if it has been maximised already. The track is flat, dead and sealed, with no access to details.
But a much more important point here is, that this often implies the question if a person can master their own music by themselves. Its a bit a closed circuit in thinking. One of the most evident legitimacies for the mastering process is, that some one else you can trust finishes your music.Software might be able to replace my racks of hardware, but it cannot replace the engineer.
If you have to master by yourself, do not attempt to do so in the mix. Mix from a musical perspective, do not mix into a limiter or compressor, and carry the finished mix to twenty places to listen to it in different conditions, and let some time pass by.And then do the mastering. The mastering should be a totally separate process.
...tricky question. What you need is a mastering console where you can easily compare and access signals, because mastering is all about comparison. If you cannot compare properly you cannot master properly. So that part should be the centrepiece, and the starting point of your considerations should be the signal flow, and how to control it.The mastering desk is also very important because this will be one piece of equipment that will always be in the signal chain and everything which is remarkable about it will be remarkable on every piece of music which goes out of your studio.
You would need some decent parametric EQs, and I would suggest getting one or two colourful broader EQs. If you go for equipment that has lots of character, you have to be careful, because if it is too recognisable, all your masters will sound the same. It should be equipment which is capable of giving character but which is also capable of being quite neutral.
Im turning my back on multi-band compressors recently. I have been a fan for some years but the more I work with them the less happy I am. It is good to have at least one frequency selective compressor, especially for the bass range, but the classical multi-band designs have three ore more bands and you get more trouble than benefit from that. But to have a frequency selective compressor is very handy, you can use for de-essing, you can use it to control the low bass etc...
Also you have to ask yourself, will this be a piece of music that will mostly be listened to very loud, or is it music which will be listened to normally in a home listening environment? That should change your way of approaching the piece and how you are looking at it in regards of sonic consistency. Not the same results will be consistent in both scenarios. Of course people will also listen to club music at low volume at home, but if you have a reference it should be taken into account.And besides that, I find it hard to judge compression at normal listening volume. At normal listening volume its easy to make wrong judgements about compression. At very very low levels compression mistakes become very evident and at high levels also. So the first thing I do is to listen to a track at a very very low level, I also do AB comparisons often at the limit of the capability of hearing. Besides that, the longer you work on a thing, the lazier you get, so you have to really discipline yourself to change the listening levels, but it should always change. When I work I always have one hand at the volume control. One should change the level two times a minute! (laughs).
The mastering engineer Bob Katz tries to establish a fixed listening level for mastering, similar to the fixed level approach for mixing movies for cinema, where it really makes sense. But I am not very confident about this when mastering music...
And then of course it depends on how close the sound is to what you wanted to achieve. Actually, in contradiction to all the expectations of the clients, the closer the sound already is to what it is supposed to be, the harder it is to master it. The more repair it needs, the easier it gets. Things that sound unachieved ... every little help will give a lot. But if it sounds already really, really achieved, and someone sits there and says: "I want you to get that extra notch, I want you to add your character" or something like that, this is sometimes much more work and much harder to achieve. If customers come with an album, and say "it really sounds very good already, it shouldn't be a lot of work", I'm getting pretty scared actually.
objectivitySome might focus on the voice, and others might focus on the kick-drum. And so their perception of an even level throughout the album might be different. It depends on the complexity of the music and your selection form that complexity, so even these parameters can't really be objective. You could want that it sounds good in a club, and you could think that is objective, but then again what is 'sound good'? You know, even with the same piece, different people might select different elements of prominence. Some are totally happy with the snare carrying the track, while others are really intrigued by how shy this one element in the background is in relation to all the other elements in the track. Objectively you can say that this piece will be more dependent on spatial parameters than that piece, because the frequency-distribution is more even, so it will be not as much affected by any room resonance as the other one which has one very strong and prominent resonance in that tonal element around 120Hz, and whenever a room matches this, we'll get into trouble. You could say that there's an objective to aim for an even distribution of frequencies, but then again, that might not be musically right. Because some song might need that totally exaggerated low mid peak which does not sound even at all, and then it would be a miss-achievement of the mastering if it would attempt to get rid of it completely.
You would say objectively speaking this or that will not sound good on all hi-fi systems. I don't think that this should be the only reference for mastering - to make music sound reliably represented on all sound systems. If that blurs the character of the music beyond the intention of the artist, then I say its a miss-achievement.Though my mastering approach is probably very much dependent on my personal experience, I still feel like my mastering efforts are very much apart from taste. I don't feel like: "I really like that or I really don't like that" - I always see the piece as an entity outside of me. I feel what is right for the piece or what is wrong for the piece, and customers will have to rely on that. Of course, I will discuss what I feel, but it will be my personal judgement at the end of the day. Does the music address me in a way I can receive it? Yes or no? Does it it address me in a way I like to receive it? Yes or No? That's the questions, and that needs to be subjective.
I don't take much pleasure from music that's really greyish. And sometimes if there's a long row of tracks coming in which all have the same greyish type of consistency I get a bit wasted. It's hard for me then to still feel if it's right or wrong, because it just generally feels unpleasant.
By producing music that's always aiming for maximum loudness you don't give people good reasons anymore to properly listen to the music. And that will retroactively shape the reasons for making music, and that will lead to a situation where all music is kind of exchangeable.And that's a real pity, I reckon. I try with my mastering to give people a reason for not only listen to their music on their PowerBook speakers! Clients do tell me that it needs to sound good on PowerBook speakers. When all music is made in that manner, is made to fit the technique the music is played back with, then all music will be shaped from the same wood after a couple of years.
And I can understand that people don't take the effort anymore to get proper hi-fi systems, because they're just not given the reasons to do so anymore. I think the general evaluation of musical products suffers from that. Like also the total availability; you don't have to go searching for music that you're interested in anymore. Music became even more push media, it's not music on demand anymore.
loudnessThe topic of loudness: There's music which needs to have loudness - there's a certain idiom in music where you always have the contrast of quiet parts and blasts and dramaturgical there's a huge potential in loudness of course. But if you make all music always loud, steady-state, you lose that potential. Besides from the physical effects it has of tiring your ears. You know about transients and you how they shape your cognition of sound. Transients are a phenomena or an aspect of music which are really vanishing from music. More or less, there are no transients included in music any more, and this is taking away so much of musical information.
I'm not willing to take part in that race for loudness anymore, because it is so damn redundant at the end of the day. Each amplifier has a volume knob.I have often heard "Well, you need to pump up the master a little bit more because the mp3 player don't give enough level on the headphones." I don't know if that's really the case because I've never used an mp3 player but it would be disgusting if it was the case, but then again what damn levels are people listening to on headphones? This is really making me feel sick! And I don't want be responsible for that. If it's really the case that current mp3 players don't have headphone amplifiers strong enough to cover up all the noise from the city, then they will have them in half a year! I don't know if it should be the ideal aim of mastering to support music-hostile environments.
I knew that, I could smell that in the very beginning - so I thought I'd do that for a year. It wasn't like mastering was something I always wanted to do, that's really not the case. I started mastering without ever really having watched anybody doing it. Also when I started at Dubplates, most engineers were leaving as you know; you started being involved with Ableton, Andi wanted to make more music, and so I had maybe two or three sessions where I learned how to handle the vinyl cutting machine, and then everything just fell into place somehow. I just started doing it. I don't even know if mastering is so much apart from any other sound treatment - you know, the most important thing is that you know what you want, because mastering is about decisions. So you can master any track in twenty-five different fashions, and at the end of the day if you screw around with the track and you think well, that's probably good, let's do it like this, it might be good, but it might also sound like its "probably good".As I said earlier, for me the achievement in mastering or in sound is that it sounds decisive - that's my idea of 'good'. I always found this kind of easy, I think that probably the only talent I really have - making decisions for other people. I'm not so easy in making decisions on my own music, but I can relate to other peoples material. I can feel safe with the decisions. Of course I also learnt that more by confirming the decisions I made in the outside world, by listening to records I've cut in clubs or something like that. But this is what feels like my qualification: to feel trustworthy on the decisions I make. I've just started doing that, because in all the working scenarios we had since the early nineties, that's the only qualifications that all the people had to have: Take responsibility, not necessarily really being capable of doing something, but taking the responsibility of what you do.
So I just started taking it.