Mastering is the voodo magic of audio production. Everybody knows of it but few hazard themselves into the subject, leaving mastering up to the professionals. Michael Romanowski is one such professional. He's been mastering in the Bay area for almost 20 years, doing work with clients such as Norton Buffalo, The Radiators and Dredg to name a few.
I had the opportunity to discuss the creative aspect of mastering, the quality of analog as well as the evolution of the home studio. Due to amount of insight Michael shared with me during our conversation I've edited it into two parts, first of which can be read below.
How did your career in audio engineering start?
I played in bands when I was younger, and went to a lot of concerts. After a while I started noticing the sound engineer that was working in the back of the room. I knew he was obviously affecting the sound. And so I started hanging out, and I asked him all sorts of questions. At the same time, the band I was in was trying to do some recording. I started to realize that there were some similarities between the live sound and the recording process.
I kept going and hanging out and asking this live sound guy questions. After bugging him for several months, one night he said, "Hey, I'm going to go take a break. You want to run it for a little bit?" And so I did that a few times. After a little while he said, "I'm taking the night off. Do you want to be the sound guy for the night?" And I said, "Okay, great." Finally he said, "I quit. Do you want to be the house sound guy?" And that’s how I started.
As I started doing live sound, I was also starting in the studio and saw that there were a lot of similarities. I was always the one in the band paying attention to what was going on. Where the mics were being placed, gear and stuff like that.
People would come to me and say, "Hey, I heard something you worked on, and wanted to know if you want to be involved with my project?”
At home I started with my own four track Fostex cassette deck, and that was awesome. And then I went to the Tascam and from there I moved to ADAT. ADAT and a little Mackie console. That was just the beginning of, well, the end of the beginning.
Well, the beginning of where you are now, I guess.
Yes. So I started recording people at home. At the same time, I was still working in studios in Nashville, mixing and producing records for other people. I just kept doing it as much as I could, learning what I could, and things grew from there. I still record a handful of records, ten or so each year. I really enjoy the perspective of recording and mixing. I believe that helps my understanding of what an artist has to go through before they come in for mastering work.
One of the things that I really like to do is live recording. I like to capture live moments and I do everything from symphonies and choirs, to folk bands, rock bands, pop bands etc. I like the process of capturing a live event, something that is spontaneous.
Yeah, I totally get that – I do live sound. And that's something you only get during a live performance. And especially when you're a sound technician too you get to be a part of it.
Yeah, right. And there are two kinds of live sound techs. Those that care and those that don't...
And unfortunately there don’t seem to be as many that care. But if you do it's great.
When I did live sound, it was a tremendous learning experience because I was working with one, two or three different bands a night. And they all had different requirements. They needed different things: different vocals, different acts, different guitars, different drums, different monitors, etc. And it all had to be dealt with quickly and correctly applied.
I think that's a great sort of pressure for somebody who cares about sound. As much time as I spend in the studio, I love live music. But you can generally tell if the sound person is paying attention, not just hanging out by the bar, or out smoking a cigarette, drinking and talking to a friend. Not really involved in the music at all.
Right. He's just there.
Yeah, to do a good job and treat the band with respect you have to be active.
Yeah. It's bad when you see that. When they're not taking part of it, because their role is important.
Very much so. And the thing is, a band can be out there rocking, having a great time and playing really well; and out front it could sound like crap and everybody may think, "Oh, well, the band is not that great." Well it's not the band's fault, you know.
It's the house sound guy who is working on generally not a great sound system and he's not paying attention on top of that.
So from live sound you ended up in mastering. How did you get into mastering from there?
It was sort of a happy accident. I always knew that things needed to be mastered, but I didn’t know what that meant. I was on a motorcycle trip riding around the country, camping and staying with friends along the way. And I ended up in San Francisco and got a call from the owner of Rocket Lab. Rocket Lab was THE mastering place in San Francisco, in the Bay area. They said, "We hear you're thinking about relocating. Would you like to send us a resume?"
I thought, "Well, that's an interesting door that just opened up." And so, I called them and said, "I’m actually in town. How about I come by?"
That’s a funny coincident.
It is, isn't it? And so I went by, talked with a few people, and hung out for a couple days. The owner took me around the area, and offered me a position and said, "Do you want to come in and be an apprentice and learn the craft? "And I said, "Okay, I've got two more weeks of riding before I get back to Nashville. I'll call you then." And when I got back to Nashville, I think that was about the 11th of July or so, I called her up. And I said, "All right, I'm in." And by the 2nd of September I was here.
And what year was this?
This was 1994.
Yeah, that was 16 years ago. I packed all my stuff and came out and I thought, "You know, I'm going to learn mastering, do it for a couple of years here and then go back to Nashville."
But I loved the Bay area and I’ve been fortunate enough to have some wonderful opportunities. I've worked at several facilities here. And 16 years later I'm still here doing it, and loving it.
Were you always interested in mastering over other aspects of engineering? Since you went there to take a job solely doing that.
Yeah, that was the first step for me to really learn. Here's this voodoo, this thing that has to be done but nobody could really explain to me what it was, and I had a chance to learn about it. And once I started to understand it, it really resonated with me. I like stepping back and looking at the broad view of “here is what's going on”.
But I also enjoy the technology that’s involved. What I really like about mastering is that it's half technical and half artistic. You are preparing a master for manufacturing and so it needs to be correct in a lot of different ways. You need to work with the manufacturers, about delivering formats and resolution and all those kinds of things. There's a lot of technical expertise needed in making sure that everything along the way is as tight as possible. Things like formats and clock sources. It’s very detail oriented. There's also a lot of deep listening. And so the artistic side of it is really focusing in on listening to what's going on.
I get this question a lot. What is my genre? And what genres do I work best in? And not to be dismissive when people ask me that but it's all genres really, because it doesn't really matter.
Yeah, how does a mastering approach differ from genre to genre? How do you master a rock record as opposed to a jazz recording?
Well, there are definitely a few things to consider when you're crossing genres. But for the most part it's listening to sonic presentation and what frequencies are happening here? What are the dynamics that are happening here? What are the transients and the levels, and the dynamic range and the overall total balance within the song and in a body of work? Those issues are consistent no matter what the genre is.
A folk record and a rock record still have all of the same components. They have level dynamics. And with the sonic presentation, they have to be translatable between a really nice system and perhaps headphones, laptop speakers, or a car. You really want things to be as translatable as possible so that when people are listening to them – every place you listen to is going to be different – but you still want the best listening experience that you can get and in as many different types of systems as possible.
That's one of the goals of mastering. Translatability. And those things aren't really affected by content, you know, by what the music style is. That happens no matter what.
All those things are there.
Yeah. And there are certainly some things you just don't do. For instance, you don't compress classical music at all, you just don't. And there are some things that you would do for LP masters that you would do a little bit differently for CDs.
For example, you need to make sure the low frequency information is not out of phase below certain frequencies. And not too much sibilance or high frequency component. You can't cut a loud record with low bass and kick drum on one side only, it screws with the cutting head and it won't cut a good lacquer that way. Just things to pay attention to. Again those aren't really genre specific.
But how would you become creative? Sometimes the impression is that the mastering engineer just puts like all the songs together and levels them off, and makes them sound like a whole. But where does the creative process come into play when you're mastering?
That’s a great question. There are only two channels. We're only dealing with left and right. Let's presume we're not dealing with surround material but we're only dealing with left and right. And when you have only two channels to work with your choices are still frequency based. Although you can't get in and change the mix you can get in and change the perception of the mix.
Let's say somebody brings me something where the kick drum and the bass are pretty close together. It's hard to tell which one is which. But I can go in and notch out some frequencies or boost up some frequencies that give you more tones at the bass and a little more impact from the kick and separate what they're doing so they sound more separate as instruments. Just because I can doesn't mean I should, but because I can, I either choose to or not, and that's part of the art.
The same thing with frequency stuff and mid-frequency and being able to get in and say, "Okay, cool. Here's a rock thing and it's got three guitar players, but I want to make them feel like they have a little bit more edge or bite to them." Maybe it's a hard rock record and they're supposed to be really angry sounding or something like that; and if I go in and look for places where presence calls out the distortion or the character of the distorted electric guitar that is an artistic call that pulls out the emotion of the performance.
Or not. You know I can leave it the way it is, and it's fine. The same kind of thing goes with the loudness and how things become very dynamic. There's an art form to making things appear louder than they really are.
That's an art form too – part of the idea is translatability. If you're doing things that are – let's say 12 songs and they come from three different engineers at four different studios over the course of the year. And you have to make a body of work out of a collection of songs I could choose any one of those to sort of be the template and try to pull the others in the same direction. So finding a song that feels close, feels balanced, feels nice, and then taking the other ones and doing the best I can to bring them all into the same circle.
Yes, and kind of making them sound and groove together.
Yeah, exactly. And that's more again of the art form.
Is the gear in mastering different than the equipment used in the usual recording or mixing studio?
Well. The gear is similar in a way because we only have a few types of tools to work with. And the most common tools are EQ, compression and limiting.
There are creative ways to use those tools and there are other things involved as well but in a mixing session you have all of those tools. And you use those on a per channel kind of basis.
In my case I'm just dealing with left and right. One of the things about mastering equipment is it must be much more exact. It must be very repeatable from left to right but, you know, one dB change on the left shouldn't be .8 dB change on the right for the same kind of thing. But the equipment is really kind of a more bare bones, perfected version of each of the typical mixing and recording processors.
What kind of gear do you use? Do you use only outboard hardware, or do you use plug-ins too?
I use hardware. I don't think plug-ins sound good enough for mastering yet. I think there are some that are quite good and I may use some to retouch occasionally if I have to. I've done some work to it and the client comes back and says, "You know, that's really great but just, the low end is a little strong." Instead of going back and remastering the whole thing I can do a little touch-up. And I feel comfortable about that.
Perhaps I have something that I really like and I think I did a really good job on. But I wouldn't rely on them (plug-ins). I use a lot of analog hardware. And obviously you can't use plug-ins for analog hardware. The digital stuff, I have a digital EQ and a digital compressor that I like. But there's so much processing going on in each of those boxes to handle one specific task. Whereas even the fastest computer that does all of this stuff still has to run an operating system, drive the monitor, manage the file systems etc.
There are all sorts of things going on. And I really think they affect the sound. I hear things that get affected by that. I use a Mac and I turn off all of the spotlight features so it's not constantly queuing up and looking for files. Doing all of those things in the background are creating noise inside the computer that affects what's happening...
And you just want your computer to be silent, you know.
The audio – exactly. I want it to do nothing but run my capture software and a screen.
What software are you using?
I use SoundBlade from Sonic Studio.
You just basically use that to run that to load tracks into?
Yes, that's my editor. And so, I use that for capturing files, for editing together. I put files together and do the assembly, the fades, do the spaces, and so on. It allows me to do all the PQ information and all the track markers. And handle ISRC codes and make PDP images. All of the things that are specific for mastering.
I do have some functions about it being able to put plug-ins in into that software if I want and I can do some rudimentary volume automation kind of thing if I need to. But I generally handle all of the stuff before capturing. Most work stations sound different. Some sound better than others. But to me, most don't sound that good.
And some definitely sound crappy, some sound acceptable, some sound fabulous. And I believe SoundBlade is one of the best sounding work stations available which is why I choose it. Its task is to capture the sound of what I load into it. Its sole purpose is to sound as transparent as it can to what I'm loading in. I'm going to make bad decisions with my EQ if I'm changing the sound to accommodate what the work station is doing to it sonically. I'd rather that the work station do nothing to it sonically, so that I don’t have to compensate for something that my software is doing without my permission.
So you basically want the computer to act like a tape machine really and do nothing except run the song.
Right. Then I'll put leader in between songs when I want to put space between in and out. And how much space I put, and so, I'll leader the tape and put it in a box and I'll label it and send it off to a manufacturer. Other people choose to do things differently. That's what works best for me.
That's always the question. What works best for you?
Yeah, you know, that's kind of the thing too. Magazines and stuff like that all have great opinions but I think it's something they don't always relay is that that's what works for that person. I've seen forums and things where people go, "Oh, well, so and so said that this is the way it is, and so, that's the way it is." Well... not necessarily.
It's like if a reviewer gives something five stars it just means it sounded good to him.
Right. There's an experience that that user or that listener had that said, "You know what. This speaks to me. I get this." Whereas somebody else could go, "You know, this is two stars; I can't believe they actually released this." They could end up listening to the same thing but they bring different experiences to the table. The room they listen in, their speakers and past experiences all contribute to how people hear. It's all opinion and personal preference.
Yeah. Of course it is.
I think one of the greatest skills a mastering engineer can have is to develop an opinion of the sound. If you ask me about equipment, I think the most important piece of equipment in a mastering environment is the room itself, the environment, the monitoring experience. If you can't hear what's going on accurately, every decision you make is a bad decision.
I was going to ask you about acoustics and if they might be the most important factor to a mastering studio?
I think so, yes. We're making decisions and developing an opinion. I have two opinions to develop. Let's say I'm working on something. Somebody brings me a song. My first opinion is I listen to it and I have to decide what I think it sounds like. And then I have to decide what I think it can sound like. And then, how do I get from the first opinion to the second opinion.
How do I get from where I think it is to where I think it should go. There are as many different ways to do it as there are engineers. And for that very reason, because the engineer has personal experiences and perceptions, different monitoring environment and different kind of equipment; those kind of choices are the kinds of the things that dictate what they hear, what they think it can sound like, and where it should go. How they get from point A to point B – that's different for absolutely everybody.
Even two engineers in the exact same room with the same equipment are going to do things differently. That's why there really is no one way to master, which takes it back to being an art form. Or to the eye of the beholder and the person doing and making the changes it has those effects on it.
But what is your approach? You said you listen to a song and you're trying to create an opinion and join those two but what's the process like technique wise? Where do you start?
The first things I do is consult with my client. Talk with them about what they are bringing me. So I'm prepared when they show up. Maybe they're bringing me files, maybe they're bringing me a half inch of tape at 30 IPS. Just to know these things ahead of time. Where do I start? Those are sort of pre-starting things.
The first thing during the session itself is to calibrate my system. Because I go through a combination of digital and analog gear, I want to make sure that everything is operating at an optimal level. Then make sure the clocks are all locked with my clock source.
And then, really, the first step is listening. It's just to sit there and listen. Then to develop an opinion. What does it sound like? If it's one song or 20 songs, go through it and listen to a little bit of every song and sort of get a handle on the overall vibe. I consult with the client to find out if it was mixed by the same person, at the same place. There's a sonic footprint going on there because the mix environment has a sound. And the mix engineer makes decisions based on that.
And so, one of my first goals is taking the mix environment out of the equation. I may start adjusting some of the things like it being really bassy all the way throughout, for instance. Let's say somebody is mixing in a small room that has really no bass response. And so, they crank up the bass because of it. And then everything has too much bass on it.
You know, it's kind of like whittling a bird. You're given this big block of wood and you say, "Oh, there's a bird inside there." I mean you have to take away all the big chunks first. You have to start taking the big corners off and all of that stuff to get into the essence. Once that's whittled away a bit, then you go into the detail. "Oh, there's a bird inside there and here's a wing. Let me get in around that foot. And here's a beak."
And so you start getting more and more detail oriented with fine tuning EQ and each little change you make will affect what some of the bigger changes are. Then you back up and make more adjustments. You alter what you did and knock off the corner pieces. And then you start bringing it back in again. And it's sort of two steps forward and one step back, two steps forward and one step back, until you get to a spot where it feels like you’ve moved from Point A to Point B.
That's a great metaphor.
It is, you know. The song is in there. I've had people sit and talk to me the whole time and wanted to tell me stories about the songs. “This song is about my dog and he was really special to me and I want to make sure that comes across. My dog used to lay on the porch at my feet when I would sit there and play guitar and sing songs to him.”
That's really cool, and as a listener I can really appreciate that, but that doesn't really affect my decisions at all. I mean, I care, because I love music. But this is not what mastering is about. At this stage, we are polishing the final mixes and helping the songs to translate the feeling that should already be in there.
Some of the things I love about mastering is that I'm at the back end of the production phase, but I'm at the big picture stage. I'm at the stage where I'm helping people get their music sounding as good as it can before other people hear it. And what that hopefully means is that the listeners get a window into their art.
If it doesn’t sound good, people are not going to listen to it very long. But if it invites them in and says, "Hey, come on in and listen. You know, I got a thing going”. Then, people are going to start to dig into the content as listeners.
Right. Well, that's kind of goes into what you said in your article, Five Things About Mastering on your website. That you shouldn't master your own work, because giving it to you or a mastering engineer in general gives the mastering engineer an unbiased opinion and it doesn't include a dog on the porch and all that stuff.
Exactly. The more that the mastering engineer is familiar with the content, the more he or she could get sympathetic to it and less objective.
So you wouldn't say that mixing engineers should master their own work either?
I don't think so because, a mix environment is designed for mixing, and a mastering environment is designed for mastering. And if you're making those decisions about what a mix should sound like, you don't have the perspective to understand and to look at the big picture. You're too tied to it.
But also, on a technical level, your mixing environment is telling you something and is geared for one thing where mastering is something else. I have broken my own rules occasionally, and mastered projects that I mixed. I have some artists that I work with that say, "You know what? I don't care. You're doing this." And I say, "Well, I'd really like somebody else..." "I don't care. You're doing this."
So I will do that sometimes. But when I’m doing that, I certainly mix somewhere else. I don't master in the same place I mix. That's one of the very biggest problems of people who are mastering their mix. Not because they don't have any perspective mentally, but they don't have any perspective sonically. They're listening to the same thing. They are in the same place with the same speakers with the same amp and the same equipment that are all designed for combining different tracks.
But mastering is about listening to the bigger picture, and making a body of work out of a collection of songs. Rather than making a song which is a body of work out of a collection of instruments. The picture gets bigger. When the picture is bigger like that and you're listening in the same spot that you've put it together in, it's not...
Well, it's not going to change at all. That mixing engineer will just want it to be a little bit louder because he likes how he mixed it.
Right. And then here's the most common problem – it's too bright and it's too loud, or it's over compressed. They go in and they compress the final mix because they think it has to be louder and they think, "Oh, mastering is making it brighter and louder," when they're not really listening.
I mean, it's a learned skill to listen past content. I still can have a hard time with it. I'm a musician. And I'll tell you one of the things that I've noticed the most about playing instruments over the years from when I started playing to now, is that because of how I've trained myself to listen and to do my job and be effective for my clients, my playing suffers because I get wrapped up in sounds.
I'll be playing along with somebody and think, "Okay, this is cool." And then I'll think, "Crap, where are we?" Because I'll be thinking, "Wow, that guitar sounds great with that other guitar. Listen how those frequencies interact. And listen to this slap back, the back hall, and that singer has too much compression on their voice." And then I think, "Wait a minute. Is this the first chorus or the second chorus, I don't remember."
So it totally takes me away and it's really hard to separate those listening objectives sometimes. It's like pleasurable listening. I have a setup at home where I sit down on the couch and I listen to music. It's not generally on as I'm just making dinner, wandering around the house, talking on the phone or things like that. I respect the art form.
For me, it's something to be active with, not passive. But it's still hard to sit and dig in to pleasurable listening without being analytical. Or to be analytical without trying to flip over to the pleasurable side.
Because once you decide on these types of careers, you're always going to be analytical on every type of music, because you're always going to be listening past the pleasure zone, you know.
Yeah. Exactly, looking at ways to be able to be better, and to dig in further to be able to help my clients more.
You talked about mixing engineers over-compressing their own masters. If somebody came to you with a mix and wanted you to master it, what kind of dynamic range and headroom would you like their final mixes to be?
Well, okay, so dynamic range and peak level and then RMS are three different things.
What I would prefer as a level goes, is give me 3dB or so headroom. If it's already up to the top, I can't do much with it without turning it down first. So there's no reason for it to be a stick of butter on a workstation. Or the meters pegged on a tape machine.
As far as dynamic range goes, it needs to be appropriate to the music. I mean, like the analogy of the difference between analog and digital. With analog, you're pushing electrons that actually go through physical pieces of gear and create resistance and capacitance and heat and all those kinds of things that affect that sound sonically. Whereas digitally, you have ones and zeros and you're just either adding more or less. You only have a finite amount of information to work with.
It's kind of like a glass of water. If you look at every track, say you've got six tracks of drums and you're running through a workstation and you have all your information analogous to water in a glass, and you have 80 percent in this glass and 90 percent in that glass. You have all of that but you're creating a stereo drum bus, and so you have six glasses mostly full of water pouring into a glass the same size as the individual glasses, because the math involved is the same per channel as it is per bus.
So you're throwing away all of this information. All of the low level information like ambient cues, or overtones. So what people do is they compress it to get the difference between the high peaks and the low peaks, or high-level information or low-level information. So the dynamic range is less.
They think that you're hearing more information and then you combine that together, then you have to compress it more and limit it because you're now adding it together to where you would be throwing away all this stuff. And so everything becomes more and more compressed along the way. Now you do this with all of the other busses, like guitars, keys, vocals etc. to come out the master bus and you have to compress and limit it more to squeeze this thing out of it and when you get to the end there's no life in it.
We have the capabilities today with 24-bit words and higher sample rates to be able to have a tremendous amount of dynamic range, but very few people really use it. It would be much better to be able to use this information and let things breathe as much as possible. I can make it louder than the mix possibly could because, one, I'm dealing with two channels.
But the other of my tools are designed for those kinds of things where plug-ins on every channel that compress everything can't handle material the same way that a mastering engineer can and what they've got going on. So what I prefer that somebody brings me, and what works best, is something that breathes as absolutely as much as possible.
So as far as dynamic range goes, it just has to be appropriate for the type of music. You know, classical music is going to have a lot more dynamic range to it than a pop record. But, you know, there's a difference in density of material as well.
If you've got an acoustic guitar and vocal on one track, and then the next track is six guitars, three keyboards, four vocals and drums and all that, and it's the same artist, there's still the same amount of energy. And if one is really compressed to be able to fit into the same sized space, they're going to have such a different presentation to themselves that there's a lot of work to try to make them sound like they belong on the same album.
If we continue on the analog versus digital. Do you get all your tracks in digital format to be mastered? Or do you get them on tape too?
I still get them on tape. Less now – it kind of comes and goes in waves. But I would say I get a certain percentage of material on tape.
One of my pieces of gear that I use is an analog tape machine. I have clients and engineers that I work with that instead of them mixing some tape, they know that I can take it to tape in essence with better D/A converters than are coming out of a workstation. And I'll take it to a one-inch two-track machine. So sometimes it's a case that people do bring me tape. Sometimes it's the case that they know that I have the capabilities and that affords them the option of not having to mix to tape, knowing that I can.
But generally, people do bring me digital files.
Yeah. You just have that option to put it to tape. But do you put it through tape even if you get it in digital format? For a better sound?
Oh, no, not always. And it's not always a better sound. I do it if it's appropriate. Just because I have it doesn't mean I do it or I need to do it. Just because I have a salt shaker on the table when I have dinner doesn't mean I put salt on everything.
You know, it's there for a flavor if I want, but it's only because I think it's needed. Or with too much compression of final mixes would be like me needing to take flour out of a cake once it has been baked. Yes. I tend to use a lot of food analogies!
Yeah, well, like just the casual day-to-day analogies is what really explains complicated things the best. Like for somebody that wouldn't know all the technical jargon, you know.
Yeah, I think it does help with understanding and comprehension.
Stay tuned for the next installment next week as we further discuss the effects of the home recording revolution, the differences between mixing and mastering rooms and his experience working with creative rock bands such as Dredg.