Bob LudwigBob Ludwig has long been one of the pre-eminent mastering engineers in the U.S., respected by both musicians and his peers. He's mastered albums by countless 12/01/2001 7:00 AM Eastern
Bob Ludwig has long been one of the pre-eminent mastering engineers in the U.S., respected by both musicians and his peers. He's mastered albums by countless major (and minor) groups in every style. Though he was a fixture in the New York mastering scene for many years, for the past several, he's been working out of his own facility, Gateway Mastering, in beautiful Portland, Maine, and he's still attracting many of the best in the business.
Traveling to Portland to interview Ludwig, we drive through stunningly beautiful rural areas, where the highway exits are few and far between. The weather is typical New England extreme, changing from sun to rain in a split-second. The cobblestone streets of the old downtown lead to a picturesque harbor where people gather for lunch or stroll around looking at the many fishing boats that reside here. It's definitely a small-town feeling.
“You can even leave your car running while you step inside a store,” Ludwig says once I arrive at his impressive studio complex, which consists of two immaculate control rooms, a room for DVD authoring, a DVD screening-room, a production room, a gear storage room, a recreation room with a pool and a 52-inch TV, and a few office spaces. A staff of 12 helps Ludwig maintain what is obviously a thriving business. “We have plans to build another mastering studio soon, in order to take on more independent artists at a lower rate,” he says. The walls are covered with Platinum-selling albums, many of them considered classics. He has also received 14 TEC Awards, in addition to the Les Paul Award he was given in 1992. Despite all of the recognition and the extraordinary success he's had, Ludwig comes across as a regular guy, albeit one with a deep understanding of pro audio that extends well beyond his part of the process.
Why did you move from New York to Portland?
I moved to Portland in 1993 for the simple reason that I have a family up here, and I figured if I were to invest that much money into my own recording studio, it better be at a place that I liked. After being VP at Sterling Sound and a long time at Masterdisk, I felt I didn't have a real influence over my career. There were a lot of limitations on what I could do, where I could live and so on. I realized it was time to start my own business. Not to mention the quality of life, which is much better here in Maine — less crime and a cleaner environment.
I have to admit, I was a bit skeptical about whether my clients would choose to travel several hundred miles north of New York, where a lot of my clients are based. However, most of my clients from my Masterdisk period are still with me, and, surprisingly enough, more often than before, we get people coming to sit in with us through the sessions. We've got clients coming from Europe, Russia, Japan, Argentina and London, and I think the reason is that people have heard about our control room and want to hear it for themselves. I don't think there are many mastering rooms like this one around the world.
In addition, Portland has a modern international airport, a prerequisite for being in the business we are in. We often work with short turnaround times due to the fact that we are last in the production chain. Most producers would like to have the job done by yesterday, so we are totally dependent on being able to deliver and receive material without delays.
How did you get into the mastering business originally?
I was about to finish my Master's at Eastman School of Music, where I also was involved in the sound department, when Phil Ramone came to teach a summer recording workshop and I worked as his assistant. Afterward, he contacted me and asked if I wanted to come to work for him in New York. Phil had just started A&R Recording at the time, so we started doing sessions together on projects with The Band, Peter, Paul & Mary, Neil Diamond, Frank Sinatra.
After a few years with him, learning the art of mastering, I got an offer from Sterling Sound. They had an awful lot of nice gear. For instance, I remember their advanced Neumann cutting machine with the SX-68 cutter heads — a huge improvement over what we used to have. They also had some tape machines from Telefunken and Studer, and I always dreamt about working on world-class gear. I knew that no matter how good I got, I would never be able to compete with places that had that kind of gear, so leaving A&R wasn't too hard a decision to make. After seven years at Sterling, I was offered a position at Masterdisk, where I worked until I decided to start my own facility here in Portland.
How do you approach a new project?
The secret to being a good mastering engineer is to be able to listen to a mix to hear how it could sound, and finally be able to push the right buttons to achieve the sound you have in your head. When I start a new project, I first listen to find out if there is enough vocal in the mix and whether the EQ is correct. Too much vocal in a mix makes the music uninteresting, and too little is annoying to the listener. People want to hear the words. At one magical point, something happens that makes the vocal blend perfectly. I work pretty fast at finding the right EQ and compression that keep the vocal in place, and I spend longer time refining the settings.
It is important to set aside time to listen to parts of the whole album before starting to work. Very often, the A&R people have placed the hit songs at the beginning, and sometimes you get the impression it's a pop album you are about to work on, where as if you listen further, you'll find that the music is more artistic and deserves a different approach.
Usually, I master an album in one day, which, on average, gives me one-half to one hour per song. Under certain circumstances, I have to ask the song to be remixed. In cases where the vocal is too bright and the instrumental tracks are too dull, there is nothing I can do. I do not listen [to the track] loudly when I work, and by loud I mean more than 90dB SPL. I try to keep it around 85 dB if possible.
Can you tell me a little about the control room and how it was designed?
To me, the mastering stage is not the place for surprises, so my philosophy was to build an ideal listening environment, acoustically speaking, that would reveal any limitations with the source. Together, with Dr. Peter d'Antonio from RPG Diffusors, we designed the studio piece by piece. He has developed his own software to mathematically calculate room dimensions in order to minimize bass buildup due to standing waves. A standing wave is a wave whose half-wavelength is corresponding to one of the three room dimensions. The ratio we arrived at was based upon a complex relationship between height, length and width that gave the least bass accentuation. That same ratio also meant that our room had to be really big. Our room has the highest standing wave at 19 Hz, which, again, meant we needed 16-foot ceilings and 32-foot sidewalls front to back.
The control room is floating; it's a room within the room, completely isolated from the rest of the building, and we used seven layers of Sheetrock in walls and ceilings. D'Antonio's idea was to build a room with an acoustically absorptive wall at the speaker's side, and an active rear wall opposing the speakers. To prevent damaging reflections from the rear, we needed to spread the sound in random patterns. He made the world's first third-generation diffractal, which means that the geometrical patterns are being repeated in three different layers, from macro level to micro level, from the way the bigger elements are arranged on the wall, to the way the fibers are being organized. On the ceiling, we put in another RPG device called Flutter Free. The purpose is to disperse the ceiling reflections from the tweeter and throw them into the rear-wall diffractor. The room is otherwise like a regular control room. All noisy fans from equipment are securely kept behind glass doors in cabinets, and we have a low-velocity AC.
What does your monitoring system consist of?
The speakers are not mounted but are resting on spikes sitting on top of a concrete base, which, again, goes down to bedrock, deep down below the floor. This is for preventing resonance from the floor, [which would] compromise our listening environment. The placement of the speakers was calculated down to the very inch based upon the room dimensions in reference to the angles of the absorption and diffraction material, measured from an ideal listening placement. In a mastering environment, I believe in having the highest-resolution speakers possible. We settled on Eggleston Works “Ivy” speakers [www.egglestonworks.com/ivy.htm].
I don't claim to have listened to every speaker that's available, but for what I do, those were the best pair I've ever heard. The speakers are three-way with one tweeter. The physical dimensions are impressive: 6 feet in height and weighs 89 pounds each. The upper surface is solid granite. For powering the speakers, I use two bridged Cello Mark II amplifiers. One is connected to the positive pole of the speaker, and one is connected to the negative. We did this to increase the slew rate, which is the speaker's ability to reproduce high transients. In our case, we have 4,000W, although I don't think we are ever going to be listening that loud!
Are there certain things mixing engineers can do in order to make sure you can achieve the best final result?
It is important to give us different versions of each song, especially where the vocal is at different levels. Some people, like Bob Clearmountain, give me final mixes with the vocal level at ± ½dB and +1dB. When you are in a mix and know the lyrics by heart, it is hard to objectively evaluate exactly where the vocal should go in the mix. Sometimes, the album producer decides which take I should go for, and other times I have the liberty of choosing myself. A lot of these people have worked with me for years, so there is always this element of trust involved.
One other thing to avoid is the excess use of compression. I am really glad that the new digital compressors weren't around when The Beatles were making records. It is a common belief that music that sounds louder sounds better and more exciting, especially when you are listening to it for shorter periods of time. However, it is my experience that a louder album will wear you out faster than an album that has greater dynamics, and it might make you not want to listen to it again.
Compression is also something that you can never undo, unlike EQ, which we always can tweak later. If you are in doubt about how much compression to use, the advice is to use less and to let us do the job. The studio monitors today are generally a lot better than what they used to be, and from a mixer's perspective, it is easier today to get the EQ right. The best thing is to get the EQ as close as possible in a mix environment, and let us only do some touch-ups. This obviously requires that you are familiar with your speakers and the room you are working in.
You started mastering vinyl. Is that something you are still doing?
Until recently, I had a DMM [Direct Metal Mastering, where the album is cut to a copper disc] system at Masterdisk, and at Gateway a lacquer-cutting lathe, a VMS-80. We sold our machine to Sony a few months ago, and the reason was that I was gradually getting more and more disappointed with the level of quality control that took place at the record companies. I realized that the A&R people didn't have record players anymore in their offices, and the amount of work we were getting was gradually decreasing to maybe one lacquer-cutting session a month.
The machine always needed a lot of calibrations and tweaking in order to perform at its best, so it wasn't really justifiable to keep the machine around any longer. Although we don't do vinyl mastering any more, we are still considering doing it again if there is a demand for it in the future. Masterdisk still has their DMM machine, and, as a side note, I can mention that Bruce Springsteen's Live in New York City was mixed by Bob Clearmountain on a DA-98 with Apogee PSX-100 converters and was mastered here at Gateway. Masterdisk, in turn, used our 24-bit master to master the LP using DMM. We often get mixes at higher resolutions: 88.2k, 96k, 176.4k, 24-bit, and for CD we work at high-resolution before converting to Red Book standard. We utilize Apogee UV22 or Pacific Microsonics HDCD dithering in order to reduce to 16-bit resolution.
It is customary to believe that the CD is superior to the LP in terms of bandwidth, but this is not the case. The CD is limited to 22,000 cycles, whereas the LP is able to reproduce frequencies up to 50,000 cycles, which in the PCM world equals a sampling rate at 100 kHz. The bottom line is that LPs mastered with DMM still sound really good.
Which platform are you presently working on?
We basically use Sonic Solutions [USP and HD]. The reason for landing on Sonic eight years ago was that it was the only viable machine at the time. Later, we acquired two SADiE Artemis systems and two Pro Tools|24 machines. A major part of our stereo productions takes place using Sonic, mainly because of the networking capabilities. All of our five Sonic USP machines are tied together using the MediaNet network, and we can access the projects and play or record to the disc drives from all machines.
On a project with Laurie Anderson that we recently worked on, we had some problems with mouth noises and had to work in parallel in two different studios. One of my staff was de-clicking the vocals using the manual de-click program, while I was working on EQ, compression, etc., in the main room. I could then go back and continue working, never leaving my mix environment, with the ticks removed. Sonic is the only system that lets you do that, and it makes for very efficient mastering.
We are also one of the few studios that does DVD-Audio authoring [192k]. Unlike DVD-Video, which has a graphic user interface, the authoring software for DVD-Audio is on code level, which makes it extremely difficult to do. One wrong spelling or an extra space between letters could introduce a functional error, so you have to be really careful and accurate during the process.
At the moment, Sonic is the only company selling authoring software for DVD-A, so we are totally depending on them to develop a more intuitive software in the future. Recently, there have been a few other companies that will offer simple DVD-A authoring, like SADiE, and soon one in Germany will offer a complete system. For recording 6-channel DVD-Audio 96k/24-bit, we are really happy with the SADiE software, which has proven to be rock-solid.
A lot of the projects come in on Pro Tools|24 on SCSI ultrawide discs. In terms of multichannel external processing gear, we purchased the TC Electronic System 6000. This unit has an extensive number of presets and is extremely flexible. It has five channels of reverbs, chorus, EQ and compression, in addition to 3-band compression/limiting, solo/mute, and the list goes on forever. Apart from a Pro Tools plug-in called RealVerb 5.1 from Kind of Loud, this is the only processor that allows five discrete reverbs from five dedicated inputs. Lexicon, Yamaha and Sony have only four for some reason.
For high-resolution audio, we use converters from dCS, Apogee and Pacific Microsonics, the latter for HDCD. These converters are able to handle 96k. The Pacific Microsonics and dCS can also do 192k, but the dCS is the only one able to handle Direct Stream Digital [DSD] for Super Audio CD [SACD].
Although you can get the impression that it's all about digital, we still have a huge collection of analog gear such as EQs from Manley, Massenburg, the Sontec 430C, Neumann, Avalon, in addition to Millennia, Manley and NTP compressors. We have digital and analog mixers such as the Harmonia Mundi 24-bit and a special George Massenburg/Sony Music level control. In terms of analog 2-track, I'm pretty proud of my 1-inch tape machine from Esoteric Audio Research [EAR] developed by Tim de Paravicini. In addition to the standard playback correction curves CCIR and NAB, Tim has included his own called “Tim's Curve” [TIMZ]. At 15 ips utilizing TIMZ, this machine has a flat frequency response from 8 to 28,000 Hz! Pretty impressive, if you ask me.
While we are in the analog domain, I would like to add that I think analog lends itself really well to pop music. The unlinearities added in terms of compression and harmonic distortion are, in many cases, desirable, unless you ask all-digital people like Bob Clearmountain, who doesn't at all like the fact that analog machines aren't reproducing what you are feeding them. [Laughs.]
With all this great gear around, how do you decide what to use on different projects?
If we receive an analog mix, we usually stay analog until we transfer ultimately to PCM 1630, so the receiving format by itself is a determining factor. Analog equalizers sound to my ears much better than most digital gear, because they have their own sound. I'm obviously not opposed to breaking my own rules if the project calls for it, and I sometimes run my digital signal into my analog devices for tweaking. The main rule, however, is to avoid more than one A/D or D/A conversion, which is the weakest link in the chain. The right placement of A/D in the chain is of great importance to the final result.
What is your relationship to MP3, and in what way has the advent of MP3 changed the way you are working?
We perform MP3 encoding at Gateway, usually to send a song to a producer to get his/her approval on a downfade, the spacing between songs, etc. We never use it for anything of master quality, however. The most important thing, regardless of format, is to work on high-quality monitors and to know how they sound.
When you're working on great equipment, making it sound as good as possible in that environment, it is my experience that your audio will translate well into cheaper playback systems such as boom boxes, computer speakers or MP3 players as well. The problem with MP3 is that you never really know how it's going to sound. On certain types of music, it sounds okay, whereas on other stuff it sounds disastrous. We did some MP3 encoding for the Paula Cole homepage, and even at the highest MP3 bit rate, her voice sounded harsh and unmusical. On the other hand, a very dynamic classical production we did was, to my surprise, acceptable.
MP3 has changed the production chain in the way that mastering no longer is the final stage before the consumers. Music distribution is now changing how the audio is sounding, and beyond the control of audio professionals, a lot of things can go wrong with the audio. There are a wide variety of encoders floating around on the Internet and they are all different in terms of quality, so it makes it impossible for us to take MP3 into account when we work on our mix. We focus on our job here and now, creating the best-sounding album possible without focusing on what's happening to it as soon as it hits the Internet. This is not a new phenomenon after all. Music-delivery channels like FM radio stations have always utilized heavy compression and EQ to create their own “sound,” if you like. When I hear a song I have worked on, on the radio or in a shopping mall, it's hard to recognize the mix sometimes.
Is there a problem that the consumers are unwilling to change their playback systems in order to keep up with the demand for improved audio quality within the professional audio community?
It is always a problem that the consumers have a system they are sort of happy with and don't see the reason why they should buy something new and even better. Some people are even happy with Internet quality! However, the audio industry has a problem in conveying knowledge about the various formats that exist to the consumers.
Personally, I think consumers today are getting to be more concerned and familiar with the concept of surround sound, more so than increased resolution and higher bit rates. We see that surround decoders are starting to become more and more common in people's homes, and the sales of DVD players is rocketing. In hi-fi magazines like Stereophile, the Super Audio CD has gotten rave reviews, so this new audio format is probably going to be more common in the future, as well. At the moment, there aren't that many titles available for SACD — around 300 maybe — but as soon as we get more titles, the demand will increase, the same way it happened with DVD-Video
What are typical problems in the receiving material, and what are the most common formats?
Believe it or not, wrongful or the lack of labeling is a recurring problem. Correct labeling is particularly important for surround mixes. One should use the DTS standard: L, R, LS, RS, C, LFE, or ITU/SMPTE: L, R, C, LFE, LS, RS.
We also see a lot of timecode drop-outs, timecode discontinuity, wrong reference tones, wrong sampling rates, etc. These things should not occur, but they do over and over again. Another problem we encounter is projects with distortion. You'd be surprised to find out how many of our projects contain distortion beyond what is considered acceptable — even major productions. My clients often come in and hear distortion they've never heard before when they take their projects to Gateway. The reason for this is that I think people are listening too loud and/or on low-resolution near-field monitors during mixing, such as the NS-10.
We also get projects where the EQ balance is all over the place. From listening to a mix where the bass is almost absent, and to having a mix with ridiculous amount of subsonic information, it's sometimes beyond me how some of this even got through the mix.
Sixty to 70 percent of the projects are still analog 2-track quarter-inch or half-inch. The new analog format 1-inch 2-track is also starting to pick up. The single “Arms Wide Open” by Creed was the first Number One single that was mixed to 1-inch. We also receive DATs, Pro Tools, SADiE and Sonic Solutions sessions.
For surround, we see a lot of DA-88/DA-98s, but also 2-inch 8-track. For stereo delivery to the CD-production plant, we still trust our Sony 1630, although we sometimes send a CD-R if demanded by the record company, but I always get very nervous when that happens. Another advantage with the 1630 is the automatic error-counter we have and that the glass mastering is taking place in real time, unlike the CD-Rs, which are glass-mastered at 4x. The Genex 8500 MO might become the new standard production format, since the 1630 is hard to maintain. We haven't acquired the Genex machine yet, but we are considering it in the future. The surround music is sent to the DVD plant by DLT.
Is there anything top engineers have in common that makes it easier to achieve maximum results during mastering?
It is hard to point out what my favorite engineers have in common. To name a few, Joseph Puig, Hugh Padgham, Tchad Blake and Bob Clearmountain are individuals with totally different approaches to audio, but they are all great engineers.
The way I see it, recording and mixing audio involves an indefinite amount of compromises, and the really good recordings are made by people who have evaluated every step in the recording process carefully, and have chosen the best compromises. What it all comes down to is everyone's ability to use his or her ears.
Øystein is an Oslo, Norway-based musician and recording engineer for the Norwegian Broadcasting Company and freelance music projects.
Bob Ludwig was a pioneer of Super Audio CD and has been a major proponent of the format's widespread acceptance. We asked him to share his views on the fidelity and the merits of SACD.
“The Sony/Philips Super Audio CD is a relatively new  format with a bandwidth up to 100 kHz, sampled at 2.822 MHz. The format is dual-layer, which means a 6-channel surround mix and a discrete SACD stereo mix are on one layer and a CD stereo mix reside on a different layer. Thus, the stereo PCM mix can be played back on a DVD player. All SACD channels can have full bandwidth [6.0], but the channel format is optional, being 6.0, 5.1, five channels, four channels, three channels or two channels. Unlike a PCM-encoded signal, the word length is only one bit, and because of the extreme sampling rate, the output of the converter yields a pulse close to an analog waveform.
“The output of the converter is called Direct Stream Digital [DSD], and is considered to be as close to analog as you can get in the digital domain. The format has a dynamic range of 120 dB, which corresponds to 20-bit in the PCM world, in theory. In order to fit on a disc, the signal is encoded using lossless encoding from Philips.
“The first DSD 6.0 surround pop album was done here at Gateway, the Guano Apes on BMG. We have great hopes that this will be a format for the future, and I really like the way it sounds. For SACD, there is no digital processing gear, which means that most DSD mastering has to take place in the analog domain. Sony is presently working on their digital editor, Sonoma, which will make us able to use Direct Stream Digital EQ, compression, limiting and mixing within SACD as well.” [Editor's note: There are already a couple of Sonoma systems in use, and getting rave reviews from users.]
— Øystein Eide