UA's Art and Science of Modeling UAD Plug-Ins, Part 1 of 2

July 25, 2010 6:15:24 PM PDT
Dr. Dave Berners and Jonathan Abel
UA Chief Scientist Dr. David Berners (left) with
Stanford University’s Dr. Jonathan Abel

How Universal Audio became the world’s leading developer of emulation plug-ins.

For anyone observing music recording over the past 20 years, audio technology has evolved at a mind-boggling pace. Giant tape machines, consoles, and racks filled with gear have quickly given way to sleek desktops, laptops and even tablets. Yet while workstation technology has made some older audio equipment obsolete, the lure of certain vintage analog hardware endures, thanks to a palette of warm sonic colors that is difficult to achieve “in the box.” For professional engineers, there’s an undeniable appeal in using certain classic equalizers, compressors, and effects processors; the bad news is that such vintage hardware is often cumbersome, hard to find, and can be impossibly expensive. All of which explains why “modeled” analog emulation plug-ins hold such a major appeal in today’s computer-based audio production.

As the thinking goes, by accurately emulating the way in which classic effects processors add their characteristic “analog” details, we can bring sonic layers and textures from the great recordings of the past into our modern-day sessions. Of course, plug-in emulations like these do not just spring into being overnight. And just because an audio plug-in looks like a highly prized classic, doesn’t ensure that it actually sounds anything like it. Rather, highly accurate plug-in emulations are best created in an environment that gathers together some of the brightest and most creative minds in our industry, ranging from system designers and programmers to evaluators and professional users. In other words, an environment like Universal Audio.

A Unique Approach to Plug-Ins

At Universal Audio, we follow a philosophy of using well-practiced scientific discovery to produce the world’s best analog emulation plug-ins. “We believe — and, more importantly, can prove — that our experienced methodologies produce the most accurate hardware emulations possible,” asserts company founder Bill Putnam Jr. “A very important difference between UAD plug-in emulations and competitive offerings is that our engineers employ true physical and circuit modeling. Many of our competitors just rely on signal modeling or limited forms of circuit modeling. But in essence, we 're-build' analog hardware in the digital world. That way, we can better emulate the interesting non-linearities of classic hardware—the hysteresis, feedback loops, distortion, and other circuit elements that give analog units their characteristic, and much cherished, appeal.”

Because Universal Audio’s software development approach is innately more complex and time-consuming, UA’s plug-in algorithms require dedicated processing, in the form of the UAD DSP Accelerator platform (including the UAD-1 and UAD-2 cards). Based on Analog Devices SHARC floating point processors, the UAD platform powers a growing library of more than 50 plug-ins that deliver high-resolution “analog realism”— without taxing the host Mac or PC. In this way, users gain both high-quality analog sound, reminiscent of the classic gear of the past, and the ability to run many more instances than would be possible without dedicated processing.

"We believe—and, more importantly, can prove—that our experienced methodologies produce the most accurate hardware emulations possible.”—Bill Putnam Jr. 

Our History

Of course, Universal Audio is a company whose origins are engrained in both music and scientific discovery. Bill Putnam Jr’s father, Bill Putnam Sr., was a leading engineer in the 1940s, 50s, and 60s, who recorded and mixed an astonishing number of landmark sessions.

“Universal Audio was founded over 50 years ago as a natural extension of my father’s role as a recording engineer, studio designer, and inventor,” recounts Bill Jr. “He was a favorite engineer of many music icons of his time, including Frank Sinatra and Nat King Cole. He also designed and operated recording studios that were highly regarded for their acoustic qualities. These spaces provided a fertile environment for his innovations, including early modular consoles, mic preamps, artificial reverb, and the 1176 compressor. In simple terms, my father was a guy who built equipment to solve problems.

“In 1999, my brother James and I decided to re-launch Universal Audio,” continues Bill Jr., “and merged it with my then-new company, Kind of Loud Technologies, which at that time was a leading audio software company. We had two primary goals: firstly, to reproduce classic analog recording equipment designed by our father and his colleagues; and, secondly, to develop the world’s best digital emulations of classic analog gear.

“In essence, our plan was for the newly refounded Universal Audio to bridge the worlds of vintage analog and digital processing, and create a company in which musicians, analog designers, and DSP engineers could intermingle and freely exchange ideas.”

Product Manager Will Shanks and Dr. David Berners in the UA Studio
UAD Powered Plug-Ins Product Manager Will Shanks (left)
with UA Chief Scientist Dr. David Berners in the UA Studio

Academic Links with Stanford University

That spirit of scientific discovery extends to the close professional and personal relationships that Bill Putnam Jr. and key UA staff enjoy with Stanford University’s Center for Computer Research in Music and Acoustics (CCRMA). Putnam is a former CCRMA graduate student and is still involved at the center; UA’s chief scientist, Dr. David P. Berners, is also a former CCRMA grad, and regularly teaches a graduate class in audio effects processing. Dr. Berners has also held positions at the Lawrence Berkeley Laboratory, NASA Jet Propulsion Laboratory and Allied Signal.

“We put a lot of effort into maintaining a close relationship between Universal Audio and CCRMA,” says Bill Jr. “CCRMA researchers collaborate on a number of our projects — a process that allow them to reserve the right to publish and use the results — while UA secures the benefits of the research to use for product development. In this way, theoretical advances can be published without creating a blueprint for our competitors to clone specific products.”

Dr. Berners considers the close, day-to-day connection between Universal Audio and Stanford University’s CCRMA, "fantastic for developing and maintaining signal-processing products. The constant exposure to new research [and] the interaction with highly motivated and trained engineers and scientists… contributes greatly to our problem-solving mindset. Our regular immersion in that environment is the best benefit, over and above results from brainstorming sessions and the like.

"In essence, my plan was for Universal Audio to bridge the worlds of vintage analog and digital processing, and create a company in which musicians, analog designers, and DSP engineers could intermingle and freely exchange ideas."—Bill Putnam Jr. 

The Story Goes…

“My father passed away in 1989, and we ended up starting the company slightly less than 10 years after that,” recalls Bill Jr. “I always knew that I wanted to get into the audio business, specifically to exercise my passion for developing signal processing and technology. Because my dad often set up his lab at our house, I was able to get involved in product R&D. During this time I was exposed both to a number of engineers, as well as folks on the recording and production side who came to our house for A/B listening tests of new products. This exposure got me excited and gave me a head start in product development.

“When my brothers and I finally sold the family home back in the late '90s, we were faced with the daunting task of going through mountains of documents as well as equipment,” he continues. “It was this emotional event, almost a decade after our father had passed, that was the catalyst to getting back into the audio business.”

The immediate plan was to develop both analog replicas of classic effects, as well as plug-ins for digital audio workstations. “At that time, there were passionate and heated arguments between analog and digital aficionados,” Bill Jr. recalls. “I saw the benefit in both approaches, and felt there was an opportunity for a company that embraced excellence in these complementary areas. Up until that time, and through today, companies tend to align themselves with one side or the other. I believe that UA holds a unique position in this regard.

“Digital systems offer major advantages in workflow and efficient operation,” he continues, “while analog offers warmth and an intrinsic link to the great recordings of the past. Having come from an academic environment focused on physical modeling, I didn't see why we couldn’t bring the beneficial aspects of analog processing to the digital world.”

Putnam’s planned Ph.D. thesis at Stanford University was centered on optimization theory and its applications to audio problems. “Some of the application areas were designing audio filters that had been optimized in a psychoacoustic sense, as well as loudspeaker and microphone processing. My interest was focused on emulation and physical modeling of analog systems," he explains, "a perfect adjunct to our pursuit at Universal Audio on modeled plug-ins.”

To this day, UA provides funding for research at CCRMA, a connection that has a strong impact on the quality of research the company undertakes. “Universal Audio has been an affiliate for years, pretty much since the company started. While I was studying at CCRMA, much of the technical research by several Ph.D. students and faculty was on physical modeling of acoustic instruments. While this isn’t exactly the type of emulations we now do, philosophically they are very similar. My exposure to the thought processes and techniques was very influential in moving towards physically modeled emulations of classic analog equipment. So, in many ways, my time there set the stage for the future directions and success of Universal Audio.”

Check back with Ask The Doctors next month for Part 2 of “The Art and Science of Modeling Plug-Ins”

— James Douglas