Alvy Ray Smith: RGBA, the birth of compositing & the founding of Pixar

This article is a companion piece to our exclusive fxpodcast #222.

Dr Alvy Ray Smith (photo by Kathleen King, copyright)

And then officially I joined New York Tech. I was the old man at 31; Ed and David were just barely 30 and 26 respectively. Thus began the NYIT – Lucasfilm – Pixar computer graphics dynasty, a marriage of the house of Xerox and the house of Utah, pixels and geometry, art and technology. The movies we dreamed of then – completely generated on computers – was shown in November 1995, 20 years later. Toy Story was that movie.” – Dr Alvy Ray Smith referring to Dr Ed Catmull and David DiFrancesco.

Dr Alvy Ray Smith has quite the track record. To summarize just a few of his many achievements reads like more than three lifetimes.

Dr Alvy Ray Smith helped or personally invented Paint as we know it, and the Alpha channel (the very idea of RGBA – and he picked the name Alpha). He directed the Genesis sequence in Star Trek II: The Wrath of Khan while at Lucasfilm, his first directing gig ever, and then followed it up with directed the first short The Adventures of André and Wally B. That film would become the prototype for the short film tradition that would follow until today in a company called Pixar, which Smith co-founded with Ed Catmull. Ed is a friend, who Alvy pushed to hire a young animator called John Lasseter.  Smith sold technology to Disney, helped Steve Jobs invest half his fortune in Pixar ($50M), and then helped make him billions, finally coming to blows with Jobs over a white board. However, Jobs still invested in Smith’s next company which Smith would sell to Microsoft, making Jobs even more money. Along the way he worked at Xerox PARC at its height, inspired Photoshop, defended Adobe and lost to Quantel. It is a remarkable story.

Mike Seymour got to talk at length with the man himself in this week’s fxpodcast. This in-depth discussion is a celebration of not chasing the money, but succeeding by passionately pushing to develop the art of computer graphics to tell stories.

Xerox PARC, NYIT and Paint

Dr Alvy Ray Smith’s story begins perhaps at Xerox PARC, the most famous research private facility in the history of computing. Founded in 1970 as a division of Xerox Corporation, PARC has been responsible for such well known advances as:

  • Laser printing,
  • Ethernet,
  • the modern personal computer,
  • graphical user interface (GUI),
  • the mouse as we know it,
  • object-oriented programming,
  • advancing VLSI for semiconductors.

It was the conceptual home of the Apple MAC/LISA, which adopted many of the principles of computer design invented or refined at Xerox PARC. PARC’s developments in information technology have had great long-term impact, but one of its lesser known roles was in effects paint.

8 bit paint, 1975 by Paul Xander. Original was 512×486 pixels.

The history of Paint, Superpaint, AVA, Aurora, Paintbox and Photoshop is the birth of 2D compositing as we know it today. Nuke, Shake, Flame, Henry and Harry were all built on the building blocks of research that certainly can be traced back to Paint. Today any computer can provide digital painting but it all started in the 1970s, and its story is the story of all the 2D packages that would follow.

Richard Shoup (today).

Richard (Dick) Shoup was one of the first employees at the Xerox Palo Alto Research Center (PARC), where he spent the following decade researching computer graphics and animation, digital video, and theories of computation. At Xerox he built one of the first digital frame buffers and developed painting software for graphic arts. The resulting ‘SuperPaint’ system now resides in the Computer Museum History Center collection at Moffett Field in Mountain View, California. SuperPaint was an 8 bit paint system

 

 

In recognition of his work in computer graphics, Shoup would go on to be awarded an Emmy by the National Academy of Television Arts and Sciences, Technical Oscar Plaque, and a Computer Graphics Achievement award by ACM SIGGRAPH.

Right is an image of Tom Porter, Alvy Ray Smith, and Dick Shoup being honored at the Scientific and Engineering Award from the Academy of Motion Picture Arts and Sciences for their work on Paint Systems in 1988.


But in 1973 no one dreamt of Oscars or Emmys – it was a challenge to even get an image on a screen. Given the ubiquitous nature of images today, this era of only live imagery or 2″ tape playback with no way to freeze a frame, display a picture of any type is almost impossible to imagine. There were no computer screens, the only TV anyone watched was broadcast live (or from tape), with no graphics other than live mixes in the studio, there was no GUI on anything – of any type – a world unable to put up a frozen image unless you printed it multiple times and played the film back with the frame repeating.

Then came frame stores…

The Xerox Framebuffer:  16 cards held 2K bit shift register chips totalling 640 pixels x 486 pixels in 8 bits. 
The very first image captured on it (R. Shoup self portrait).

Above right – “This is the very first picture captured in the SuperPaint frame buffer when it first came to life in April, 1973.  The system took a standard video signal as input, digitized it to 8 bits, and could capture or combine it with other data. To take this picture, I pulled a cliplead off of the backpanel using my knees. The card says ‘It works! (sort of)’. At this moment the interface to the Nova 800 CPU had not been debugged or even plugged in to the system yet. In order to preserve this picture, I had to plug the interface card in and get it working with the power on.  Later, I was able to clean up some of the missing bits with a heuristic program and some fiddling.”  – Dick Shoup


Shoup left Xerox in 1979 to co-found Aurora Systems, a manufacturer of graphics and animation systems. While serving as President and Chairman of Aurora, he continued as designer of two generations of the company’s graphics systems, including PC and workstation-based software packages, and user interfaces. Left is a shot of the Aurora Paint system.

 


Today it is nearly impossible to understand how huge a deal a frame buffer was to the graphics community in particular. To give one an idea of the significance, at the University of Utah, prior to the frame buffer, the Computer Graphics team had to do a long exposure of a Polaroid aimed at a monitor, which would line by line display just part of the picture. Only when the whole series had been time-lapse combined into one image, and the Polaroid was developed/processed, could a team member like say Ed Catmull see the image he had been working on all at once. (The exception to this was vector displays that ‘drew’ the image on a vectorscope, at much lower fidelity and complexity). It is sobering to think of this…to Pixar’s Toy Story in a quarter of a lifetime.

Ed Catmull in 1995 on stage.

The story could be traced back further to the University of Utah, the real birthplace of computer graphics. Ed Catmull was an alumnus, as was Jim Clark (SGI, Netscape), John Warnock (Adobe), Alan Kay (Apple Fellow) and so many others. One of the other University of Utah key players is James F. Blinn, or Jim Blinn as he is known to the computer graphics community. Blinn also wrote an 8 bit paint system in 1974 for the Evans and Sutherland Frame Buffer at the University of Utah. Blinn is a towering figure in his own right, both physically and as a pioneer in computer graphics, inventing such things as environment mapping and bump mapping and being behind the famous JPL Voyager spacecraft fly bys of Jupiter and Saturn animation that would inspire his friend Dr Alvy Ray Smith’s storyboards for the Star Trek Genesis sequence years later.

Blinn, like many others, would also migrate from Xerox PARC to the New York Institute of Technology. Although Blinn would only be there a brief time, he would pop up time and time again, with Smith at NYIT, then Lucasfilm Computer Division, Pixar and finally Microsoft.

Smith moved to New York with David DiFrancesco (the two would stay working together for years, and DiFrancesco is still at Pixar today as senior research scientist in the Pixar Research Group).

Up until this point all paint systems had been 8 bit, but at NYIT ‘Uncle Alex’, Dr Alex Schure, would end up buying a range of Evan and Sutherland frame stores (extremely, extremely expensive back then) and hired Dr Ed Catmull in 1974 to run them. Smith would join Catmull and become friends.

Ed Catmull wrote a 2d animation program ‘Tween’ first in PDP 11 assembly language (re-written later in C), at NYIT which would be the forerunner of the CAPs animation system Pixar would sell to Disney years later. Schure had a great interest in animation, and was working on a project with an animated feature called Tubby the Tuba. Frustrated with the slow progress, he had been looking for tools that could help speed up the production – but as it would happen nothing from Smith or Catmull would even make it into the film.

Smith left Xerox PARC when the company decided to not follow color, they wanted to stay black and white. While Smith argued with the decision, Xerox was adamant that their future did not involve color imagery, and they fired Smith, as he revealed in a 2005 interview.

In 1975 Smith, who had written an 8 bit BigPaint system at NYIT, discovered while he was out of the office that Schure had sold the system to Ampex, then bigger than Sony in broadcast. Ampex did not have enough actual cash to be able to afford to buy the paint system which would go on to become the AVA, so, they traded “a PDP 11 computer to run it on and some peripheral equipment I don’t recall for I think an HS 200 disc recorder,” says Arnold Taylor, who was at Ampex at the time and would become the AVA’s product manager.

Note the HSB (Hue/Saturation/Brightness) SuperPaint menu, by Dick Shoup

Smith would travel to Ampex’s headquarters in Redwood City, California to install the software and it was while there alone one night, bored in a hotel room, that he conceived of Paint 3, the world’s first 24 bit RGB paint system. This used three of the expensive E&S frame buffers in parallel. There were many advances at this time. For example, Smith decided a better way to define colors for artists was not RGB but Hue/Saturation/Value, thus inventing HSV or HLS as we know it today.


NYIT, top right is where Smith met Catmull for the first time.

The year was now 1977 and while sitting around one night with Ed Catmull, who was complaining about the annoying aspect of having to re-render whole frames, Smith and Catmull then invented the Alpha channel, Smith staying up all night so that his friend Catmull could use it in the morning. He named it Alpha “as we always used greek letters for things,” he says.

The integrated RGBA is of course central to computer graphics to this day, but the team rarely stopped back then to document anything and certainly rarely were involved with patents.

In 1978 there was one exception to this lack of documentation. Smith had agreed to give a tutorial or talk at SIGGRAPH ’78 which was being held in New York.  This lecture would be the basis for several years of successive SIGGRAPH courses and a key point in later litigation.

Alvy Ray Smith and Ed Emshwiller at NYIT. The floating exposure of the monitors here is due to a bumping of the tripod during a double exposure! Note the Paul Xander picture on the wall.

But the point of all this technology was to tell stories and in 1979 Alvy Ray Smith started on what he considers to this day to be the most important and rewarding creative exercise, the Sunstone video. Artist Ed Emshwiller came to NYIT and made an art piece that today forms part of the exhibition at the Museum of Modern Art in New York.

Emshwiller had a grant to make a film and approached the team at NYIT. “He showed up at our door,” says Smith, “and I said I have a Guggenheim fellowship in 6 months and I would like to turn out a 3 hour movie with you fellows. We laughed. I said, ‘In 6 months you’d be lucky to get 3 minutes not three hours!’ and he said ‘OK, I’ll change my ideas then!’, and he and I then went into a partnership that is the most important I think of my life, artistically. He was definitely the artist and I was the technician.” The result of this close working relationship was Sunstone, which we link to below.


Copyright NYIT.


Lucasfilm, ILM and Paramount

After implementing the Paint system AVA at Ampex, with the team there, Smith moved to Lucasfilm’s Computer Division. Tom Porter knew the Paint code very well, so Smith hired him away from Ampex to implement a paint system at Lucasfilm. Smith joined Lucasfilm at the beginning of the 80s, as Director of Computer Graphics Research and between himself and Ed Catmull they assembled an amazing team. In addition to Porter, there was Mark Leather, also ex-Ampex who would write LayerPaint later on the first Pixar Image Computer, plus David DiFrancesco, Loren Carpenter, Rob Cook, Bill Reeves and many others.

A ‘graftals’ plant made by Smith in 1983. Grass made with fractal style particle systems. (copyright Alvy Ray Smith).

Porter wrote the 32 bit paint system that the team would use in 1982 for the Genesis sequence in Star Trek II The Wrath of Khan. It was not used as has widely been reported to move a fractal mountain that grew in front of the lens – that was manually edited out by – but it was used to texture a receding planet. This was the first use of a digital paint system in production at Lucasfilm and it happened – surprisingly – because almost no one at ILM knew much about computers!

Ed Catmull was Director of the Computer Division at Lucasfilm and Alvy Ray Smith was Director of Graphics Branch, but at the time of Wrath of Khan, neither were making films nor even contributing to any – the computers that were being used were involved in controlling motion control rigs but there were no direct graphics required for The Empire Strikes Back, for example. Yet there at Lucasfilm was the most incredible array of talent. Lucas wanted computerization in editing, audio and video (laserdisc) but not graphics per se. At this time the team included Bill Reeves, Tom Duff, David DiFrancesco, Tom Porter, Loren Carpenter, Rob Cook, and even Jim Blinn briefly. A dream team if there ever was one.

Their big break at Lucasfilm came not from director George Lucas but Wrath of Khandirector Nicholas Meyer, who expressed “a desire to explore some spectacular effects using this new fangled technology called computer graphics,” recalls Smith. “Luckily for us although ILM did not know computer graphics, one of its employees used an Apple II computer (this is still 3 years before the launch of the Mac), and was therefore aware of us computer folks next door. I got a call to go next door for a meeting.”

At this time Smith had never directed a major sequence like this, he was a complete unknown to Paramount Studios. “I had just been given the opportunity to design 60 seconds of this film. Complete novices never get such a chance. But I had it and I knew it. I worked sleeplessly all night on a concept employing everything I had one hand – principally an extremely talented team, including Carpenter and his fractal mountains, Reeves and his particle systems, Porter and his paint program, the 3D rendering skills of Cook and Duff and the digital filmmaking expertise of DiFrancesco. I was still under the influence of Blinn’s Voyager fly bys of planets (JPL). I threw together and sketched a six-card storyboard. ILM accepted my ideas completely. This was my first directing job for the big screen, and our first real movie job, a very lucky break,” Smith wrote in a 1998 IEEE article.

The piece would become a landmark in computer graphics, and awaken interest from Lucas himself, exactly what Smith wanted all along. After all, Smith designed the piece to catch Lucas’ eye, knowing that the filmmaker had a passion for interesting camera work. “I knew one secret thing about George,” says Smith. He doesn’t watch movies like the rest of us. He is really aware of the camera and what it is doing”.

Smith calls the Genesis sequence “a 60 sec commercial for George Lucas so he knows what he’s got.” Smith designed a camera move to ‘blow his socks off’. It worked, a fact borne out when Lucas stopped by Smith’s office after the premiere and quietly said, “‘Great camera move’ – and then whoosh he was gone. And then we were in his next movie, sort of, but more importantly we were in his buddy’s next film.” Smith is referring here to Steven Spielberg: the Porter paint program would again be used in anger when in 1985 Amblin Productions made The Young Sherlock Holmes.

1984
The original Pixar 1984 motion blur breakthrough (copyright Pixar).

This 1984 image was key as it demonstrated that the team had solved motion blur, a key issue and one that allowed the Lucasfilm team to also make The Adventures of André and Wally B.

André and Wally B. was originally going to be called ‘My Breakfast with André’ and the character waking up and appearing at the beginning of the film was intended to signify the arrival and birth of computer animation in films. The piece was shown for the first time on July 25, 1984, at SIGGRAPH in Minneapolis, but it was not completed when it was first shown. The later part of the film could not be rendered in time and was shown as just paper sketch drawings crudely animated. The final rendering of the film was released a month later, on August 17 at the International Film Festival (IFF) in Toronto.

The director was Alvy Ray Smith, technical lead: Bill Reeves, technical contributions by Tom Duff, Eben Ostby, Rob Cook, Loren Carpenter, Ed Catmull, David Salesin, Tom Porter, and Sam Leffler, filmed out by David DiFrancesco, Tom Noggle, and Don Conway, and computer logistics by Craig Good. The animation was done primarily by a young animator John Lasseter. Lasseter had earlier been at Disney and both Catmull and Smith knew him from one of their earlier visits to the animation studio. While many other animators had been scared by the concept of computerization, Lasseter was not. Smith recalls not only getting on with Lasseter at Disney but a delightful afternoon when Lasseter took him down to the Disney Archives and happily showed him original drawings and cels from the great Disney movies of all time. This included the dancing hippos from Fantasia, the pink elephant scene from Dumbo and many more. When Ed Catmull later mentioned to Smith that he had run into the now unemployed Lasseter at a conference on the Queen Mary in Long Beach, – “and I said Ed get off the phone right now and go hire him… and he said ‘Oh OK that’s a great idea’, .. we had to bring him in not as an animator, we had to hire him as a user interface designer or something (at Lucasfilm)”.

When the team started The Adventures of André and Wally B., Lasseter had not yet been hired, and the be was not part of the story and André was actually a robot. “He would rise and greet the new day,” says Smith, “and this story just sucked, but he looked at that and said, ‘I have a few recommendations’, and we said ‘Of course that’s why we want you here!’, and he then took the story and made André into a little lovable character, with soft floppy shapes and he added a bee. John is a natural talent of the first order, he gets paid a movie star salary because he is a movie star. He is really good, and its magical,” says Smith proudly.


Knoll Brothers

It was during the end of this time that David DeiFranesco kept urging Smith to visit another building in the Lucasfilm compound, where two young brothers were starting work on implementing a paint system using some of the team’s ideas on what Smith described as a ‘toy’. “I said, Who cares? You couldn’t do any serious graphics on them (the computers), so I said, ‘Oh Ok’.” The computer was a Mac, and the brothers: John and Thomas Knoll and the program: Photoshop.

Thomas Knoll, a PhD student at the University of Michigan, began writing the program on his Macintosh Plus (in monochrome). This program, called Display, caught the attention of his brother John Knoll, at Industrial Light & Magic, who recommended that Thomas turn it into a fully-fledged image editing program. Thomas took a six month break from his studies in 1988 to collaborate with his brother on the program. Ironically, Smith did not take the project seriously, but he would later figure greatly in defending Adobe who bought it (via BarneyScan who sold about 200 copies as a way to help sell scanners) and who then continued its development while marketing it worldwide.

Pixar

One of 5 copies of the original Pixar deal documents.

Pixar was born out of a feeling that Catmull and Smith had that their time was limited at Lucasfilm. After nearly being sold to GM – as you can hear in detail in the podcast – the team bought the technology from Lucasfilm, funded by Steve Jobs. Jobs had left Apple, or rather been fired from Apple by John Sculley (someone history will no doubt be extremely harsh on). Sculley had removed Jobs from his managerial duties, Jobs resigned from Apple and then founded NeXT Inc. the same year. Jobs invested $5 million to fund the Pixar technology buyout and another $5 million to capitalize the new company. Jobs bought exclusive rights to the technology developed by the Pixar team from Lucasfilm (excluding Lucasfilm which was allowed to use any of the technology, and still to this day does not pay any fees to Pixar to use RenderMan). Initially Jobs was just the primary investor. The founding documents cite Ed Catmull and Alvy Ray Smith as the management of Pixar. Steve owned 70% of the new company, the management and employees the other 30%. Ed Catmull and Alvy Ray Smith were the second two largest and equal shareholders, as President and Vice President, respectively, of Pixar management. Over time, Jobs would be forced to invest a further $40 million dollars (for a total of $50 million – half his fortune at that time). In the process Jobs gained 100% ownership of the company by 1991.


Among the founding employees, 38 of them in addition to Catmull and Smith included Cook, Carpenter, DiFrancessco, Lasseter, Porter, and Reeves. They and the rest of the key team moved to Pixar. The new company made hardware – the Pixar Image Computer which was used in medical imaging and elsewhere. They also made TV commercials and continued to develop the CAPS system for Disney (see below) – not that Disney would allow Pixar at that time to publicly say so, “The magic will go away if people find out that computers are involved,” Smith recalls being one of the issues.

 

The original list of Pixar employee’s NDA list
A page from an original Pixar business plan.

CAPS

One project started at Lucasfilm and finished at Pixar was the Computer Animation Production System (CAPS) project. CAPS forever changed the way Disney made its ‘cel animated’ movies. It was also the first income for the new company Pixar after the initial investment.

Left to right: Lem Davis, Bob Lambert, Alvy Ray Smith, Peter Schneider, Roy Disney (Walt’s nephew). Photo courtesy Bob Lambert.

The deal took 18 months to negotiate, but Smith says, “That was the deal that made Disney and Pixar love each other. Both companies had great admiration for each other from there on.”

One of the key techniques developed by the early Disney animators was to multi-plane. This allowed all the layers of the art to pan, but at different rates. This adds great image depth, but was extremely complex to achieve. Adding multi-plane to a shot would adds tens of thousands if not hundreds of thousands of dollars to the cost of a shot. It was therefore reserved for only key shots such as the opening of say Pinocchio.

The original multi-plane camera stands worked by having various parts of the artwork layers left transparent, to allow other layers to be seen behind them. The movements are calculated and photographed frame-by-frame, with the result being an illusion of depth by having several layers of artwork moving at different speeds – the further away from the camera, the slower the speed. The multi-plane effect is sometimes referred to as a parallax process (Wikipedia).

An interesting variation is to have the background and foreground move in opposite directions. This creates an effect of rotation. An early example is the scene in Walt Disney’s Snow White and the Seven Dwarfs where the evil Queen drinks her potion and the surroundings appear to spin around her. The pre-CAPS multi-plane system was limited to only five layers and “almost pulled down the entire power grid of Burbank, it was so complex,” jokes Smith.

CAPS was the first digital ink and paint system used in Disney feature films, although the Pixar crew had been looking at this problem since NYIT. In addition to simply doing multi-plane shots much more easily, the CAPS system meant that multi-plane camera shots were not limited by artwork size. Extensive camera movements never before seen were incorporated into the films. The other aspect of the CAPS system was its paint system functions where areas and lines could be easily colored in using an unlimited palette. Transparent shading, blended colors and other sophisticated techniques could be extensively used that were not previously available.
CAPS was paint now as a compositing system and a true landmark for all other VFX compositing systems that would follow, even in live action. In 1992, the team that developed CAPS won an Academy of Motion Picture Arts and Sciences Scientific and Engineering Award.

CAPS was used on The Rescuers Down Under, a film some called the first 100% digital feature film ever produced, although not Smith himself. Subsequent films using CAPS included Beauty and the Beast, Aladdin, The Lion King and The Hunchback of Notre Dame. “People just started saying ‘wow there is something different in these Disney movies’,” recalls Smith.

Patents

Smith left Pixar before Toy Story was released after a heated fight with Jobs, over a white board. (You can hear the details in his own words on the podcast), but before Smith left he was approached about a patent issue in the UK.

Spaceward was being taken to court by Quantel, who claimed that Spaceward’s paint system violated Quantel’s patents.The trials began with an approach at a SIGGRAPH in the 1990s, probably led by Robin Forrest a longtime colleague of Smith. Forrest was accompanied by a solicitor acting for Spaceward in the UK.

“With great earnestness they asked me to aid them in a patent challenge from Quantel, also of England, but a company I had obviously watched for years. Spaceward, on the other hand, I had not heard of, but I did understand their problem and immediately sympathized with them, annoyed by Quantel’s claims on our technology.

Quantel had for years built and sold a beautiful realization of a paint program. When we had first seen the Paintbox at a National Association of Broadcasters convention, we had known immediately what we were seeing: paint done in hardware. There was no other way to get the speed they were showing at that time, in the early 1980s,” Smith would later write.

The trial was not a pleasant experience for Smith or for some of the other key expert witnesses called in defense of what would be a losing argument for Spaceward. Many people, this author included, recall this key trial and the outrage it stirred up at the time.

“The point is that we knew immediately that Quantel had dropped it [painting] into hardware. We never for a moment thought they had ‘invented’ painting. Paintbox appeared in the 1980s, after all, and we had done painting in the 1970s. This, of course, was the point of the upcoming trial Quantel v. Spaceward. Quantel claimed to have invented ‘airbrushng,’ or soft-edged painting—or ‘wetpainting’ as I had called it in 1977. This was why Spaceward had approached me: to establish my priority and get them off a patent infringement hook. The Spaceward product was an airbrushing paint program aimed at the video market, directly competitive with the Quantel Paintbox, and cheaper.

Quantel also claimed, via British patents, to have invented digital compositing. To my mind, airbrushing and digital compositing are the same thing.”

Smith accepted the pro bono work with the Spaceward team with the blessing and support of his colleagues at Pixar and the computer graphics world in general, but Spaceward would lose the landmark case.

“This trial affected us in different ways. Jim Blinn was so scorched by the process that he refuses still to help anyone else in patent battles. I was furious at having my word brought into question and my priority stolen. But I learned a lot about the patent process and was therefore much better prepared, and much less naïve, when the call came a second time.”

Microsoft

Microsoft ended up with many of the original Xerox PARC alumni not just Smith, but also Butler Lampson, Chuck Thacker (ex DEC), Charles Simonyi, Gary Starkweather (ex Apple) and one almost Xerox PARC alumni Jim Blinn. Smith joined Microsoft in 1984 as its first and contractually meant to be its only ‘Microsoft Fellow’. Blinn would soon join as Microsoft’s second only ‘Fellow’.

Also ex-Xerox, having started in 1978, was John Warnock who left Xerox PARC in 1982 to co-found Adobe, where he was CEO until he retired in 2001.

While Warnock was not at Microsoft, Adobe did seek Smith’s help as Quantel moved from Spaceward to Adobe seeking patent infringement fees for every copy of Photoshop ever sold. Warnock was an old friend of Smith’s. Smith agreed to help Adobe only if they “took the case all the way”.

Microsoft also supported the move, lending Smith Microsoft’s brightest litigation attorney Tom Burt and together they took on Quantel once again. This time however the result would be very different. The Adobe trial took place in Delware in 1997, and “the Quantel team waved around their Emmy, and I my technical Academy Award for Pioneering inventions in digital image compositing.” (awarded to Smith, Tom Porter, Rom Duff and Ed Catmull the year before).  Smith actually had a copy of Paint 3 which the defense team got running again, Smith proudly showing his paint program with code written 20 years before.

“Then I left for Italy, not having heard most of the arguments. The only clues that perhaps I had scored were the grins that the jurors gave me as I departed the stand, with my Academy Award that I almost left behind. I believe they would have cheered if not for the dignified courtroom setting. Much to my amazement, my old colleague Ed Catmull searched me down via telephone in a remote Tuscan village a few weeks later to give me the remarkable news that not only had the jury found Adobe innocent but all the Quantel patents invalid, and even recommended that Quantel be found guilty of defrauding the US Patent Office (the latter recommendation not followed by the judge). How sweet it was! The decision also saved Adobe several hundred million dollars in demanded royalties on Photoshop.”

Today

Digital photography became a passion for Dr Alvy Ray Smith – images of a different digital source, and ones that led him to travel to Africa and other places.

Digital photography by Dr Alvy Ray Smith.

As soon as digital photography became possible, Smith got a Nikon D1X digital camera with 5.5 megapixels.

Since then Smith has been on trips to places such as Tanzania in Africa, on one trip taking over 2000 shots.

The other great passion for Smith today is nothing to do with Paint. He has for sometime been active in genealogy.

“I have been keeping a family database, from which I generate the Family Tree, on computer for a over a decade. The family database now contains over 12,000 individuals, many anecdotes, and over 2200 surnames.”

We wish to thank him for taking time to talk to fxguide and provide such a wealth of information.

5 thoughts on “Alvy Ray Smith: RGBA, the birth of compositing & the founding of Pixar”

  1. What an excellent article and podcast. It really put context on what was happening in my work in the early 90s, when I was working in video and these new computer guys, with their SGI machines, arrived. Why did they call a ‘key’ an ‘alpha’? And a ‘frame store’ a ‘frame buffer’. Why were they messing with stuff which took ages to render one frame, when our DVE could fly two channels of video around in real time? While they didn’t seem to quite get the video world, they were doing some very very cool stuff and had some great demo’s on their SGI Reality Engine. And they were talking about new software called Flame…
    A lot has changed since then. This interview is a gem and it’s great to get a perspective on what was happening in the film, computer and business worlds in those early days.
    Great stuff – Thanks guys.

  2. shashikiran jeppu

    The best podcast i have ever listened to. feel so connected when m working now. Great job guys. you are the best.

  3. What a great interview. Thanks so much for capturing this and sharing it. It is truly inspirational and crazy to think we now live in a world where the thought of this technology not exciting is absurd.

    Interesting that Dr Alvy mentions Apple computers not being serious graphics machines all those years back and how that situation hasn’t changed today in 2012.

  4. Jonathan Kendall

    I have been a DVE/CGI enthusiast since watching Westworld in the sevenites I have to take my hat off to the producer of this article. It is the most comprehensive and, to my knowledge, historically accurate, account of what are the most important development stages in CGI history. Well done!

Comments are closed.