Multimedia
Blogs
Forum
Affordable HD
Whitepapers
Newsletters
Events
Advertisers
Blogcast
MillimeterAlias, Avid, Discreet, Quantel, SGI
1998 was a year to remember. That's when we first heard the phrase that “it all depends on what the meaning of the word ‘is’ is”. The US had its first balanced budget in 30 years, and Siggraph celebrated its first 25 years.
For anyone watching the top post companies — the ones that supplied the editing, graphics, and effects hardware and software that literally transformed post in the 1990s — a snapshot of 1998 would have revealed manufacturers who were at the top of their game (Alias|Wavefront), down on their luck (Avid), just about to catch fire (Discreet), about to make a near disastrous move (SGI), and unsure of their way (Quantel).
Here's a look back at the tumult these companies faced since 1998, the challenges they dealt with, and the decisions they made. Like the rest of America, they went on a five-year rollercoaster ride with more twists than a soap opera plotline. A happy ending? Who knows? (Soap operas don't have endings). But each company remains a vital player today. Here are their stories.
In July of this year, as part of its 20
But back in 1998, the company was focused on the highest end of the graphics and animation markets. That year, Bill Kovacs and Roy Hall of Wavefront received Academy Award plaques for their development of Advanced Visualizer. The product was acknowledged by the Academy as the first widely adopted commercial software package for modeling, animating, and rendering with sufficient quality for use in motion pictures. (Alias|Wavefront received an Oscar statuette in 2003.)
In 1998, Alias|Wavefront also introduced Maya, its new flagship 3D product, which would soon become one of the most popular animation and effects programs in history. ILM, for example, purchased enough seats of Maya that H.B. Siegel, then CTO at Lucas Digital Limited, said, “Maya is going to change the way we do things at ILM.”
So why, five years later, did Alias need to move into a wider, more difficult marketplace?
In 2000, Alias management predicted that the high-end DCC software market would contract throughout the first five years of the century. Software makers, they decided, would need to make significant price drops just to compete. Alias decided to move quickly, ahead of that revenue decline. From 2000 to 2002, it built the infrastructure necessary to support its entry to a larger marketplace. Alias would still target the entertainment industry pros, but its products also needed to attract educators, corporate, and even the prosumer markets. In March 2002, the company discounted its flagship Maya Complete to $1,999. Suddenly, buying a top-drawer 3D graphics and animation package made sense to a whole new range of customers, their appetites whetted by years of Hollywood blockbusters and flashy commercials.
“If you wanted to run a sustainable, global business,” says Doug Walker, president of Alias, “you needed to have a significant share of the market, since it's costly to run such a global business. There's a tremendous amount of infrastructure you need to invest in to make it work effectively.”
Global marketing, of course, requires well-known products. With Maya, Alias knows it has a brand leader, although that's still more of an industry insider's perception. Even so, company literature regularly touts that every Academy Award-winning film since 1995 used that software, and that's something that can help sell to a larger audience. “Maya's brand is one of aspiration,” says Walker. “I contend that it's a lot easier to take a product (like Maya) down market — especially when you have a brand like that — than to take (a cheaper solution) up market.”
On the road to becoming a “volume” market supplier, Alias released a free, non-commercial version of Maya Complete (Maya Personal Learning Edition); pushed into 3D education via 20 new self-paced Learning Tools books and CDs with publisher Sybex; launched Sketchbook Pro, an inexpensive pen tablet-based drawing program that's now bundled by a number of Tablet PC makers; and started a Web-based sales effort with PC Mall.
But while Alias wants to sell into a much larger audience, the company won't try to teach high-end 3D skills to prosumers. Such skills, of course, take years to build up. Meanwhile, pro users don't want a dummied-down version of the software either, says Walker.
So Alias will provide “smarter software.” Coming out of the company's well-regarded R&D group — which has included the renowned “human-computer” interface designer Bill Buxton — future software will deliver a new way to interact with the 3D world, says Walker. “There's lots of room to grow 3D in the future.”
Five years ago, the only place Avid had to go was down. At least that's what the stock market seemed to think. Avid had an overwhelming share of the postproduction marketplace, especially the Hollywood market segment, but some questioned whether the market for NLEs was saturated.
In 1995, Avid had purchased digital audio workstation leader Digidesign and 3D graphics and effects companies Elastic Reality and Parallax Software, which evolved into the Media Illusion product. 1998 saw the acquisition of Montreal's Softimage, which eventually provided Avid success in the high-end graphics/effects compositing and animation market and later became the platform for Avid's first entry into high-definition video. But the struggle to achieve success caused culture clash and development pains along the way.
By 1998 Avid was promoting its “Online” editing technology, and many were noting that its NLEs did not yet have online audio three years after Avid had acquired Digidesign. The low-end NLE products, such as Media Suite Pro and then Reel Impact (which evolved into MCXpress and years later Avid Xpress), also appeared to have trouble gaining the market share that many would expect from an Avid product.
Avid's prototype SGI-based Workgroup Postproduction Servers never came to the market. Its OMF interchange format never received the needed acceptance from competitors to be an industry-wide interchange standard and was later canceled.
The company had also began to attract complaints concerning what owners viewed as unreasonably expensive upgrade paths. Many Media Composer owners felt betrayed with the introduction of a new flagship, The Avid Symphony, and soon after that, the Avid|DS.
Five years ago the Avid user community was frustrated and vocal. At NAB 1999, an untrue report that Avid had decided to cancel development on the Apple platform spread like wildfire. This was a low-point, and the board of directors knew it.
By the end of 1999, Avid began a major restructuring. David Krall was brought in from his position as the COO of the successful Digidesign division to be president and later CEO — the third in three years. He built a new team with Mike Rockwell as CTO and Paul Milbury as CFO.
Good things were beginning to happen. Avid won an Oscar statuette for the Film Composer system for motion picture editing. Moreover, what can arguably be one of the most important introductions of the past five years, the Avid Unity, was introduced in 1999. Avid also introduced a low-end Avid ShowBiz Producer (which would evolve into the successful Avid Xpress DV). This was occurring at the point where the Apple Final Cut Pro was beginning to take sales from Avid in several market segments.
The year 2000 saw the tide begin to turn for the company. Additional acquisitions took place, including The Motion Factory, which had a positive impact on Avid's Trilligent rich media server and MetaSync technology.
Avid also acquired Pluto Technologies, which offered new low-cost capabilities in the broadcast division. At the time, Avid was part owner with Grass Valley Group of the corporate entity known as iNews. The following year Avid would purchase GVG's share of iNews, and it would become a successful broadcast division of Avid, just at a time when digital broadcast news and operations were soon to take off.
More importantly, 2000 saw the introduction of Avid's first editing solutions to run on a laptop — Avid Xpress DV and Avid Newscutter XP — as well as the launch of Avid Unity for News, and the rebranded Avid|DS. Avid was selected by NBC for Summer and Winter Olympics coverage and won an Emmy for pioneering development of full-motion broadcast-quality PC video and compression plug-in cards used to manufacture nonlinear editing systems or video servers.
But the most important event no one knew about. Avid engineering started a separate, secret R&D team to begin what we now know to be Avid DNA, developed without the knowledge of most Avid personnel and completely separate from all the company's other R&D efforts.
Today Avid is again considered a major success with strong showings in the stock market, and with what may be one of the most daring corporate transformations our industry has seen. At NAB 2003, Avid introduced a new line of products — Digital Nonlinear Accelerators or DNA. This combination hardware/software solution re-energized Avid's product with state-of-the-art technology and incredible price/performance. For example, Avid boasted that the price of its mainstream Media Composer was cut in half while its quality and performance increased significantly. Avid promises more surprises and has established itself once again as an industry leader.
1998 was a big year for Discreet. It introduced Discreet Edit with three different flavors, Discreet Effect and Discreet Paint, which later would evolve into Discreet Combustion, and it demonstrated version 2.5 of both Fire and Smoke — the latter introduced with a much lower price-tag and new HD capabilities. That year, Discreet won an Academy Award for Flame and Inferno and introduced realtime HD editing capabilities to Fire. Both Discreet and the industry thought the days of HD were upon us.
In 1999, Autodesk acquired Discreet. Fire, Smoke, Edit, Paint, and Effect were all doing well in the marketplace with feature-rich new versions. Version 5.0 of Edit was so good that headlines proclaimed it to be the best Windows-based NLE. 1999 ended with an announcement that Discreet had sold its one hundred thousandth copy of 3ds Max.
Discreet Combustion, an evolution of the Effect and Paint products, was introduced to great acclaim in 2000. Version 4 of Fire and Smoke at IBC advanced each product's feature set, and Discreet appeared to be emphasizing web streaming, HD/universal mastering postproduction, and infrastructure tools, including the introduction of Backdraft, in its product lines. The emphasis on streaming and games content creation continued in 2001. That year, Discreet announced it had acquired Media 100's streaming video software product division, including the Cleaner family of compression products, Cinestream, and IntroDV. It also saw the final version of Edit at 6.5.
Last year, on the 10
This year saw continued development of its advanced systems products: Inferno 5, Fire 6, Flame 8, Smoke 6, Flint 8 (all SGI Irix-based), and a new Linux version of Smoke 6, but there was no sign of the promised Windows-based advanced systems products, still in development.
Discreet also introduced Lustre, a digital color-grading solution developed in conjunction with Colorfront, which is capable of working in all digital cinema resolutions. Discreet's other HD and digital cinema resolution products continue their success. Also, the trend away from YUV toward RGB high-rez in this high-end market segment — especially with digital film transfers and graphics/animation to digital intermediates — has benefited Discreet.
Discreet has also done very well with the Cleaner products — especially the new Cleaner XL for Windows — and with Combustion. The latter product had its price significantly lowered to $995 at the beginning of the year, which successfully spurred sales. In September Combustion 3 was introduced with many improvements and new features. While the high-end products continue generating the bulk of the revenues and profits in a competitive marketplace, Discreet Cleaner and Combustion, together with version 6 of 3ds Max, have done their part and aided the company's success in this tumultuous economic period.
Five years ago, Quantel could hardly be thought of as “hurting,” but many thought it was not living up to expectations in a very small and competitive high-end marketplace. Quantel had help set those high expectations itself when in 1988 it introduced Henry, the first online nonlinear editor/compositor/finishing system, followed a year later by Editbox, products that were generally considered the “ultimate” in quality and performance.
On the one hand, Quantel remained a profitable company with great prototypes at the various NABs and with many awards, including 10 Emmys. On the other hand, the company grew more frustrating and arrogant. The software featured what many considered the best ergonomics but retained features that exasperated many.
For example, its editing technology did not offer an undo feature. When I asked a Quantel representative why this was the case, he replied, “The more pain it causes, the sooner you learn to do it correctly.” That pretty much summed up the attitude that many were put off by.
Then, in 2000 Quantel management bought the company from its mega-parent Carlton Communications, and things changed radically. It became an employee-owned company, and there was a newfound urgency to please the customer. Quantel metamorphosed into a creative systems company practically overnight. It was as if it became a start-up company with 30+ years of intellectual property.
In July 2000, Quantel introduced the first member of what would become the GenerationQ family, the iQ. “Acceptance was immediate,” Quantel says. “iQ has rapidly become the workhorse of the digital intermediate business around the world, with customers ranging from Lucas Digital and Peter Jackson's Film Unit to Cinecittà Digital in Rome and London's Moving Picture Company.”
Last year, Quantel introduced the full GenerationQ. The result of incredible R&D efforts, this represents no less than a recreation of the company and a replacement of its entire line with three families of higher performing and price-competitive products. These include a new broadcast newsroom server, sQserver, and new software-only broadcast editing products such as QCut, QBrowse, and the QEdit which evolved into the software/hardware turnkey hybrid QEdit Pro; a graphics family from the software-only QPaintbox, to the powerful new gQ and the hybrid QPaintbox Pro, and a postproduction family, including the iQ flagship, the powerful eQ, the software-only QEffects and later the QEdit Pro and QColor
GenerationQ also includes exciting new-to-Quantel concepts such as “resolution independence,” AAF interchange support, common toolsets, advanced media asset management tools such as Frame Magic, as well as its previous 8-bit to 10-bit Dynamic Rounding technology.
With a bit of astonishment still lingering in his voice, SGI's Greg Estes, vice president, corporate marketing, said “We're working with Quantel!”
Years ago that would have been really remarkable, as the two rival companies each assumed top positions in the highly competitive graphics and effects markets.
But, as the saying goes, things happened. Once-expensive computer hardware — including powerful GPUs — became a commodity item. Coders became ever more adept at designing lower cost, off-the-shelf but sophisticated graphics and effects software to run on those radically less expensive workstations.
Over the past year, losses built up, forcing SGI to lay off some 1,000 workers to lower operating costs. But SGI's markets, which include high-end effects, research, geology, and especially a free-spending military, continue to buffer it from the worst of the dot bomb meltdown. With a lower cost structure — which included subleasing most of its Amphitheatre Technology Center campus to Google — SGI has turned the corner, says Estes.
Another concern you'll see in the press: SCO Group, Inc., claiming ownership of crucial Linux core code, threatens to take SGI to court for a supposed breach of license. In its well-funded moves against the Linux community, SCO brought a $3 billion suit against IBM and has threatened other companies as well as individual Linux users.
What's the issue? SCO claims that SGI had violated licensing contracts, both in code used in SGI's Unix IP and for its donation of XFS code to the Linux community.
“Our position is that there is no merit to their claims,” says Estes. “We don't need anything from them, so there are no engineering issues (to resolve). We purchased a fully paid-up Unix license from Novell some years ago (when Novell owned the disputed code), so we're saying back to them, ‘Our Unix license is non-terminable’.” At press time, SCO indefinitely extended an October 14
Meanwhile, SGI has moved far beyond its beginnings as a supplier of graphics terminals and workstations.
For example, company executives acknowledge that they won't compete in the large commodity desktop graphics market; a move into the Intel/Windows NT market in the late 1990s cost the company dearly when SGI attempted to include its own non-commodity graphics hardware in its workstations.
Tezro, however, introduced in July, sparked a renewed interest in the company's ability to deliver competitive high-end hardware. Pricing around $50,000 for one configuration of the full four-processor version, Tezro enables Discreet to offer a lower cost turnkey workstation combined with some of the latest versions of its suite of software, including the soon-to-ship Smoke 6 HD. According to Estes, Tezro sales to date are well beyond initial projected sales figures.
Another bright spot: storage and networking. This summer, SGI joined Discreet's Infrastructure Sparks partner program, which fully qualified SGI's storage and networking systems with Discreet products. Discreet will now resell SGI SAN systems worldwide.
The company also collaborated with Quantel to notch an important proof-of-concept sale with a recent DI installation at The Film Unit, Peter Jackson's New Zealand-based post facility. To create a leading-edge digital intermediate facility, Jackson and staff chose the two companies to integrate Quantel's iQ finishing system with SGI's InfiniteStorage SAN storage and networking setup. Other key ingredients are SGI's CXFS shared file system and Pandora's color correction hardware. Now, editors can work off massive amounts of centralized storage to manipulate, edit, and color correct 2K film images in realtime, which can require up to 300MBps throughput.
“The bottom line is that SGI is doing better now than we have in a long time,” says Greg Estes. “We've got a lot of momentum with the new products, a renewed presence in the production community, and even our new storage effort is working out well. We're on the upswing now, with more positive news out of SGI than people have seen in a while.”
Sidebar
Media 100 Resets
Five years ago, Media 100 had seven Mac-based Media 100 models and had just announced the Windows-based Finish NLE. The company was recognized for high-quality digitization of analog video. Media 100 was collaborating with Macromedia's Key Grip software. As we now know, Apple acquired that software which became Final Cut Pro, and Media 100's collaboration efforts were for naught.
By the end of 1998, Media 100 announced version 5.0 software with a DV-based version of the application. What no one outside Media 100 knew was that the company was also completing the second year of secret development for something new and exciting. Media 100 was utilizing its hardware expertise to develop a hardware-based compositing/FX product unlike anything on the market. The heart of the system would be the GenesisEngine, an ASIC-based “media supercomputer.” The result would be Media 100 844/X, introduced in February 26, 2002.
In 1999, the company had acquired Terran Interactive, leaders in streaming video compression technology. In 2000, with Media Cleaner sales about $4 million per quarter, the product was the defacto standard for producing video on the Internet. When the Internet bubble burst, it cut into the profits. The extraordinary R&D for the new project was costly. When Discreet made a $16 million offer for Media 100's streaming media products in 2001, it was too good to turn down — even though profits were still being generated from the division.
While the company is still in the red, sales of the Media 100 844/X products have seen a 33% increase this past quarter over the previous quarter, totaling 300 systems. Version 2 (known as the “Finishing Release”) has been shipping since May, as well as the realtime XBLUR Gaussian blur option. Media 100 has continued to develop and sell its dual-stream NLE products and its Final Effects software plug-ins. New HDX technology for high-definition input and output — available on both the 844/X products and the dual-stream Media 100 products — has been promised for release before the end of the year.