After the tremendous success of ShaderX a new book with an entirely new set of innovative ideas, techniques,and algorithms will be published by Wordware: ShaderX 2. Game developers of all levels will find insightful tips and tools from this unique collection. Written by game programming experts, each contribution will cover new Shader tips and tricks with DirectX 8/9 hardware. ShaderX 2 will get around 500 pages and it should be in stores in August 2003. Read about it here.

Microsoft has finally released DirectX 9 to the public. New features are

Microsoft's DirectX 9 download page.

Speculation is that this will be the last drop for a while. The next MS Desktop-OS release (Longhorn) will be/is closely tied to DirectX. I've heard that DirectX9.1 will be released mid/late next year to sometime in 2004. I've also heard that there'll be various point releases as new hardware vendors release DX9 flavored hardware so MS will be doing point releases as they become necessary to keep DX9 "fresh" until the DX10/Longhorn release. This is all rumor at this point. The interesting thing is that Longhorn is supposed to take advantage of the GPU just like a graphics program should, that 3D will become part of the desktop. The Longhorn screenshots I've seen aren't that impressive. I worked at Xerox PARC for a while and the stuff they had in 1996 was way cooler. Let's not waste 3D computing power on visual gewgaws.

NVIDIA used Fall Comdex to announce that the NV30 will be called the GeForceFX, and had a Hollywood/cinematic-themed release. The GeForceFX is pretty much as expected - it's a beefy chipset that's probably going to make it slightly faster that ATI's RADEON 9700. The interesting parts are that NVIDIA's using 128 bit DDR2 memory instead of the more easily available DDR memory. This means that the throughput for the GeForceFX will be twice as fast as other cards. But since a lot of other vendors are using 256 bit memory bus widths, this tends to even out performance. The other interesting bits are the shader "plus" bits. Basically you'll be able to check some caps bits to detect the additional programmability that is available in the NV30's DX9 shader capabilities. This lets you program some conditional branches on an GeForceFX that you can't on other DX9 chips. NVIDIA calls this their CineFX shading engine. Full color pixels (16-bit and 32-bit per color) which are necessary when you consider that pixel shader can now not only have over 1000 instructions but can loop as well, so round-off error becomes very important.

It'll be interesting to see how developers take advantage of the extended shader capabilities. This seems like a flip on the PS 1.3 (NVIDIA) vs. PS 1.4 (ATI) question where developers had to program to both shader versions to cover all bases. In this case you can choose not to use the extended capabilities and get full ATI and NVIDIA support (at least on DX9 capable parts). There's a nice summary of the features of the GeForceFX on Tom's Hardware Guide here. The real question is availability. It's getting a bit late to think that NVIDIA is going to ship any significant amount of GeForceFX cards this year. However, I've been told the NVIDIA is planning some aggressive pricing, with bringing some slower variants of the GeForceFX to the $US100 range next year.

Microsoft takes Xbox "live" today with its multimillion-dollar investment into online broadband console gaming. For $50, gamers will get 12 months of online gaming through Xbox Live, along with a Xbox Voice Communicator allowing them to talk to friends while playing games over a broadband connection and two online game demos -- THQ's "Moto GP" and Microsoft's "Whacked!"

After the initial year expires, Microsoft is expected to charge about $9.95 a month for Xbox Live. Consumers, though, also must pay for a broadband connection. Xbox Live will be supported by 14 games this year, most of which allow for competition among two to 16 players. "Over time, all games will have some form of online functionality," said J. Allard, Microsoft's general manager of the Xbox platform. "We haven't created an Xbox Live division; online will be part of everything we do. It's like a chip."

Microsoft expects a quarter of next year's Xbox games to incorporate some type of online functionality. Of the 50 online titles to ship by the end of 2003, some of the biggest brands will make use of broadband connectivity, including Microsoft's "Halo 2," "Project Gotham Racing 2," "Midtown Madness 3" and Activision's "Return to Castle Wolfenstein." "When we started designing Xbox three years ago, we looked at the machine as the next shift in the evolution of entertainment," said Scott Henson, director of platform strategy for Xbox. "Online gaming was a major part of that strategy, which is why Xbox ships with a hard drive and broadband Ethernet built-in." Xbox Live is part of a 10-year plan by the software giant to stake a claim in broadband-enabled living rooms around the globe. There are about 15.7 million broadband users in the United States, according to Leichtman Research Group.

Gaming is seen as a key element in growing broadband, and DSL companies like Speakeasy.net are giving away free Xbox and PlayStation 2 systems to customers who commit to their broadband packages, which begin at $60 a month and are tailored to video games. Microsoft has spent 18 months building the worldwide infrastructure for its online gaming. There are two servers running out of Seattle, with additional servers in London and Tokyo. "Broadband gaming opens up the ability for publishers to post free game demos online for anyone to try out," said Jon Thomason, director of software development for Xbox. "It also allows developers to try out a few levels of a game 'pilot,' to test the waters before investing on a complete game." Microsoft also will allow publishers to sell expansion packs and other downloadable content to gamers via their Xbox Live connection. And massively multiplayer online gamers also are expected to flock to Xbox Live. "What if Mountain Dew wants to target gamers with a new drink?" Allard asked. "They could go to Activision and ask for a few free Tony Hawk levels, which Mountain Dew would fund and then distribute free online to promote the drink. The possibilities are endless."

It's been tough times for ATI in the last few years. They've been filled with anguish that upstart NVidia, less than (back then) a tenth ATI's size, was getting all of the attention. From my dealings with folks inside ATI I understood that they were attempting to change their image as a stodgy, mass-market graphics chip manufacturer with perpetually late, buggy drivers. They were actively looking for evangelists, looking to create a developer relations team, and generally looking to kick NVIDIA's butt. It look's like they may have pulled it off.

While NVidia has had rumored production problems with the NV30, ATI has beaten them out the gate by a wide margin. It was entertaining to see ATI front-man & engineer Jason Mitchell at Siggraph, just grinning as he demo'd ATI's latest. Using beta DirectX 9/ OpenGL 2.0 drivers, Jason demonstrated the advanced vertex and pixel shading that the new RADEON 9700 was capable of. It was pretty impressive. While NVidia was talking about cinematic techniques on hardware, ATI was demonstrating it. (To be fair NVIDIA's NV30 will cover the same ground as the 9700, and NVidia did demo some of the features as well.) In all the test results I've seen, the 9700 is not only faster and more feature rich than any other video card out there, it's better by a wide margin. Given that ATI is shipping months before we can expect to see the latest from NVidia, ATI may be able to take a huge chunk out of NVIDIA's market share. Of course, having great hardware is just part of the story.

What's really impressed me is that ATI's finally "got it". Not only do they have really great hardware, not only are the drivers nearly fully featured and stable for a change, but ATI is actively rolling out developer tools, whitepapers, a developer conference (Mojo Day), all ala NVidia. Getting behind developers, assisting them in understanding the hardware (particularly with the introduction of shaders), posting demo code, etc. is a sure way of leading game companies down the path of enthusiastic support. So not only is the RADEON 9700 a kick-ass bit 'o graphics heaven, ATI has gone the extra distance to make sure that developers have the resources available to them to put that hardware to work. ATI has also teamed up with 3DLabs to drive the OpenGL 2.0 spec process. Yes, I think they've got it. I get to vote on the Game Developer mag's Front Line Awards this year - I think I know which way my vote is going.

Santa Clara — Wheeling and dealing its way into the third quarter, NVIDIA Corp. (NASDAQ: NVDA) started off the week by announcing plans to open source its NVIDIA Cg Compiler technology and provide the ingredients for creating Cg compilers for other platforms and architectures.

The Santa Clara, Calif.-based semiconductor provides 2D, 3D, video, and multimedia processing technology for PC-based applications in the manufacturing, science, e-business, entertainment, and education industries. Cg Language Specification - C for Graphics, a high-level shading language, was developed earlier this year in collaboration with Microsoft Corp. (NASDAQ: MSFT) and is compatible with Microsoft's recently released DirectX 9.0. Cg has met with widespread developer and market acceptance, although "petty licensing disputes" have created some setbacks for Cg Compiler technology, prompting NVIDIA to open source the compiler code in hopes of spurring Cg adoption and removing the need for developers to program directly to the graphics hardware assembly language."NVIDIA's decision to open source some of their development work with a very liberal license is a positive step that I hope other vendors will follow," said John Carmack, co-founder and owner of id Software, Inc. According to NVIDIA, the Cg Compiler code will contain the parser that reads the language and creates intermediate code for compilation as well as a generic back-end. Under a nonrestrictive, free license, Cg Compiler code will be available for download in August from NVIDIA's site or www.cgshaders.org.

NVIDIA will also provide the full source code for sample shaders in the Cg Toolkit, which includes the Cg Compiler 1.0, the Cg Browser, a CgFX file format, and the Cg Standard Library. The Cg Toolkit is a relatively new offering from NVIDIA designed for game development applications, digital content creation, and design. This week, NVIDIA also announced its acquisition of Exluna, a Berkeley, Calif.-based 3D software company that was founded by former Pixar employees. Exluna originally caught NVIDIA's eye as an ideal marketing and development platform for Cg technology. Until Tuesday of this week, Exluna had been the subject of a lawsuit filed earlier this year by Emeryville, Calif.-based Pixar, a digital animation studio, charging Exluna with the misappropriation of trade secrets, copyright infringement, and patent infringement. More specifically, the charges were aimed at infringement violations against Pixar's patents on RenderMan, a software package used by animation studios. The two companies have reportedly resolved all pending litigation and disputes over copyright, although details of the settlement were not disclosed by NVIDIA or the two other parties involved. NVIDIA was also tapped this week by the Ohio Supercomputer Center (OSC) as the visual processing technology provider of choice for the Blueprint for an Advanced Learning Environment supercomputing project. OSC has deployed and installed more than 50 NVIDIA Quadro4 900 XGL workstation graphics boards for an integrated visualization cluster. The cluster will be used for training purposes by students and researchers as well as the underlying visualization technology for supercomputing projects. "To be able to decompose the computation and graphics within the same node will provide real-time results for massive computations that have long been a need of researchers across the state of Ohio," said OSC Associate Director Al Stutz.

Game god John Carmack has given OpenGL his blessing in the ever-thickening shader wars. Carmack says that given the 3Dlabs P10 card, he opted to use the OpenGL 2.0 extensions for the shader path, even though the Nvidia Cg extensions offered "the most expedient choice". And the port went so well, he writes in the latest addition to his .plan file, he's committed to using OpenGL as the back end for all implementations of Doom 3.0.

It isn't so much a shader war (not our term) as a language war, and Carmack sees such higher level languages as NVIDIA's Cg as inevitable for graphics developers. "It won't be too long before all real work is done in one of these, and developers that stick with the lower level interfaces will be regarded like people that write all-assembly PC applications today," he writes.

Nvidia has garnered much industry support for its Cg language, but this could help swing critical opinion behind a vendor neutral approach. Carmack has enthused about the quality of NVIDIA's OpenGL drivers in the past. Carmack's comments are received with great interest in the community, particularly because he chooses his words so carefully. He's no rambling blogger; Carmack had made just three updates to his .plan file this year. You can read his comments here and while we're on the subject, here's a gratuitous link to our favorite Carmack interview.

Santa Clara — NVIDIA Corp. stepped forward Thursday with a new language standard intended to radically simplify graphics programming. The Santa Clara, Calif.-based semiconductor maker is best known for creating advanced 2D, 3D, video graphics and multimedia processing technology for the consumer and professional computing markets. Developed in collaboration with Microsoft Corp. (NASDAQ:MSFT), the Cg Specification -- C for Graphics -- is a programming language that gives content developers the tools to create real-time cinematic renderings with greater speed and without the need to program directly to the graphics hardware, which can be a slow and difficult process. With the C-like syntax development of real-time shaders and visual effects for graphics platforms, Cg enables developers to create, share, and re-use shaders for richer, more cinematic visual effects. Cg is also compatible with Microsoft's recently announced High Level Shading Language for DirectX 9.0. "We are entering a brave new world of graphics software development," said Peter Glaskowsky, editor-in-chief of the Microprocessor Report, In-Stat MDR. "The announcement of Cg signals a revolution in computer graphics, and for the first time in years, we can say this without exaggerating." Cg has already been greeted warmly by the game developer and digital content creator communities, according to Jen Hsun Huang, NVIDIA's CEO, who in a like-minded announcement unveiled the Cg Toolkit to complement the Cg Specification. The Cg Toolkit will be made available to developers attending the London-based conference "The Gathering 2" on June 13-14 and includes a Cg Compiler, Cg Browser, CgFX file format, Cg Standard Library, and a collection of pre-written Cg shaders which can be used for game development, digital content creation, and computer-aided design. The toolkit will also be available for download from the company's site starting June 13. The Cg Compiler is a cross platform that generates optimized graphics assembly programs for real-time visual effects and was designed to take full advantage of NVIDIA's GPU features and pipeline. The resulting program is a more advanced vertex or pixel shader that is equal to or better than hand-coded graphics programs, according to the company. The Cg Compiler also supports programs written for Windows, OS X, Linux, Mac and Xbox. "By taking care of the painstaking work required to optimize special effects for today's GPUs, Cg will make the developer's life much easier," said Dan Vivoli, vice president of marketing at NVIDIA. "Because we do the heavy lifting with our Cg Compiler, developers can spend more time on the creative side of game development. This begins a new era of dramatically compelling story telling."

LOS ANGELES — Two of the creators of Microsoft's Xbox have joined a pair of video-game industry veterans to launch a venture with a familiar Hollywood feel: find raw talent, create edgy entertainment, and then sign with a major publisher to sell it to the world. But the product for Capital Entertainment Group is not movies, but video games, and the startup is in the process of raising $50 million in venture backing for just that purpose. The executive team includes former Xbox chief technology officer Seamus Blackley and Kevin Bachus, who handled relations with game developers for Microsoft, the company said. Former Sierra Online senior vice president J. Mark Hood and Dotted Line Entertainment founder Eugene Mauro are also behind the start-up, which plans to find video game developers with ideas that established publishers would normally shun. CEG will back the development with money and production resources and find game publishers to market and distribute the resulting games, very much like an independent movie studio. The game startup plans to make its money by taking a cut of net game sales. The company believes its investors can break-even on a game after sales of 250,000 copies. Developers will get a royalty, assuming a "breakpoint" is achieved within the first 12 months of sales, equal to anywhere between 5 percent and 15 percent of net sales. The new venture comes at a time when the video game industry is in the first year what most expect to be a cycle of unprecedented growth. The top three U.S. game publishers recorded sales of $714.3 million in the first quarter of this year, up 45 percent from a year earlier. Investors, analysts and even major media companies have started to pay closer attention to big-name game publishers, as they produce stock gains and quarterly returns far better than those coming from other technology or media companies.

TRYING TO INCREASE INNOVATION
"I've found over the last year that it was becoming increasingly difficult to get innovative, original titles funded," Mauro told Reuters. "I put together a model that was a little bit VC (venture capital), a lot of bit of production." Mauro serves as chief executive of the new venture and will work from his home base in Connecticut. Blackley, vice president of development, Bachus, vice president of publishing, and Hood, vice president of production, will work from Seattle. A third office will eventually be opened in Los Angeles. Bachus left Microsoft months ago, but Blackley did not resign from the software giant until April 22, only days after it cut its Xbox sales forecast for fiscal 2002 by as much as 40 percent after poor international sales. Blackley denied a correlation between the two events and said he was leaving simply because he wanted to get back to developing games. While the company does not plan to have any titles available at retail until the 2003 holiday season, it already has a deal in place with Sega Corp. for its first two games, with what Mauro described as "extraordinary latitude" in what the games will be and for which platform they will be developed.

WILL DEVELOP FOR MOST PLATFORMS
Bachus told Reuters that while the CEG team has a natural affinity for Microsoft Corp.'s Xbox, it will also back the development of games for the PC, Sony Corp.'s PlayStation 2 and Nintendo Co. Ltd.'s GameCube. For reasons of logistics and time commitment, he said they will largely stay away from working with developers on games for handheld platforms and mobile phones, as well as massively-multiplayer online games. "For Seamus and me ... it was really borne out of our desire to increase the level of innovation in the industry," he said. "CEG is the industry's first independent production company." A background sheet on CEG reads much like those for independent Hollywood producers, discussing fine details like negotiable net sales splits and completion bonding for titles in development -- essentially a form of collateral to ensure that game projects are finished. While the group intends to work with a number of different publishers, Bachus said they agreed to the initial production deal with Sega because "Sega entertaining this opportunity is very flattering."

San Jose— So where does an 800 pound gorilla sleep? Where ever he wants. Microsoft has told NVIDIA that they deal that they had for pricing chipsets last year isn't so cool. Microsoft points out that - hey, you must be producing them for less now, eh? So let's renegotiate, so what if a goodly chunk of your revenue come from us, I know you'll do da right thing. Officially NVIDIA disclosed that it's in arbitration with Microsoft over the pricing of its chips used in the XBOX. Just look at the corpse of SGI lying by the roadside, you'd think all those ex-SGI NVIDIA employees'd know better. Microsoft has a term for working with other companies..."GUBO".

Hamilton Bermuda — 3DLabs announced today that it has signed an agreement to be acquired by Creative Technology LTD. (NASDAQ:CREAF) Creative will purchase 3Dlabs stock at $3.60 per share, with two-thirds being converted to Creative stock and one-third to cash, valuing the company at approximately $170 million.

“We see great opportunities to leverage 3Dlabs' high-end professional market leadership into higher volume PC desktop products that satisfy power gamers' insatiable demand for faster graphics,” said Sim Wong Hoo, Chairman and CEO of Creative. “We have analyzed 3Dlabs' forthcoming products and technology roadmap and we believe that their development of a scalable visual processing architecture with new levels of programmability and flexibility will provide a significant competitive advantage in the graphics space. The technological breakthroughs that 3Dlabs has achieved to date, coupled with the strong market we see for high-performance graphics processors, provide a tremendous growth opportunity for our company.”

“We are very excited about the opportunity provided by Creative's worldwide distribution network, huge customer base, world-famous brands, and financial resources to maximize the technology we produce,” said Osman Kent, Chairman and CEO of 3Dlabs. “Integrating the resources and expertise of these two industry-leading companies can create a truly potent force in the graphics market. The timing of this acquisition is fortuitous, as we are close to releasing the most exciting high-performance graphics lineup in our history. We have experienced a difficult financial period during which we continued to devote significant resources to the development of our breakthrough technologies. Creative can provide us the resources and opportunity to expand our leadership position beyond the high end professional graphics market. We can leverage Creative's vast experience at building global brands and market share to take full advantage of the market potential for our upcoming release of an exciting new family of graphics processors for the desktop.”

This is a good thing for 3Dlabs, since it lacks the marketing muscle to compete with NVIDIA and ATI, even though 3DLabs has consistently produce excellent (though not always bleeding-edge) products. With 3DLabs posting a loss f nearly $5 million for Q2, the needed some infusion of cash to compete in the cut-throat consumer graphics market. While 3DLabs has typically sold to the workstation market, it's getting tough to ask premium prices for high-end graphics boards when the sub $400 boards could do nearly as much. With 3DLabs new chipset coming out and with the OpenGL 2.0 proposal, 3DLabs can certainly use the infusion of cash that the Creative acquisition will bring.

UPDATE: The acquisition was completed May 16,2002.

HONOLULU (February 1, 2002 9:33 p.m. EST) - The stars of the digitally animated movie "Final Fantasy: The Spirit Within" are being retired. The Honolulu-based studio that created the groundbreaking - but critically panned - movie announced Wednesday that it planned to close up shop, largely because the film failed to cover its $145 million cost. Aki Ross, the movie's heroine, had begun to be accepted as a real star. Last year, she posed in a bikini alongside real models in the men's magazine Maxim. Square USA, a subsidiary of Japan's Square Co., said it will shut down its powerful computers - along with Ross and other actors they generated - on March 31. But the company is still seeking an investor to keep the studio opne. Efforts to form a partnership with a major U.S. studio have failed, studio president Jun Aida said in an interview Wednesday. Aida lamented the operation's likely demise."We could have used the same characters in different roles," he said. Gov. Ben Cayetano had lauded the operation, which he hoped would form the heart of a new high-tech industry base for the islands."I'm sorry to see Square USA leaving Hawaii," Cayetano said, "but the 'Final Fantasy,' the movie that they made, was a financial disaster for them and they expressed to me that they have to consolidate and so Hawaii was one of the casualties," Cayetano said. Although panned for a weak story line, "Final Fantasy" is widely regarded as the most realistic animated film ever made. In the movie, actress Ming-Na provides the voice for Aki Ross, a beautiful scientist who is gradually succumbing to a phantom that has infected her body. The movie was the fourth-highest-earning film when it was released in theaters, but it slid after that, taking in $32 million in the United States and $72 million abroad. A DVD of the movie, in which sequences can be viewed from several angles and some scenes can be rearranged by the viewer, was released in October and was the top-selling DVD in the nation for a time. Aida said sales remain strong, but not enough to save the studio. Before closing, he said, his studio will complete a 10-minute feature tied into the sequel to the highly popular film "The Matrix." Also, "Final Fantasy" may be in competition for the first-ever Academy Award for a feature-length animated film, to be presented in March. Nominees are to be announced Feb. 12. Square USA built the high-tech studio five years ago, choosing Hawaii because of its proximity to both Hollywood and Japan, Aida said. The studio, whose parent company produces the popular "Final Fantasy" computer games, has employed up to 220 people in Honolulu during peak production. Aida said it is unlikely that many employees of the operation will be able to stay in the same business in Hawaii."Obviously, I'm very disappointed," Aida said. "But as far as what we were able to achieve in the last five years, I'm very proud that we met the challenge and were able to finish what people in the industry consider a very groundbreaking project."

Sunnyvail CA (Jan 30,2002) -SUNNYVALE, Calif. — High-end graphics designer 3Dlabs Inc. Ltd. is driving an effort to enable a new class of programmable graphics processors with an upgraded version of the OpenGL application programming interface in a chip it may introduce later this year. 3Dlabs is also spearheading a move to draft a version of OpenGL as an API for embedded systems. 3Dlabs has put forward proposals for OpenGL 2.0 that would simplify the high-end graphics interface, in effect turning it into a high-level, C-based graphics-programming language. The plan would enable software for a hybrid graphics architecture the company is designing that will meld pieces of a traditional hardwired graphics pipeline processor with programmable SIMD arrays in a device that could require 70 million transistors. The chip, which could be announced at Siggraph in August, "is enabling technology," but won't help 3Dlabs break out of its niche of high-end workstation graphics, said Jon Peddie, principal of Jon Peddie Research (Tiburon, Calif.). "It's a breakaway development from the traditional linear graphics pipeline processors we all know and use," he said. Separately, 3Dlabs has proposed a subset of the existing OpenGL 1.3 specification that could be the software basis for a simple set of 2-D, 3-D and imaging cores it would sell to embedded markets ranging from traditional avionics and automotive displays to potentially high-volume Web-enabled cell phones and handhelds."There's a real need for this," said Peddie. "Without this, graphics processing on handhelds has to be directed at the particular characteristics of the processor that is unique to that system." Once in the vanguard of graphics, the OpenGL API has stagnated, losing steam to Microsoft Corp.'s Direct3D, part of that company's DirectX suite of multimedia APIs for Windows. "Direct3D is doing very well and has pulled ahead of OpenGL in certain features, such as programmability," said Peddie. "OpenGL needs to expose programmable functions." "Future graphics chips will be more like visual processors," said Neil Trevett, vice president of marketing for 3Dlabs. "They will be less hardwired graphics pipeline accelerators and more like programmable CPUs." Graphics chip makers will also need to start developing good compilers to enable high-level languages like those envisioned for OpenGL 2.0, he added.

The OpenGL 2.0 proposal creates three classes of programmable processors inside the traditional graphics pipeline. Presumably, the programmable parts correspond to planned single-instruction, multiple-data arrays in 3Dlabs' upcoming chip architecture. Programmable vertex processors will replace coordinate transformation and lighting engines and allow arbitrary per-vertex operations. Programmable fragment processors will replace texture mapping and fog engines and allow per-fragment operations, something that developers have long wanted, Peddie said. And programmable image formats — also known as pack/unpack processors — will replace fixed-format image-packing and unpacking operations. The goal of the pack/unpack operation is to convert so-called application pixels to a coherent stream of pixel groups, Peddie said. Other features 3Dlabs has proposed include a more flexible frame buffer configuration, application control of texture memory, a unifying framework for OpenGL objects, and cleaner and more efficient synchronization mechanisms, according to a white paper written by Peddie. Part of the idea is to replace the complex burden of extensions that exist for OpenGL today with a more manageable set of programmable capabilities. An initially backward-compatible version 2.0 would be followed by a "pure" OpenGL 2.0 that "would be significantly smaller than version 1.3," said Trevett. The OpenGL Architecture Review Board (ARB) reviewed the proposal in September. Graphics leader Nvidia Corp. (Santa Clara, Calif.) has not given its OK to the concept. Trevett said he hopes the ARB will ratify a final spec at Siggraph.

Embedded API
Separately, an ad hoc industry association called the Khronos Group is reviewing 3Dlabs' proposal for a subset of OpenGL 1.3 for embedded systems. The group includes 3Dlabs, ATI Technologies, Intel, Nvidia, SGI and Sun. "You can't get good graphics in embedded devices, and one of the problems is there is no good API," said Trevett. "This could help define advanced graphics in the embedded community." Microsoft already supplies a subset of its Direct3D APIs in its latest version of embedded Windows, called CE.Net. But other embedded-OS vendors lack a common graphics API. "We'd like to have a standard graphics API because it's a big challenge for us and our customers to get value without such a standard," said Christopher Zapf, a product manager for Wind River Systems Inc. (Alameda, Calif.), which sells the lion's share of commercial real-time operating systems for embedded. Zapf said Wind River will participate in defining embedded OpenGL. The embedded OpenGL API could have as few as 50 function calls for its 2-D version and 100 for its 3-D subset, down from 350 for the entire version 1.3 of the API, according to the 3Dlabs proposal. Drivers should require less than 200 kbytes of memory, down from about 1 Mbyte for 1.3 drivers. The API will also conform to such embedded-system quality standards as the DO-178 spec. The Khronos Group, previously a closed-door association, is opening its membership to anyone interested in being part of the work groups to define the embedded API. The group hopes to start signing up members in February. Trevett hopes the expanding group can ratify a final spec before the end of the year. The OpenGL ARB has indicated it will approve the group's work.