nTersect |
- GPU Tech Conference - A Thank You
- GPU Technology Conference – Day 3 Video Recap
- GTC Closing Keynote Takes GPUs Into the Future with Robot Cars
- Sportvision Augmented Reality Enabled by GPU Computing
- CUDA at GTC Day 3
- Startups Come Under the Spotlight at Emerging Companies Summit
- Sounding off on 3D Audio
- Cancer Research and Supercomputing
- No Smoke Here: Fireside Chat with Jen-Hsun Huang and Quentin Hardy
| GPU Tech Conference - A Thank You Posted: 23 Sep 2010 08:00 PM PDT Three jam-packed days. Attendees from 50 nations. 120 sponsors and exhibitors. 200 black-shirted employee volunteers. 280 speakers. And one balloon-crafted NVIDIA logo. GTC 2010 by the numbers was impressive. But it's impossible to put a figure on its depth of content, its intensity of conversation, its spark of ideas.
This week's event – with all of its energy and imagination -- would simply never have happened without so many of you contributing in so many ways. It was, in some ways, the ultimate act of parallel processing. On behalf of all of NVIDIA, I want to thank each of you for infusing this event with a passion that lifted it beyond what most of us encounter in our business lives. With a bit of a break, we look forward to beginning to plan GTC 2011. And we can't wait to join you in doing this all over again. |
| GPU Technology Conference – Day 3 Video Recap Posted: 23 Sep 2010 07:11 PM PDT GTC Day 3 kicked off with a fireside chat between Jen-Hsun Huang and Forbes national editor Quentin Hardy, covering everything from fundamental changes in the computer industry, to skills needed by the next generation of entrepreneurs, to products that will be hot for the holidays. Sebastian Thrun, robotics pioneer at Stanford and distinguished engineer at Google, closed the show with a passionate closing keynote. Much happened in between. In this video, senior VP Dan Vivoli recaps the day, and the entire show.
|
| GTC Closing Keynote Takes GPUs Into the Future with Robot Cars Posted: 23 Sep 2010 07:08 PM PDT A furious three days at GTC concluded on a soaring note as Stanford's Sebastian Thrun gave a fascinating, at times comical, look at the evolution of driverless cars. Taking the stage to the strains of Gary Numan's "Cars," Thrun took a packed house through a litany of sorry statistics. Cars are the leading cause of death for young adults. They consume a third of our nation's energy. They are only used for 3 percent of their life. And the highways they drive on are only about 6 percent fully utilized. The way to do better is with GPU-based technology, which helps drive the computer vision that robotic cars depend on. Thrun's research began in 2004 when DARPA, the research arm of the U.S. military, sponsored a 150-mile challenge race for robot cars. The initial challenge was nothing short of disastrous, with no car making it more than seven miles. Most failed right out of the gate. The next year, Thrun entered the challenge with a team of students and a car called Stanley (which now lives at the Smithsonian and was top of the heap in Wired Magazine's 50 best robots of all time). Taking us through Stanley's evolution, Thrun touched on the incredibly complex engineering feat of creating a driverless car. The solution involved wedding computer vision through lasers and cameras with machine learning algorithms so that Stanley could find the road was and drive confidently along it. Showing some nail-biting footage of the race itself, Thrun included an unforgettable blooper reel of robot cars in confused, hapless encounters with traffic cones, hay bales and makeshift tunnels. One underdog robot motorcycle was a crowd favorite, and it was almost hard not to cheer when Thrun ran video of it finally staying upright. Profound challenges remain in creating driverless cars, although the sight of a robotic car executing a flawless parallel parking maneuver indicated that the day may not be too far off. Thrun noted that advances in GPU computing could bring the day closer. He closed with a list of what he called "2x challenges" – among them, making cars twice2x as safe, 2x as efficient, and to live with only half the cars we have today.
|
| Sportvision Augmented Reality Enabled by GPU Computing Posted: 23 Sep 2010 05:46 PM PDT A great opening highlight reel perked up a late-afternoon crowd in the "Enabling Augmented Reality with GPU Computing" session, which showed sports broadcast moments enhanced by Sportvision augmented reality (AR). Ryan Ismert, Core Technology lead at Sportvision, took session members behind the scenes of on-air graphics such as the company's famous NFL "1st and Ten" yellow line.
AR imposes a layer of computer graphic information onto real-world environments. If you've watched pro sports in North America, then you've seen Sportvision AR effects in action. The company – which works with ESPN, Fox, ABC, NBC, and others – is perhaps the single largest AR provider in the sports world (based on last year's football broadcast of 106 million viewers) and provided visual effects for the NBC broadcast of the 2010 Olympics. Ismert described how Sportvision outfits broadcast cameras with specialized hardware to capture and process information, apply and output the effects –with a delay of only about 30 miliseconds, or, less than 3 fps. The extremely high standards for sports broadcast TV, make GPUs essential to the process. Joking that what his talk was becoming a "shameless shill" for NVIDIA Quadro, Ismert showed a production pipeline slide in which the only aspect not touched by GPUs was the camera data itself. Looking ahead, Ismert discussed how GPUs will allow AR to spread pervasively to smartphones and other mobile devices. Already, some of this technology is in the works, and Ismert predicted it won't be long before fans in stadiums will be able to get enhanced views and video replays delivered directly to their phones. |
| Posted: 23 Sep 2010 04:18 PM PDT A major theme emerging from this week's GTC is the staggering – and increasing – breadth of applications that CUDA is enabling. Tuesday, we highlighted sessions that expanded on CUDA's use in medical applications; Wednesday, the efficiences developers are gaining through CUDA libraries. Today, we take a look at some of the sessions covering CUDA in high performance and large scale computing:
|
| Startups Come Under the Spotlight at Emerging Companies Summit Posted: 23 Sep 2010 03:00 PM PDT The chief execs of two dozen startups – focused on areas ranging from robotics to cloud-based video rendering – showed their stuff at the Emerging Companies Summit, pitching their companies in eight minutes flat and taking another eight minutes of question.
Subjecting them to questions were members of a rotating panel drawn from VC firms, as well as industry analysts and corporate developments execs. CEOs from Scalable Display Technologies, RTT, Aqumin, Otoy, Universal Robotics, and ICD took to the stage this morning, in front of judges including Savitha Srinivasen of IBM Research, leading industry pundit Rob Enderle, Norman Winarsky, of SRI International, and Jeff Herbst, who runs business development at NVIDIA. Here's a quick breakdown of what they're bringing to market: Scalable Display Technologies, of Cambridge, Mass. – CEO Andrew Jamison portrayed a compelling vision of giant displays comprised of many individual sources to facilitate corporate collaboration. He said his software is being used by Fortune 50 companies to display information for presentations and meetings. The technology, originally developed at MIT, allows curved displays at a resolution of 5,000 x 1,000 pixels, up to 25 feet wide. Realtime Technology (RTT), of Munich – CEO Jeroen Snepvangers said his company helps clients use their 3D data to deliver visualization solutions for product development and marketing. Clients, including Harley Davidson and Nissan, use RTT technology to allow customers to select configuraton options on a computer, in some cases for vehicles still in development. Another client, BMW, used RTT software to produce a commercial showing hundreds of CADCAM images of the vehicle's components melded together. Aqumin, of Houston – CEO Mike Zeitlin's company is working on solving large data problems by converting huge spreadsheets of financial data into 3D graphs which depict market changes in real-time. The two-year-old company is working with partners like Bloomberg, CNBC, Deutsche Bank and Fidelity to bring GPU-accelerated technology to market traders. Otoy, of Los Angeles – CEO Jules Urbach showed an impressive video of an upcoming movie called Gaiking, which features Transformers-like CG effects rendered from the cloud. Otoy's technology allows them to deliver "unlimited rendering power" for film and video games with up to 1,000 GPUs from the cloud. It promises major disruption in the film industry, which typically creates movies using massive CPU-based rendering farms. Universal Robotics, of Nashville, Tenn. – CEO David Peters described how robots can increase efficiency and reduce injuries by providing intelligent materials handling. While today's industrial robots must often be programmed for specific actions, Universal Robotics uses software called Neocortex which allows the robots to learn how to move packages of different shapes and sizes. Innovative Converged Devices (ICD), of Seattle, Wash. - CEO David Hayes likened ICD to a new generation of ODM that can reduce the time it takes to bring new products to market. ICD provides rapid product development of Tegra-based mobile devices while reducing cost by up to 20 percent. Hayes showed a product design called the Floob, a very thin device with a 3D interface for social media and movies streamed via Netflix. |
| Posted: 23 Sep 2010 02:00 PM PDT There's a natural tendency at GTC to focus on 3D visuals, but 3D audio is also emerging as a topic here. Sound engineers who work to produce 3D audio – commonly thought of as "surround sound" – can leverage GPUs for the processing required to simulate sound in actual spaces, with implications for gaming, home entertainment systems and film. In a talk for audio professionals, Nicolas Tsingos, of Dolby Laboratories, described the inroads GPUs are making in 3D audio rendering and simulations. His presentation, "Interactive 3D Audio Rendering Systems" covered the ways that GPUs can be used in audio processing and in creating simulations of the way acoustics sound in three-dimensional environments. Starting with an overview of GPUs and parallelism, Tsingos showed how he and other acoustical researchers have applied many-core processors to acoustical simulation challenges, and then shared optimization secrets they've learned along the way. They have found that, depending on the scenario, GPUs can accelerate processing speeds by 2x to 300x versus using CPUs on their own. The key to achieving faster performance is in writing to the GPU's strengths, he noted. Because GPUs are well suited to processing large chunks of data, Tsingos advised session members that they would see better results by increasing data per transfer and reducing the number of transfers. Attendees showed keen interest when Tsingos showed movies of various acoustical simulations. One displayed differences in sound, according to changes in an area's contours. Another showed acoustical modeling of an echo from a Mayan temple in Mexico. Several dozen audience members stayed behind for extended Q&A, which ranged from acoustical design and the "computing stereo of the future" to microphone array processing in real time. |
| Cancer Research and Supercomputing Posted: 23 Sep 2010 12:39 PM PDT Watching the session "Application of Tesla C2050 to Flow Cytometry and Cancer Detection," led by Bob Zigon, of Beckman Coulter, I got thinking: How many problems could we begin solving if we had access to an infinite amount of computation capacity? Zigon's product, Kaluza, enables doctors to detect cells in the body that aren't functioning correctly, typically cancer cells. Powered by Tesla GPUs, Kaluza can run 400x faster than a CPU-only based solution, enabling doctors to look for cancer cells in real-time with more accuracy and data from each cell. "Results from our product will lead to better diagnostics for detecting leukemia in blood cells," said Zigon, rather humbly. He and his team use lasers to analyze cells. Through light diffraction, Kaluza can measure attributes of the cells, their internal and external characteristics, and, more importantly, the presence and absence of certain proteins. With Tesla GPUs, Kaluza has vastly greater, faster visibility into each cell. Why does this matter? Well, I'll answer that with a question: How would it change your life, or lives of people you love, if you could just walk into a doctor's office, get a quick screen for cancerous cells, and walk away knowing the results? Zigon's final, somewhat indirect, request to NVIDIA was to "just build hardware that is infinitely fast and let me look at every cell." His dream is that infinitely capable computers will detect all cancer cells in every human being on earth, before these cells become dangerous. While GPUs aren't infinitely fast, (and never will be), they are enabling some amazing work. And I'm glad to be part of the solution. |
| No Smoke Here: Fireside Chat with Jen-Hsun Huang and Quentin Hardy Posted: 23 Sep 2010 11:35 AM PDT Jen-Hsun Huang, NVIDIA's co-founder and CEO, just wrapped up a thought-provoking discussion about the future of the computer industry and NVIDIA's place in it. Billed as a "fireside chat" with Forbes national editor Quentin Hardy, the wide-ranging conversation covered everything from the fundamental changes in the computer industry, to skills needed by the next generation of entrepreneurs, to what products will be in demand for the holiday shopping season.
Here are some highlights: NVIDIA's fundamental mission. "As opposed to other companies, whose mission was to build the fastest chip or most powerful workstation," Jen-Hsun said, "our perspective was that our work was a medium for artistic expression of all kinds." That expression could be video gaming – or it could be research. "Product comes and goes, but the fundamental purpose of our company should be everlasting." On the virtue of interdisciplinary skills. Sought-after new hires of the future won't be just specialists. They'll be wide-ranging experts. Jen-Hsun returned to this idea several times. In the interdisciplinary world, computation or IT is added to traditional fields, resulting in new growth areas, like bio-informatics. The intersection of disciplines is where "the interesting stuff is happening." Following up on the importance of being able to move among various disciplines, he talked about his own journey from "narrow and deep" to something broader. "Training for CEOs is overrated. I'm sure there are companies – ours is among them –that have lost billions on the education of the CEO," he quipped. The disruption ahead. "Over the course of the next five years, we will see an enormous disruption in the industry," he predicted. The same consumer activity – like, watching movies – will be fulfilled in different ways by different companies. People will have to ask themselves where they will buy computers – Best Buy? AT&T? Or a service provider? "Everything we know about the computer industry will change based on that answer," Jen-Hsun said. Surviving that disruption. Jen-Hsun talked about who will survive and thrive in the future. "Companies don't go out of business because they run out of cash," he said. "The reason most companies go out of business is they run out of ideas." When the only constant in the business is change," Jen-Hsun said, "you want people in your company who are comfortable operating in an environment of ambiguity." Survival, he added, means being passionate about the mission and fellow team members. NVIDIA's focus for the future. Key focus areas for NVIDIA's future will be visual, parallel, and mobile (not just low power, but also systems that have sensory systems attached to them, and are situationally aware). NVIDIA used to be fully focused on visual computing, but in recent years, its focus has shifted to the point where it's two-thirds on parallel computing. What's hot for the holidays. Jen-Hsun predicted that Motorola, Samsung, and HTC would be "going nuts" trying to fill demand for a new category he called "superphones," slightly larger than iPhones but smaller than a tablet. Tablets are already upending the PC industry. "They're disrupting the marketplace faster than anybody thought. We need to get other tablets into the marketplace to fill niches and capabilities that the iPad doesn't provide." His own technology preferences. "I carry a PC – and I've been carrying it around all week," Jen-Hsun said, "but I haven't booted it yet. The question of whether I should bring a PC is going to disappear in another year." He held up a smartphone saying that it was a "foregone conclusion" that computing of that size was the future. Asked what was on his wish list, he wrapped up with the line that got the session's biggest applause. "I'm dying to have a Tegra phone, a Tegra tablet, Tegra set top boxes. I want to have a Tegra home." |
| You are subscribed to email updates from The NVIDIA Blog To stop receiving these emails, you may unsubscribe now. | Email delivery powered by Google |
| Google Inc., 20 West Kinzie, Chicago IL USA 60610 | |
No comments:
Post a Comment