Instead, that Sunday we were awaiting the arrival of some amazing new silicon chips from Taiwan to SFO, as San Francisco’s airport is universally known. With their touchdown, our 22-day race to bring an incredible mobile processor to life in time for Mobile World Congress (MWC) on Feb. 25 would begin.

Grey’s Anatomy

Behind every great microprocessor is a great team. Most people, however, don’t realize just how big that team is, how many months of exhaustive planning is involved, or how much meticulous effort is put into changing a piece of silicon into a working product.

The Super Bowl was no match for our attention.

Even before our manufacturing partners at TSMC produced those first chips, hundreds of NVIDIA hardware and software engineers had been hard at work designing the Tegra 4i (code-named “Grey”) – our first integrated mobile processor.

The stakes were high: this would be our first product with a built-in modem. Not just any modem, either. The programmable “soft modem” technology we acquired with the purchase of Icera, back in mid-2011, promises to bring next-gen wireless networking into the mainstream smartphone market.

In an industry where integrating newly acquired hardware and software could take years, we had built Icera technology into Tegra in just 20 months, with Tegra 4i.

Every clock cycle mattered.

The Bringup

Modern semiconductor manufacturing isn’t a job you can outsource to a far-off factory and forget about. It doesn’t run smoothly by chance, but by many engineer-years of planning. It’s a tightly choreographed operation involving hundreds of engineers who not only do the upfront designs of a new part, but who painstakingly turn these designs into a working product once the part arrives from the manufacturer. We call this process the “bringup.”

Like scores of my colleagues, I was prepared to give up sleep, free time and weekends to turn Tegra 4i into a product we could unveil at MWC. We’d challenged ourselves to tackle this effort – which typically lasts many months – in just over three weeks. Our priorities were clear: once the first parts showed up in the lab, we abandoned the big game for the power-on.

The bringup was a huge undertaking. For Tegra 4i, we flew more than 120 of our engineers from all over the world to our Santa Clara headquarters. Each had a list of tasks to execute to prove that our new Tegra chip would work perfectly. Every feature was tested. Every component was stressed to its absolute limits to make sure no matter what you want to do with your Tegra, it would work as specified.

We not only pulled in the bringup schedule, but set goals for better performance. If we did our jobs well, we could save precious days, showing off everyone’s work to the world earlier than we’d planned. Often, we found that we could beat our early estimates for how fast we could make Tegra run. So, when we finally had the first hardware, we worked to squeeze every last drop of power out of our design.

Bringup was a collaborative effort.

Coming Into Focus

Back in the lab that first evening, the excitement was palpable. More and more of the chip was coming alive. The displays were showing within an hour; the system memory team worked hard to tune settings for stability; and the touchscreen responded to input.

A few hours later, Linux – the core system underlying Android – booted for the first time. The next morning, Android was running. A few hours after that, our very first integrated software modem sent a text message to our CEO. Operating in shifts nearly 24 hours a day, we were turning around weeks of work in just hours.

The dream of having Tegra 4i in time for MWC was starting to come into focus, but there was still plenty more to do. Our system engineering team worked around the clock with our manufacturing team to build enough “Phoenix” reference phones to show off the functioning processor. The software team spent day after day enabling new features. And the hardware team continued their work to tune everything to work the best it could.

Engineers worked around the clock to ready phones.

The culmination of all this work was a product that we’re extremely proud of.  Our design team does an incredible job of creating products long before we begin manufacturing them. With the Tegra 4i,  we showed that our bringup team is one of the best in the industry.

When MWC started, we were ready – not with a model phone in a sealed box looping a video, but with a genuine article that the public and the press could touch. An amazing 22 days after the first plane landed in San Francisco, we showed the world what we’d built.

Such was Anne Elster’s motivation when she resolved several years ago to start working on real-time snow simulations.

It’s probably important to point out that Elster, who was talking about her work at the GPU Technology Conference, is an associate professor at the Norwegian University of Science and Technology.  Rumor has it Norway gets a bit of snow.

Elster thought focusing on snow would be a fun project for her students, allowing them to combine graphics processing with powerful simulations. She also figured it would be a nice tool for recruiting more post-graduate students.

The precursor to the project was an effort a decade ago by Elster and her students to run real-time simulations of the behavior of smoke using a dual-core laptop. Their goals at the time seem tame today: 20 frames a second, about a third of the current standard for real-time simulations.

Fast forward to 2007, and Elster started the snow-simulation effort by parallelizing on a multi-core laptop. But she eventually saw that GPUs would be a better choice for providing the required computing power.

The following year, she and her students were able to simulate several million snow flakes as particles, complete with wind field interactions, all made possible by GPUs.

Elster and her students have since expanded their simulations to work on including features such as avalanches, geysers and waterfalls. Now, they’re working on incorporating more realistic terrains, and are considering adding ray tracing to simulate how snow interacts with light.

Five of the most promising companies bagged “One to Watch” awards, splitting a pot of $85,000 worth of prizes, at a sumptuous party concluding the event. Winners included:

• Fluid, makers of a cloud-based, on-demand virtual merchandising suite that incorporates product customization and 3D dynamic imaging.
• Morpheus Medical, creators of the first non-invasive heart disease diagnostic tool.
• Oculous VR, originators of a virtual reality headset for immersive, stereoscopic 3D gaming.
• Splashtop, providers of high-performance remote desktop applications, and the first two-time winner of an ECS award.
• SynerScope, inventors of an extreme-scale analytics tool that let users easily explore huge data sets using interactive images.

NVIDIA’s Jeff Herbst, who serves as VP of business development, welcomed attendees from around the world to our fourth annual Silicon Valley showcase for startups that are using the massive computing power of GPU technology. The companies also included those engaged in mobile computing, game development and cloud-based graphics.

Five companies nabbed One to Watch awards at ECS 2013.

From the rapid-fire “CEO on Stage” event – where execs made their best pitch to a panel of expert investors, analysts and tech leaders, who then grilled them on details – to panel discussions, ECS offers a place to share ideas, network and learn within the GPU ecosystem.

NVIDIA today rolled out a new set of beta drivers optimized for these titles, just two weeks after the release of the GeForce R313 Game Ready drivers for Crysis 3.

GeForce 314.14 beta drivers boost performance by up to 23 percent and are now available for automatic download and installation using GeForce Experience. That makes it easy to stay up to date with each new GeForce Game Ready Driver.

The new drivers – and our new GeForce Experience service – are examples of how we’re working hard to deliver more immersive gameplay to millions of gamers. We want to thank the million PC gamers everywhere who have already downloaded the open beta of our GeForce Experience and are helping us make it better.

Here are some examples of performance increases in GeForce 314.14 drivers:

GeForce GTX 680:

• Up to 23% in Sniper Elite V2
• Up to 9% in Sleeping Dogs

GeForce GTX 680 SLI:

• Up to 22% in Sniper Elite V2
• Up to 14% in Sleeping Dogs
• Up to 9% in StarCraft II
• Up to 5% in Call of Duty: Black Ops II
• Up to 5% in Deus Ex: Human Revolution
• Up to 5% in Just Cause 2
• Up to 4% in The Elder Scrolls V: Skyrim
• Up to 4% in Batman: Arkham City

For more details, refer to the release highlights on the driver download pages and read the GeForce driver article on GeForce.com.

Enjoy the new GeForce Game Ready drivers and let us know what you think. 

Work being done at the ten new centers includes three-dimensional genome sequencing for cancer research, graphics and numeric programming, and design of programming and runtime systems for heterogeneous nodes and clusters. There are now CUDA Research and Teaching centers in 42 countries.

CUDA Teaching Centers equip tens of thousands of students graduating each year with the knowledge and expertise to take advantage of the parallel processing power of GPUs (see “What Is CUDA?”). They get free teaching kits, textbooks, software licenses, NVIDIA CUDA architecture-enabled GPUs for teaching lab computers and academic discounts for additional hardware.

CUDA Research Centers embrace GPU computing across multiple research fields. They have access to exclusive events with key researchers and academics, a designated NVIDIA technical liaison and specialized training sessions.

Here are some examples of CUDA-related work taking place at some of our newest CUDA Research Centers:

Baylor College of Medicine/Rice University (U.S.)

Baylor College of Medicine and Rice are working on the analysis of the three-dimensional structure of the human genome. They will be using CUDA technology to increase the throughput of genomic analyses, to accelerate physical simulations of genome folding, and to enable a wide range of high-throughput biophysical assays.

Indian Institute of Technology, Delhi (India)

The new Exascale Research Lab IIT Delhi will provide advanced ongoing research, testing, and technology development in a variety of areas including processor architecture, circuits, memory architecture, high-speed signaling, programming models, algorithms and applications. The lab will also apply its work to address next-generation scientific challenges in computer science, nanotechnology, material science, power engineering and other domains.

Technion (Israel)

The CUDA Research Center at Technion runs as part of the Technion Computer Engineering Center, an international computer engineering research and education center. The CUDA Research Center was established to educate, research and support the development of heterogeneous systems using NVIDIA technologies in areas such as computer architecture, parallel systems, graphics and geometric computing, geometric image processing, image processing and more.

University of Missouri (US)

The CUDA Research Center at University of Missouri was established to support a variety of research activities employing CUDA technology. Two key research thrusts: the design of runtime systems for CPU-GPU clusters, and the study of programming models and parallelization mechanisms to allow the deployment of irregular applications on GPUs. Recently, it was awarded a grant by the National Science Foundation to design scheduling and virtualization technologies to efficiently enable many-core devices in cluster and cloud environments.

The new CUDA Teaching Centers include:

• Indian Institute of Technology Gandhinagar (India)
• Laboratório Nacional de Computação Científica (Brazil)
• Nanyang Technological University (Singapore)
• Tsinghua University (China)
• University of Houston (US)
• University of Science and Technology of China (China)

Separately, we are excited to announce that we’ve upgraded our Center Rewards Program. Anyone at one of our CUDA Centers can receive a 15 percent discount on all the latest Tesla Kepler GPU accelerators from any preferred solution provider. The CUDA Center Reward Pricing applies to Tesla K20X for servers, Tesla K20 for servers and workstations and Tesla K10 for servers (some limitations will apply). To verify if your institution is a current CUDA Center, check here.

For more information on NVIDIA research activities and these programs, please visit the NVIDIA Research site.

Every Christmas Eve, the North American Aerospace Defense Command (NORAD) reports on the whereabouts of Santa Claus on “NORAD Tracks Santa,” at www.noradsanta.org, as the jolly elf makes his rounds delivering presents to the world’s children.

This year’s NORAD Santa Tracker was put together with some help from Analytical Graphics, Inc. (AGI), a 250-person company that develops analysis and visualization software for the aerospace, intelligence, and defense communities.

Patrick Cozzi, a senior software developer for AGI can’t say much about how NORAD uses his company’s software.

But he can talk about his firm’s role in NORAD’s Santa tracking efforts, which, for the record, rely on the same set of satellites, fighter jets, and powerful radar installations the joint U.S.-Canadian defense organization has used to monitor the airspace over North America since 1958.

That network is supplemented by ‘Santa Cams,’ that allow NORAD to offer an up-to-the- moment report on Santa’s movements to the world’s children.

No plugin required.

This Christmas Eve NORAD will rely on an open-source WebGL globe and map engine Cozzi and his colleagues started called ‘Cesium’ to show Santa’s whereabouts as he makes his rounds.

WebGL taps into GPUs to render images in a web page (for more details, see Cozzi’s detailed explanation, here). That allows NORAD to serve up maps with jaw-dropping realism. No special plugin required.

You won’t need dual GeForce GTX 680 GPUs, either. Even an aging GeForce 8800 GT will allow users to watch Santa’s progress on a 1280 by 1084 display at 55 frames per second, Cozzi says.

But if you want a present, you’ll still need to be nice. No matter what GPU you’re running.

It takes an artist’s eye to notice details like this. And it takes a hard-core coder to do something about it. Tim Lottes is both. Maybe that’s why the NVIDIA engineer’s technology has been built into hundreds of games. Lottes specializes in a field of visual computing known as ‘anti-aliasing.’ The aim is to smooth the jagged lines created when putting an image on a screen composed of dots, known as pixels.

His latest creation is an anti-aliasing technology dubbed TXAA (temporal approximate anti-aliasing). The NVIDIA-only techology, which debuted in July, gives games running NVIDIA’s latest Kepler-based GPUs a more cinematic feel. In the five months since its debut, TXAA has been built into “The Secret World,” “Assassin’s Creed III,” and “Call of Duty: Black Ops II.”

It helps that Lottes knows how to think like a game developer. Before joining NVIDIA, he worked at computer game developer Human Head Studios.

For years, the way graphics were generated for movies and videogames were different, he says. Now, developers are looking for technologies that mimic the look of a Blu-Ray movie. “My personal goal is to get games up the quality of a feature film,” Lottes says.

That’s a goal propelled forward by the 34-year-old’s passion for landscape photography. You begin to notice small details when stitching together realistic panoramas out of high-resolution images, he says.

Lottes also has a knack for writing ‘low-level’ code. He likes stuff that’s as close to the bare metal in a machine as you can get. “I was writing assembly language before I was a teenager,” says Lottes, who holds a degree in Computer Science from the University of Illinois, Urbana-Champaign.

So when he got the chance to join NVIDIA three years ago, he jumped. He soon began working on an anti-aliasing technology now known as FXAA (Fast Approximate Anti-Aliasing). Inspired by a technique that ran slowly on CPUs, known as ‘MLAA,’ Lottes  aimed to come up with the fastest possible anti-aliasing technique for GPUs.

Rather than using hardware ‘multi-sample anti-aliasing’ (MSAA) — a process that takes place as an image is rendered — FXAA smooths the edges on the screen after all the rendering is finished. The result: more life-like images on older, slower hardware. The technology is now used in many of the top tier video games for the PC, PlayStation 3, and XBox 360.

Lottes wasn’t satisfied, however. His latest creation, TXAA aims to remove the crawl seen on the edges of objects and inside many shaded surfaces that appear as characters move through a virtual world. “I wanted to take what the film guys have been doing with anti-aliasing and get as close as I could to that,” Lottes says.

Not easy, considering filmmakers can rely on banks of servers working days, or weeks, to generate life-like effects. To do that Lottes exploits his knowledge of NVIDIA’s  GPUs to craft a compact ball of code that can deliver superior image quality with a smaller performance impact. The trick? “I do the right amount of work, and I don’t do too much work,” says Lottes, who is already working on another round of improvements to TXAA .

For many enterprises, Big Data can mean big confusion.

Fuzzy Logix, a North Carolina-based startup, is helping them make better sense of it and predict the future by deploying GPU-based technology.

Imagine a brokerage that has only milliseconds to decide whether a large investor’s big trade will be within margin requirements. Not just the requirements at the moment of the trade, but accounting for all projected trades throughout the rest of the day.

Or a call center determining which rep should field an incoming call based on the demographics of the caller’s location, past interactions and who will provide the best customer experience. And doing all this while also providing suggestions for the next likely purchase – before the second ring.

From marketing and finance to sales and customer service, there are literally thousands of applications for predictive analytics in enterprises. Fuzzy Logix is focused on making the technology adoptable and pervasive.

“We do this,” says Michael Upchurch, chief operating officer at Fuzzy Logix, “by moving analytics out of the hands of specialists and specialized applications and making them broadly available to business decision makers.”

And they make it lightning-fast by using GPUs. “Working with our clients, we are consistently identifying general business challenges that can be best solved by leveraging GPU-based analytics,” says Upchurch.

To date, the company has ported the code base of more than 500 in-database analytics models to run in GPUs. With these essential building blocks at the ready, every Java, C/C++, .NET or other programmer in the world can build statistical models faster and spend their effort writing higher order models.

Since the models can be run using common programming languages, they are easily deployed to model runners, such as marketing and finance executives and salespeople in the field. These non-statisticians can run analytics on demand to meet their immediate needs, such as forecasting sales, optimizing bid prices for online ads and projecting customer turnover.

Fuzzy Logix also offers a deskside or rack-mountable appliance that taps into an enterprise’s computing infrastructure to grab a data file and then processes the information locally using GPUs.

The benefits to this approach are dramatic: it reduces the queues that occur when using CPU-based systems, lessens the demands on model builders and lets end-users make timely decisions because they can get answers they need fast.

How fast? The Fuzzy Logix appliance with four NVIDIA Tesla cards is able to perform 1 billion calculations in 13 milliseconds. This is speed a CPU-only system simply can’t match, performed at a fraction of the cost.

By providing this enormous processing power to businesses, and giving them control over an easily adopted analytics environment, Fuzzy Logix is making desktop supercomputing real.

One in a hundred newborns suffer from congenital heart disease. 5.2 million Americans of all ages suffer from heart failure. The challenge for all of these patients: proper diagnostics.

Transthoracic echocardiograms – which involves placing a specialized transducer against the patient’s chest wall– are non-invasive, but imprecise.

Inserting a probe equipped with an ultrasound transducer down through the mouth and into the chest cavity – a procedure known as a trans-esophageal echocardiogram – is more precise, but it can be painful, risky, and time-consuming.

Morpheus Medical, a five-person startup in San Francisco may have a solution: software that uses magnetic resonance imaging (MRI) to build a real-time model of a beating heart.

Using data gathered during a 10-minute MRI scan the startup’s software can build a model of a heart in full 3D with flow and function while pulsing, and the arteries around it, that doctors can play forwards and backwards, examine from every angle, and zoom in on to get the details they need to know.

“It allows us to replace all those invasive procedures with something that is non-invasive and much shorter,” Morpheus co-founder Fabien Beckers says.

The aim is to offer a tool that can be used to help patients suffering from congenital heart disease – something challenging to diagnose with newborns – and then address the growing numbers of adults suffering from trouble with the valves that move blood around the heart.

Like so many startups in Silicon Valley, the company was hatched at Stanford University, where Beckers – a Stanford Graduate School of Business student who holds a PhD in physics from Cambridge University – ran into John Axerio-Cilies, who was working on his PhD in fluid dynamics in 2010. The two pitched the idea behind Morpheus as part of a business-school class project.

The pair was quickly joined by another pair of co-founders who were working on ways to change the paradigm of cardiac MR imaging: Albert Hsiao, who holds a PhD in bioengineering from UC San Diego and a computer science degree from Caltech; and Shreyas Vasanawala, an associate professor at Stanford University who specializes in pediatric and abdominal MRI.

Albert and Shreyas had already begun a collaboration with NVIDIA’s workstation group back in 2010. As part of NVIDIA’s Fermi launch, a prototype was developed and showcased at various conferences including GTC 2010.

So far the company is being backed by angel investors, and the four are now working to turn their prototype into a product. One challenge: finding technology powerful enough to render the results in real time. For that, Morpheus turned to workstations equipped with NVIDIA’s Quadro K5000 GPUs. “Without them we couldn’t visualize the results,” Beckers says.

Making more accessible the skills that employers need most is key. So online education startup Udacity is working with NVIDIA and other technology companies to provide free technology courses to people around the world.

The first class we’re creating with Udacity is an “Introduction to Parallel Programming,” led by David Luebke, who helped found NVIDIA Research six years ago, and John Owens, associate professor of electrical and computer engineering at UC Davis.

Students will learn to use the CUDA programming model, take part in hands-on exercises and test their knowledge with quizzes – all delivered online via (what else?) cloud-based GPUs. The course kicks off early next year, but you can enroll today at http://www.udacity.com/overview/Course/cs344/CourseRev/1.

Check out a brief description of the course here:

Plus, if you want to quickly visualize a new algorithm or program, you’ll be able to do it right in your browser thanks to new computational tools Udacity is integrating into its courses through a partnership with Wolfram Research, creators of the technical computing software Mathematica and the online search and answer engine Wolfram Alpha.

“Technologies change quickly, and universities can be slow to react,” said Udacity’s Andy Brown. “We’re working with companies like NVIDIA to close the gap between the skills graduates have and the skills employers need.”

We expect the “Introduction to Parallel Programming” course to be a boon to students, researchers and people already working in industry, so sign up today!