Exclusive: NVIDIA Talks About GT300, 40nm, Ion and Tegra

During this year’s Computex 2009 show, NVIDIA took the stage with a number of new Ion- and Tegra-based products, showcased by some of the company’s partners. Despite rumors prior to the debut of the computer show in Taiwan, the Santa Clara, California-based graphics maker didn’t reveal any details regarding its much-anticipated 40nm-based graphics processing units. However, the launch of the new Ion-based PCs, as well as the highly anticipated Tegra devices, confirmed NVIDIA’s business orientation, becoming a company that is involved in more than just graphics.
While we weren’t able to make it to Computex ourselves, we looked for the chance to find out more about what the graphics maker is preparing in the near future, as well as the company’s opinion on the current computer industry. The opportunity was just around the corner, as NVIDIA was actively promoting its latest technologies around the world, holding press events to demonstrate some of the things that were on display at Computex. We took this opportunity and sat down for a one-on-one with the company’s Igor Stanek, Product PR Manager for our region. We wanted to go through some of the most important aspects of the manufacturer’s business, taking into consideration the latest rumors that surfaced on the Internet.

Mr. Igor Stanek was kind enough to provide us with a perspective on what NVIDIA is focusing on, as well as on what to expect from the future line of ARM-based Tegra devices. Unfortunately, we could not find anything about the much-rumored next-generation GPU architecture, allegedly called GT300, whose existence is still a mystery. However, we do have some useful insights into what is to come from NVIDIA. Now, without further ado, we will let you go through the interview itself.

NVIDIA is evolving from a company that was mainly focused on delivering graphics cards for computer enthusiasts and supporting the latest games in the industry to one that is set on providing solutions that could power a complete computer system. What can you tell us about that?

I.S: GPUs are used by gamers around the world to play the latest games at their best. GPUs though have become a power warehouse, thanks to their parallel architecture. The industry and the scientists have now realized that they can use this incredible parallel computing architecture to do other things than games. CPUs are no longer increasing in clock speed yet consumers are demanding more from their PCs today than ever before. In order to provide the much needed performance to deliver on these consumer expectations, the only path available is to go multi-core or parallel – i.e., add more cores and split demanding workloads across them. Due to the very nature of computer graphics, GPUs excel at doing many things at once and are ideally suited to this new heterogeneous computing environment. If you look at what Microsoft and Apple are working on with their new OSes you understand how GPUs are rapidly becoming the most desired chip inside the PC; the upcoming Windows 7 and Snow Leopard use GPU to full extent.

Scientists are using it to accelerate calculation by a factor they were never able to achieve before with CPUs. We consider ourselves a company that provides specific solutions to the resolution of complex problems; whether they are common or most complex ones in medical, astrophysics and more. We develop tailored solutions in different fields; one example is our Quadro business where we develop products that solve specific problems like SDI capture, multi display financial visualization. Dedicated 1U GPU racks, etc.

After seeing all those new Tegra-based and ARM-based devices at Computex this year, can you project a future where an ARM-based device can provide users with everyday computing?

I.S: It all depends on your definition of everyday computing. When you use a Tegra chip your everyday computing is not limited to typing email and surf the web; you expect things that you do on your regular PC but in a smaller footprint. These devices deliver desktop-class Internet browsing with Flash video and animation acceleration and high-definition video playback, all with cell phone-class power management – making days of HD mobile Internet experiences a reality.

Consumers are looking for an always-connected device for social media applications such as Facebook and YouTube, as well as great multimedia performance for recording and watching HD movies and videos on the go. Tegra includes an entire computer-on-a-chip, including an ARM 11 core processor, so to answer your question, yes Tegra, which includes an ARM process, is a fully capable and modern device that fits your communications and entertainment device in your pocket.

Putting that into perspective, can we think about a future without x86?

I.S: I don’t think anybody is saying x86 is out of the picture, but CPUs are losing their breath and GPUs becoming much more important. It’s very simple. Just check how much time you need for transcoding of video on ION and how powerful CPU you need to match ION speed. I think that future is in well balanced PC with CPU and GPU living in harmony. CPU will have more control tasks in operating system and GPU will be focused for most of visual computing staff including some heavy computational tasks. This is the architecture of the future. I believe that PC with 1000 USD CPU and Intel integrated graphics card is already history.

Tegra has been one of the main highlights at Computex this year. How important is Tegra for your future product lineup?

I.S: Tegra is a strategic line of products for NVIDIA. The world is going mobile and expectations are for low power always on, always connected and powerful devices that can do HD is growing rapidly.

Should we expect more Tegra devices to become available in the near future?

I.S: The response from our customers is phenomenal. Tegra is strategic to us and we have a great roadmap that we are sharing with our ecosystem, as I said our partners are extremely satisfied with performance, features and especially the low power used by these devices. We cannot announce products on behalf of our customers, but there are many coming

Do you think Microsoft will play a significant role in the adoption of Tegra?

I.S: Microsoft Windows Embedded CE is a great operating system and we support it with our Tegra devices. We do also provide support for Google Android.

Can you tell us something about the alleged GT300 that has been rumored in the media recently?

I.S: I am not aware of that code name.

What can you tell us about your next-generation chips, transitioning from 55nm to 40nm?

I.S: We don’t comment about unannounced products.

There were reports about TSMC’s problems with their 40nm manufacturing node. What can you tell us about that?

I.S: That is a question you should ask TSMC. We have new mobile parts being announced shortly that will be using 40nm process. It’s not all about process though; it’s also about efficiency of the architecture. 40nm process coupled with this efficient architecture works well for Notebooks, so we are launching this first on notebook.

During Computex AMD took the opportunity to showcase the world’s first 40nm DirectX 11 GPU, revealing a couple of videos with their next-generation solutions. How do you comment on that?

I.S: Simply, that we are thrilled about what we are working on. Stay tuned!

BY Traian Teglet, Technology News Editor

© 2001 - 2009 Softpedia. All rights reserved.


ImIcarus said...

Nice post.

The part about GPUs being more and more integrated as a powerhouse of the computer is definitely interesting.

From my perspective, it seemed as though CPUs were becoming so powerful that it would no longer need GPUs, and CPUs will be able to do everything on its own.

However, after reading this article, I realize it can be quite the opposite; GPUs can be the future of computers.



Copyright 2008-2009 Daily IT News | Contact Us