Developer Newsletter: Issue #13

Since it's inception in 1989, ECTS has been solely focused on interactive entertainment. It is the only place in Europe where the entire interactive entertainment industry comes together for three days to do business!

ECTS will be held at Earls Court from the 27 - 29 August, with more than 100 exhibitors. The event is expected to attract more than 11,000 people from all over Europe.

Please stop by Booth 1180, where NVIDIA will display the latest 'The Way It's Meant To Be Played' titles from the major European publishers.

For more information, please visit the official ECTS website.

NVIDIA’s SIGGRAPH 2003 show theme was "Leading the Programmable Graphics Revolution." With application partners from DCC, MCAD, medical imaging, and visual simulation, NVIDIA showed how GPU programmability and hardware rendering is accessible to all users in every market. Content once produced off line is making its way to real-time, as evidenced by NVIDIA's Dawn demo, starring in this year’s SIGGRAPH Electronic Theater. NVIDIA recently introduced the NVIDIA Quadro FX 3000 and the genlock/frame lock model, the NVIDIA Quadro FX 3000G, which made their debut and extended the industry lead for performance and functionality right out of the gate.

In addition, NVIDIA and partner Realtime Technology demonstrated large scale visualization for the masses. Using the NVIDIA Quadro FX 3000G with full genlock and frame lock capabilities, Realtime Technology delivered a fully frame locked nine-system cluster, rendering a full, pixel shaded prototype of a Porsche Carrera.

Check out some photos from SIGGRAPH 2003, where the NVIDIA presence was hard to miss. The NVIDIA booth took center stage on the show floor--it was our biggest booth yet at SIGGRAPH!

NVIDIA also provided training for over 1,000 attendees at hands-on Cg sessions, with different tracks covering both introductory and advanced topics. The advanced track featured sessions for Technical Directors with offline backgrounds, artists interested in workflow and DCC tools, and programmers interested in shader effects, such as the skin shader from the GeForce FX 5900 Dusk demo. The Cg Tutorial was the unambiguous bestselling book of the show, making it the bestselling book at both GDC 2003 and SIGGRAPH 2003. For purchasing information, please click here.

For more, check out http://www.nvidia.com/object/LO_20030702_7512.html and http://developer.nvidia.com/object/IO_20030716_7327.html.

From July 23rd – 25th in San Diego, leaders from all areas of academia and industry presented their graphics research and successes at NVIDIA U. Held immediately prior to SIGGRAPH, this event provides the ideal forum for graphics researchers, professors, graduate students, and industry leaders to share research and analyze current and future industry trends with the movers and shakers in the graphics community.

Attendance rose again in 2003, with nearly 100 attendees representing 44 universities and affiliated organizations present. Keynote presentations were given by graphics gurus John Hart (UIUC), Wolfgang Heidrich (UBC), Peter Schroeder (Caltech), Eric Haines (Autodesk), and NVIDIA’s own Chris Malachowsky, Larry Gritz, John Spitzer, and Kurt Akeley. Introductory and Advanced Cg Sessions were taught by Randy Fernando and Simon Green, and workshops highlighted the work of NVIDIA Fellowship recipients and film industry’s Industrial Light and Magic, among others.

There is incredibly important and vital work going on in academia, and we are proud to join with the leaders of this work to power next generation research.

At SIGGRAPH 2003, we announced the NVIDIA Quadro FX 3000 and 3000G workstation graphics solutions, which extend our powerful product line in the workstation space. NVIDIA Quadro FX provides the ultimate in quality, precision, performance, and programmability. CAD and DCC applications acquire a new level of interactivity by enabling unprecedented capabilities in programmability and precision. For the first time, styling and production rendering become integral functions of the design workflow, shortening the production process and enabling faster time to market.

In addition, NVIDIA professional solutions now include the NVIDIA Quadro FX 3000G, providing advanced cost-effective visualization capabilities with genlock and frame lock support. NVIDIA Quadro FX 3000G brings the power of real-time industrial realism to full-scale collaborative engineering and design reviews.

Thanks to eVGA for working with NVIDIA to provide every attendee at Meltdown 2003 with a GeForce FX 5900 Ultra graphics card.

NVIDIA's Developer Sample boards are currently shipping LeadTek's WinFast A350 Ultra, which offers exceptional GeForce FX 5900 Ultra performance, in a very sexy package. 

For more information, contact your NVIDIA account manager or visit LeadTek's web page.

NVIDIA's Board Store is now open. The NVIDIA Board Store is a secure and comprehensive online shopping site, selling a variety of graphics boards and accessories.

The NVIDIA Board Store was created as a place to get NVIDIA products at a significant discount, without the hassle of shopping around. We strive to keep a varied inventory including everything from GeForce2-class boards all the way up to our newest flagship products.

If you have not yet been approved to shop at the Board Store, please contact your account manager or the Board Store admin (jsousa@nvidia.com).

See http://developer.nvidia.com for all the latest tools to empower your projects.

We've released an updated version of the Cg Plug-in for Maya that supports Maya 5.0. This release includes full support for all features that were available in the previous release supporting Maya 4.5.

We've released the material that we presented at our SIGGRAPH 2003 Cg sessions. Material is available for each of the four tracks we covered:

• Introductory (Diffuse and specular lighting plus environment mapping)
• Advanced [Effects] (Dusk skin shader)
• Advanced [Workflow] (Cg integration in 3ds max and Maya, hands-on labs)
• Advanced [Cg for RenderMan Users] (Cg for people with offline rendering backgrounds)

Special functionality is enabled in a combination of the latest NVIDIA drivers and the GeForce FX 5900, providing an overlay of two informative graphs on top of any Direct3D application. These graphs contain important statistics about your application, helping you to identify potential bottlenecks. The graphs display sampled data in the same way a heart rate monitor would, scrolling from right to left so you can see the values of the last 256 frames. Available to registered developers now! If you're a registered developer, simply log in and you will see NVPerfHUD as one of your download options.

In case you missed it at GDC 2003, we've put up a Video-Enhanced Presentation of Kevin Bjorke's "Cinematic Effects" talk. This presentation combines video and audio with slides, allowing you to conveniently jump from slide to slide and view demonstrations. We would love to get your feedback.

The latest Detonator FX drivers (version 45.23) are now available.

• New performance and compatibility fixes – please consult the v45.23 Release Notes
• The industry's only true Unified Driver Architecture (UDA) - supports over five generations of NVIDIA GPUs - from TNT to GeForce FX - all in a single driver binary
• The industry’s best Microsoft® DirectX® 9 support
• Ultimate “The Way It’s Meant to be Played” experience
• nView™ 2.0 multi-display technology increases productivity and provides the necessary tools to control the way users access and view applications.
• NVRotate™ enabling image rotation on today’s advanced flat panel displays.
• NVKeystone™ allows for real-time image correction on portable projectors and heads-up displays.
• Digital Vibrance Control™ 3.0 allows users to adjust color controls to achieve accurate, bright colors under all conditions.

Achieving high performance is all about removing bottlenecks-which really means that you have to balance every piece of the pipeline: the CPU, the AGP bus, and the pieces of the graphics pipeline in the GPU. The decision to use a vertex shader or a pixel shader depends on a few factors:

• How tessellated are your objects? You may want to lighten the load on the vertex shader if you have millions of vertices in each frame. This is especially true if you're using a multipass algorithm.
  
• What resolution are you targeting? If you expect your application to be run at higher resolutions, the pixel shader is more likely to become the bottleneck. So, you may want to push more computations to the vertex shader.
  
• How long are your pixel shaders? If you're doing complex shading, the pixel shader will probably be your bottleneck. If your pixel shaders compile to more than 20 instructions (on average) and occupy more than half the screen, your application will likely be pixel shader-bound. So, look for opportunities to move calculations to the vertex shader. For example, you can move from world space to light space for attenuation. Or, if you're doing bump mapping, you can make the move into tangent space per-vertex, unless you're doing per-pixel reflection into a cube map.

The most popular pages and downloads over the past few weeks have been: