Reviewer's Guide

Reviewer’s Guide NVIDIA® GeForce® GTX 280 GeForce® GTX 260 Graphics Processing Units

TABLE OF CONTENTS

NVIDIA GEFORCE GTX 200 GPUS...... 3 Two Personalities, One GPU ...... 3 Beyond Gaming ...... 3 GPU-Powered Video Transcoding...... 3 GPU Powered Folding@Home...... 4 Industry wide support for CUDA...... 4 Gaming Beyond ...... 5 Dynamic Realism...... 5 Introducing GeForce GTX 200 GPUs ...... 9 Optimized PC and Heterogeneous Computing...... 9 GeForce GTX 200 GPUs – Architectural Improvements...... 10 Power Management Enhancements...... 11 NVIDIA GeForce GTX 280 and GTX 260 Specifications ...... 12 GeForce GTX 200 Board Design...... 13 Features and Benefits: GeForce GTX 200 GPUs ...... 14 Reference Performance Results HERE...... 16 SLI Performance...... 18 NVIDIA HybridPower Technology ...... 20 PureVideo HD...... 20 INSTALLATION GUIDE ...... 22 CPU ...... 22 Motherboards...... 22 Power Supply ...... 22 Windows Vista Service Pack 1...... 22 Graphics Driver ...... 22 Game Benchmarking ...... 22 Appendix A ...... 23 NVIDIA CONTACT INFORMATION...... 24

2 NVIDIA GeForce GTX 200 GPUs

Two Personalities, One GPU Built with over 1.4 billion transistors, NVIDIA’s GeForce GTX 280 and 260 GPUs are the largest and most complex GPUs ever created. Featuring 240 shader processors, 80 texture processors, and 1 GB of frame buffer, the GeForce GTX 280 offers peerless performance in all of today’s hottest titles. But it is more than just the world’s fastest GPU. Thanks to NVIDIA’s CUDA technology, the GeForce GTX 280 is also an immensely powerful, many-core, parallel processor. When operating in computing mode, the GeForce GTX 280 becomes a fully programmable multiprocessor with 240 cores, on-die shared memory, random read and write capability, and 1 GB of dedicated memory. In computing mode, the GeForce GTX 280 turns an ordinary PC into a miniature supercomputer, providing almost one teraflop of horsepower for advanced visual computing, video transcoding, physics acceleration, and scientific computing.

Beyond Gaming With over 70 million CUDA-enabled GeForce 8 and 9 Series GPUs shipped to market, numerous rich-media and visual-computing applications are becoming CUDA-enabled. These applications are joining the ever-growing trend of apps that are enjoying phenomenal performance speedups that started with high-end professional and scientific applications upon the initial release of CUDA.

GPU-Powered Video Transcoding

Transcoding is one of the most time-consuming desktop applications today. Transcoding a movie from one format to another can take half a day, making the process extremely unfriendly to the end user. Faster CPUs can reduce the time by minor increments, but the final waiting time is still many hours. With CUDA technology, video transcoding time can be cut by up to 18×, reducing waiting time from hours to minutes. The RapidHD video transcoder from Elemental Technologies uses CUDA to achieve this exact feat, allowing video to be quickly converted from one format to another, making them easier to share and enjoy.

GPU Powered Folding@Home

Stanford University’s Folding@Home client allows the home PC user to perform protein simulations to help scientists better understand protein folding, misfolding, and related diseases. Now powered by CUDA, Folding@Home will tap into the power of the GPU, allowing tens of millions of GeForce users advance medical science with this distributed program. Even if 1% of NVIDIA users participated in Folding@Home, it would offer a stratospheric increase in the total computing power of the Folding@Home distributed cluster.

Industry wide support for CUDA

Countless math-intensive applications benefit from CUDA and the power of the GPU. Figure 1 shows the incredible speedups achieved for both consumer and professional applications. Appendix A lists references and details for these applications.

Figure 1. GPU Speedups Versus CPU

Gaming Beyond Today’s games have great visuals, but they do not behave with great realism. Animation is often unconvincing, particle effects do not take into account of physics, and geometry is static and unchanging. GeForce GTX 200 GPUs take gaming beyond static realism, into dynamic realism.

Dynamic Realism

While prior-generation GPUs could deliver real-time images that appeared true-to-life in many cases, frame rates could drop to unplayable levels in complex scenes with significant animation, numerous physical effects, and multiple characters. The combination of the sheer shader processing power of GeForce GTX 200 GPUs and NVIDIA’s new GPU PhysX technology facilitates many new high-end graphics effects including: Convincing facial animation Multiple ultra-high polygon characters in complex environments Advanced volumetric effects like smoke, fog, and mist Fluid and cloth simulation Fully simulated physical effects such as live debris, explosions, and fires. Physical weather effects such as accumulating snow and water, sand storms, soaking, drying, dampening, overheating, and freezing Better lighting for dramatic and spectacular effect, including ambient occlusion, global illumination, soft shadows, color bleeding, indirect lighting, and accurate reflections. In Figure 2 below, you can see a comparison of our previous “Adrianne” demo used to highlight the capabilities of the GeForce 8-series GPUs to our new Medusa demo used to demonstrate the power of the GeForce GTX 200 GPUs. Adrianne was a single, complex character existing primarily in a cubemap environment. Prior- generation GPUs only had enough horsepower to render a single complex character, like Adrianne, in a scene. With the GeForce GTX 200 GPUs, we have enough horsepower to render two characters (each more complex than Adrianne) within an incredibly complex environment all on one GPU.

Figure 2. Facial Expressions in Medusa

Numerous "blendshapes" are used in the Medusa demo. Blendshapes are the facial poses that are combined to produce facial expressions in animation. Hollywood movies can use a hundred or more of these blendshapes per character to capture all the nuances of human expression. Previous demos like Adrianne were able to maintain about 25-30 blendshapes with only five active in any individual frame. In the Medusa demo, the Medusa and warrior characters each have 130 blendshapes with over 30 active in an individual frame. Looking at Adrianne, you'll notice her expressions were a bit "plastic" looking. She had a few extreme expressions of happy, sad, puzzled, etc., but the Medusa characters show a much broader range of emotion. We can now create scenes where the characters are truly acting. We used a key DirectX 10 feature called “stream out” extensively in Medusa. When using the same model several times in the same frame, stream out allows processing the geometry for the model once, streaming it out to local video memory, and then reusing the processed geometry over and over. The characters in Medusa must be rendered several times in a single frame, including renders for shadow passes, and blur passes for skin rendering. Stream out relieves us from having to re-process the same geometry multiple times for each pass, which reduces the geometry processing workload and allows us to apply the remaining horsepower to shading processing. Medusa also fully utilizes the DirectX 10 feature called "geometry shading," where geometry can be created by the GPU. Geometry shading is used for the "stone effect" where the stone grows up over the form of the warrior. If you observe the demo in wireframe mode, you can see new triangles (geometry) being created along the leading edge of the growing stone. Thanks to the power of GeForce GTX 200 GPUs, blendshapes can be used extensively and, in conjunction with geometry shading and stream out, allow for greatly improved dynamic realism.

NVIDIA PhysX Technology Delivering physics in games is extremely compute-intensive and based on a set of physics algorithms that require tremendous amounts of simultaneous mathematical and logical calculations. NVIDIA’s PhysX™ technology provides a powerful, real-time physics engine used in numerous leading edge PC and console games. PhysX has been incorporated into over 150 games, is used

by more than 10,000 registered users, and is supported on Sony Playstation 3, Microsoft Xbox 360, Nintendo Wii, and the PC. The PhysX API is designed to be accelerated by powerful processors with numerous processing cores, such as the GeForce GTX 200 GPUs. In fact, all NVIDIA CUDA-enabled GPUs such as the 70 million already shipped GeForce 9 and GeForce 8 Series GPUs will support PhysX. Combined with powerful GPUs, PhysX provides an exponential increase in physics processing power that takes gaming to a new level, delivering rich, immersive physical gaming environments with features such as: Explosions that cause dust and collateral debris Characters with complex, jointed geometries for more life-like motion and interaction Spectacular new weapons with incredible effects Cloth that drapes and tears naturally Dense smoke and fog that billow around objects in motion See the PhysX FAQ for more details: http://www.nvidia.com/object/physx_faq.html

SLI The GeForce GTX 200 GPUs provide 50-100% more performance over prior-generation GPUs, permitting increased frame rates and higher visual quality settings at extreme resolutions, delivering a truly cinematic gaming experience. Support for the new DisplayPort interface allows resolutions beyond 2560 × 1600, and 10-bit color support permits up to a billion different colors on screen (driver, display, and application support is also required). Note that prior-generation GPUs included internal 10-bit processing, but could only output 8-bit component colors (RGB). GeForce 200 GPUs permit both 10-bit internal processing and 10-bit color output. NVIDIA’s SLI technology is the industry’s leading multi-GPU technology, giving you an easy, low- cost, high-impact performance upgrade. PC gaming simply doesn’t get any faster or more realistic than running GeForce GTX 280 boards in SLI mode on the latest nForce® motherboards. Two flavors of SLI are supported by the initial GeForce GTX 200 GPUs: Standard SLI (two GPU boards), which typically boosts supported game performance by 60-90% and permits higher quality settings 3-way SLI, which provides even higher frame rates and permits higher quality settings for the ultimate experience in PC gaming when connected to a high-end, high-resolution monitor. The bottom line is that GeForce GTX 200 GPUs process and display complex DirectX 10 and OpenGL game environments with amazing cinematic effects and high frame rates at extreme high-definition resolutions.

Figure 3. Extreme High-Definition Gaming

Introducing GeForce GTX 200 GPUs Implementing NVIDIA’s second-generation unified visual computing architecture, the new GeForce GTX 280 and GeForce GTX 260 graphics boards are members of the GeForce GTX 200 GPU family and include the powerful new GT200 GPU, providing the ultimate visual computing and extreme HD gaming experience. Numerous application areas are greatly enhanced and accelerated by the GeForce GTX 200 GPUs, including gaming, physics, rich media encoding, HD video playback, image processing, and High-Performance Computation to name a few. GeForce GTX 200 GPUs include a number of very significant architectural enhancements over the prior generation, and they continue NVIDIA’s trend of continuous improvement of architectural efficiency as exemplified by increased performance per watt and performance per square millimeter of GPU die area. With its 240 processing cores and 1GB of frame buffer memory at a suggest retail pricing of $649, a stock-clocked GeForce GTX 280 board will deliver 1.5x performance improvement on average over high-end GeForce 8 and GeForce 9 series single-GPU graphics boards. The $449-based GeForce GTX 260 targets price/performance conscious enthusiasts and includes a reduced number of processing cores (192) and lower frame buffer size (896MB), but still permits many top 3D games to be played in all their glory. Users can configure two or three GeForce GTX 200 GPU boards in SLI mode, using appropriate nForce chipset-based motherboards, delivering the highest frame rates, highest resolution, and best 3D visual quality possible today. The new GPUs incorporate advanced power and thermal management for optimal acoustics, power, and performance based on usage. Dynamic clock and voltage scaling based on workload, intelligent clock gating, and Hybrid Power support help keep the electric bill at bay, while lowering heat and noise output. The GT200 GPU is manufactured using TSMC’s 65 nm fabrication process. Built with 1.4 billion transistors, it is the largest, most powerful, and most complex GPU ever made, and operates well within power and heat margins required in enthusiast PCs.

Optimized PC and Heterogeneous Computing GPUs are the most important processors in the new era of visual computing. Powerful GPUs such as the GeForce GTX 200 GPU family are necessary to run intensive DirectX 10-based 3D games like Crysis at high quality and high resolution settings. Very capable GeForce 8 and 9 series integrated and midrange GPUs must be employed to obtain stutter-free high-definition video playback on the PC, while simultaneously displaying the Aero 3D user interface of Windows Vista. Users and system vendors are steadily realizing that the proper balance of CPU and GPU power in a PC configuration is essential for the best user experience and optimal price-to-performance ratio. We refer to such proper CPU and GPU balance as the “optimized PC”. Optimized PC configurations are based on the concept of “heterogeneous computing”, where the appropriate processor is applied to the tasks that it can process the fastest and with the highest quality. Configuring a system with a lower-cost CPU and using the savings to upgrade to a higher- end GPU, or multiple GPUs using NVIDIA SLI technology, delivers significant performance improvements in gaming, visual computing, and compute-intensive rich media applications without costing more money.

GeForce GTX 200 GPUs – Architectural Improvements

The GeForce GTX 200 GPUs implement a number of key architectural improvements. Not only do these changes improve performance by 1.5× on average over the prior generation, but also improve architectural efficiency, measured in performance/square millimeter and performance/watt. The GeForce GTX 200 GPUs significantly reduce idle power, hovering around 25 watts, which is pretty incredible for such powerful GPUs. HybridPower™ is also supported (see HybridPower section below), so the GPU may be turned off when used in nForce® chipset-based motherboards that support HybridPower and have integrated motherboard GPUs (such as nForce 780a and 790i enthusiast motherboards). The GeForce 8 and GeForce 9 Series implemented the first version of NVIDIA’s unified visual computing architecture. The new GeForce GTX 200 GPUs implement a second generation of this architecture. While the foundation of both architectures is a scalar/unified/DirectX 10 processing core, GeForce GTX 200 GPUs provide many more architectural enhancements than simply a dramatic increase in the number of functional and processing units. Here are some of the key advances found in GeForce GTX 200 GPUs versus GeForce 8 and 9 Series architecture GPUs: Support for three times the number of threads in flight at a given time New scheduler design for 20% more texturing efficiency 512-bit memory interface Improved z-cull and compression technology for better performance at high resolution Architectural enhancements for geometry shading and stream out performance Full-speed, raster-operation (ROP) frame buffer blending (vs. half speed on 8800 GTX) Improved dual issue for more efficient and higher performance computation Twice the number of registers for longer and more complex shaders IEEE 754R double precision for improved floating-point computational accuracy Hardware support for 10-bit color scan out (DisplayPort only) All of these improvements combined with the large increase in functional units, delivers the next generation of visual realism and ultra-high-resolution DirectX 10 gaming. Refer to the Tech Brief titled “NVIDIA GeForce GTX 200 GPU Architectural Overview” for more in- depth descriptions of all the new architectural enhancements.

Power Management Enhancements

GeForce GTX 200 GPUs include a more dynamic and flexible power management architecture than past generation NVIDIA GPUs. Four different performance / power modes are employed: Idle/2D power mode (approx 25 W) Blu-ray DVD playback mode (approx 35 W) Full 3D performance mode (varies—worst case TDP 236 W) HybridPower™ mode (effectively 0 W)

Using a HybridPower-capable nForce motherboard, such as those based on nForce 780a or 790i chipsets, a GeForce GTX 200 GPU can be fully powered off when not performing intensive graphics operations and graphics output can be handled by the motherboard GPU (mGPU). For 3D graphics-intensive applications, the NVIDIA driver can seamlessly switch between the power modes based on utilization of the GPU. Each of the new GeForce GTX 200 GPUs integrates utilization monitors (“digital watchdogs”) that constantly check the amount of traffic occurring inside of the GPU. Based on the level of utilization reported by these monitors, the GPU driver can dynamically set the appropriate performance mode (i.e., a defined clock and voltage level) that minimizes the power draw of the graphics card—all fully transparent to the end user. The GPU also has clock-gating circuitry, which effectively “shuts down” blocks of the GPU which are not being used at a particular time (where time is measured in milliseconds), further reducing power during periods of non-peak GPU utilization. All this enables GeForce GTX 200 graphics cards to deliver idle power that is nearly 1/10th of its maximum power (approximately 25 W on GeForce GTX 280 GPUs). This dynamic power range gives you incredible power efficiency across a full range of applications (gaming, video playback, surfing the web, etc).

NVIDIA GeForce GTX 280 and GTX 260 Specifications

GPU GTX 280 GTX 260 Fabrication Process 65 nm 65 nm Number of Transistors 1.4 Billion 1.4 Billion

Graphics Clock (Including dispatch, 602 MHz 576 MHz texture units, and ROP units) Processor Clock (Processor Cores) 1,296 MHz 1,242 MHz Processor Cores 240 192 Memory Clock (Clock rate / Data rate) 1,107 MHz / 2,214 MHz 999 MHz / 1,998 MHz Memory Interface 512 bit 448 bit Total Memory Bandwidth 141.7 GB/s 111.9 GB/s Memory Size 1 GB 896 MB ROPs 32 28 Texture Filtering Units 80 64 Texture Filtering Rate 48.2 GigaTexels/sec 36.9 GigaTexels/sec HDCP Support Yes Yes Yes (Using DVI-to-HDMI Yes (Using DVI-to-HDMI HDMI Support adaptor) adaptor) 2 x Dual-Link DVI-I 2 x Dual-Link DVI-I Connectors 1 x 7-pin HDTV Out 1 x 7-pin HDTV Out

RAMDACs 400 MHz 400 MHz Bus Technology PCI Express 2.0 PCI Express 2.0 Form Factor Dual Slot Dual Slot Power Connectors 1 x 8 pin and 1 x 6-pin 2 x 6-pin Max Board Power 236 watts 182 watts GPU Thermal Threshold1 105° C 105° C 1) The GPU is designed to operate safely up to this temperature. If for any reason the GPU exceeds this temperature, the clock speed will automatically dial down

GeForce GTX 200 Board Design

The GeForce GTX 200 GPUs are fully encased in a high-tech, sculpted, dual-slot metal enclosure and they implement dynamically adjustable and quiet fan technology. Board temperatures typically do not exceed 80°C in an ambient room environment with proper case cooling, allowing the GPUs to perform at full speed during the most intense 3D gaming efforts, while also keeping board noise to a minimum.

Figure 4. GeForce GTX 280

Figure 5. GeForce GTX 260

The GeForce GTX 200 GPUs feature two dual-link, HDCP-enabled DVI-I outputs for connection to analog and digital PC monitors and HDTVs, a 7-pin analog video-out port that supports S-Video directly, plus composite and component (YPrPb) outputs via an optional dongle.

Figure 6. Rear bracket for GeForce GTX 200 GPUs HDCP over dual-link allows video enthusiasts to enjoy high-definition movies on extreme high- resolution panels such as the 30” Dell 3007WFP at 2560 × 1600 with no black borders. GeForce GTX 200 GPUs also provide native support for HDMI output, using a DVI-to-HDMI adaptor in conjunction with the built-in SPDIF audio connector. The GTX 200 also supports the DisplayPort interface and is an option provided by certain vendors.

Features and Benefits: GeForce GTX 200 GPUs

Second Generation NVIDIA Unified Architecture:i Second-generation architecture delivers 50% more gaming performance over the first generation through 240 enhanced processing cores that provide incredible shading horsepower. NVIDIA PhysX-Ready:ii GeForce GPU support for NVIDIA PhysX technology enables a totally new class of physical gaming interaction for a more dynamic and realistic experience with GeForce. NVIDIA SLI® and 3-way SLI Technology:iii Industry-leading, 3-way SLI technology offers amazing performance scaling by implementing 3-way alternate frame rendering (AFR) for the world’s fastest gaming solution under Windows Vista. Microsoft DirectX 10 Support: DirectX 10 with full Shader Model 4.0 support delivers unparalleled levels of graphics realism and film-quality effects for today’s hottest games. NVIDIA CUDA™ Technology:iv CUDA technology unlocks the power of the GPU’s processing cores to accelerate the most demanding system tasks—such as video transcoding—delivering up to 18× the performance of traditional CPUs. PCI Express 2.0 Support: Designed for the PCI Express 2.0 bus architecture offering the highest data transfer speeds for the most bandwidth-hungry games and 3D applications, while maintaining backwards compatibility with existing PCI Express motherboards for the broadest support. GigaThread™ Technology: Massively multi-threaded architecture supports thousands of independent, simultaneous threads, providing extreme processing efficiency in advanced, next- generation shader programs. NVIDIA Lumenex™ Engine: Delivers stunning image quality and floating-point accuracy at ultra- fast frame rates. 16× Antialiasing Technology: Lightning fast, high-quality antialiasing at up to 16× sample rates obliterates jagged edges.

128-bit Floating Point High Dynamic-Range (HDR) Lighting: Twice the precision of prior generations for incredibly realistic lighting effects—now with support for anti-aliasing. OpenGL 2.1 Optimization and Support: Provides top-notch compatibility and performance for OpenGL applications. Dual Dual-link DVI Support: Able to drive the industry’s largest and highest resolution flat-panel displays up to 2560 x 1600 and with support for High-bandwidth Digital Content Protection (HDCP).v NVIDIA PureVideo® HD Technology:vi The combination of high-definition video decode acceleration and post-processing that delivers unprecedented picture clarity, smooth video, accurate color, and precise image scaling for movies and video. Discrete, Programmable Video Processor: NVIDIA PureVideo is a discrete programmable processing core in NVIDIA GPUs that provides superb picture quality and ultra-smooth movies with 100% offload of H.264 video decoding from the CPU and significantly reduced power consumption. Dual-Stream Hardware Acceleration: Supports picture-in-picture content for the ultimate interactive Blu-ray movie experience. Dynamic Contrast Enhancement & Color Stretch: Dynamically provides post-processing and optimization of high definition movies for spectacular picture clarity. NVIDIA HybridPower Technology:vii Lets you switch from the GeForce GTX 280/260 graphics card to the motherboard GeForce GPU when running non graphically-intensive applications for a quiet, low-power, PC experience. i. The number of processing cores varies by model. GeForce GTX 280 has 240 processing cores. GeForce GTX 260 has 192 processing cores. ii. GeForce GTX 280/260 GPUs ship with hardware support for NVIDIA PhysX technology. NVIDIA PhysX drivers are required to experience in-game GPU PhysX acceleration. Refer to www.nvidia.com/PhysX for more information. iii. NVIDIA SLI certified versions of GeForce PCI Express GPUs only. A GeForce GTX 280 GPU must be paired with another GeForce GTX 280 GPU (the graphics card manufacturer can be different). SLI requires sufficient system cooling and a compatible power supply. Visit www.slizone.com for more information and a listing of SLI-Certified components. iv. Requires application support for CUDA technology. v. Playback of HDCP protected content requires other HDCP-compatible components vi. Feature requires supported video software. Features may vary by product. vii. Requires an NVIDIA HybridPower-enabled motherboard.

Note: DirectX 10.1 Not Supported DirectX 10.1 is not supported in GeForce GTX 200 GPUs. DirectX 10.1 includes incremental feature additions beyond DirectX 10, some of which GeForce 8/9/200 GPUs already support (multisample readback for example). We considered DirectX 10.1 support during the initial GPU design phase and consulted with key software development partners. Feedback indicated DirectX 10.1 was not important, so we chose to focus on delivering better performance and architectural efficiency.

Reference Performance Results

SLI Performance Results Configuring two GeForce GTX 280 GPUs in an enthusiast rig permits games to run at high resolutions with high quality, and high frame rates.

NVIDIA HybridPower Technology

When coupled with a HybridPower-enabled motherboard, the GeForce GTX 200 GPUs can be completely powered down. For everyday computing and watching HD movies, the system uses the motherboard GPU and the user can turn off the GPU for a significant reduction in overall power use (the discrete GTX 200 GPUs consume no power at all). When you run a 3D-intensive application you can turn on the GeForce GTX 200 GPUs for maximum performance. HybridPower works by sending the output of the discrete GPU through the output connector on the motherboard. This allows the system to use both GPUs as it sees fit without physically changing the connector.

Figure7. HybridPower

PureVideo HD GeForce GTX 200 GPUs support the full list of second-generation PureVideo HD features including: Hardware decode acceleration for H.264, VC-1 and MPEG-2 Image post processing Dynamic contrast enhancement Blue, green, and skin tone enhancements HD dual-stream decode Windows Vista Aero with Blu-ray movie playback Full score for HQV and HQV HD

Installation Guide

CPU

For a single GeForce GTX 280 or GTX 260, we recommend a of a Core 2 Duo E6300 CPU. For SLI and 3-way SLI, we recommend a Core 2 Duo E8500 CPU.

Motherboards

The GeForce GTX 200 GPUs should work with any motherboard that supports PCI Express. For SLI, an NVIDIA nForce SLI motherboard is required.

Power Supply

For a single GeForce GTX 280 GPU, a 550-watt power supply unit with 40 A on the 12 V rail is required, at minimum. For a single GeForce GTX 260 GPU, a 500-watt power supply unit with 36 A on the 12 V rail is required, at minimum. The GeForce GTX 280 requires one 8-pin and one 6-pin PCI-E power connector. The GeForce GTX 260 requires two 6-pin PCI-E power connectors. A high wattage rating does not guarantee stable power. For SLI, we recommend SLI certified power supplies. Please tell end users to visit SLI Zone (www.slizone.com) for a list of SLI certified power supplies for the GeForce GTX 200 GPUs.

Windows Vista Service Pack 1

We recommend that you always stay up to date with the latest Microsoft Windows updates. Be sure to install Service Pack 1 prior to testing in Windows Vista.

Graphics Driver

If you have any earlier drivers installed, please uninstall them first. This removes the chance of any version conflicts. Install the latest recommended driver from the NVIDIA extranet, FTP, or web site.

Game Benchmarking

For detailed procedures on how to test various games, please see the NVIDIA Benchmarking Guide. It can be obtained from the NVIDIA PR extranet.

Appendix A

Figure 1 References 1. “Interactive Visualization of Volumetric White Matter Connectivity in DT-MRI Using a Parallel- Hardware Hamilton-Jacobi Solver,” by Won-Ki Jeong, P. Thomas Fletcher, Ran Tao, and Ross T. Whitaker 2. “GPU Acceleration of Molecular Modeling Applications.” 3. Video encoding uses iTunes on the CPU and Elemental on the GPU running under Windows XP. CPUs tested were Intel Core 2 Duo 1.66 GHz and Intel Core 2 Quad Extreme 3 GHz. GPUs tested were GeForce 8800M on the Gateway P-Series FX notebook and GeForce 8800 GTS 512 MB. CPUs and GeForce 8800 GTS 512 were run on Asus P5K-V motherboard (Intel G33 based) with 2 GB DDR2 system memory. Based on an extrapolation of 1 min 50 sec 1280 × 720 high-definition movie clip. 4. http://developer.nvidia.com/object/matlab_cuda.html 5. “High performance direct gravitational N-body simulations on graphics processing units paper,” communicated by E.P.J. van den Heuvel 6. “LIBOR,” by Mike Giles and Su Xiaoke. 7. “FLAG@lab: An M-script API for Linear Algebra Operations on Graphics Processors.” 8. http://www.techniscanmedicalsystems.com/ 9. “General Purpose Molecular Dynamics Simulations Fully Implemented on Graphics Processing Units,” by Joshua A. Anderson, Chris D. Lorenz, and A. Travesset 10. “Fast Exact String Matching On the GPU,” presentation by Michael C. Schatz and Cole Trapnell

NVIDIA Contact Information

Please don’t hesitate to contact us if you have any questions. For PR inquiries, please contact your region’s local NVIDIA representative. For technical inquires, please contact James Wang or Nick Stam.

Contact

Technical James Wang Marketing Technical Marketing Analyst, GPU Product Reviews Office: 408-566-7074 Cell: 408-480-4827 [email protected]

Nick Stam Lars Weinand Director, Technical Marketing Senior Technical Marketing Office: 215-504-0321 Manager, Europe Cell: 215-514-0400 Tel. +49 89 6283 50013 [email protected] Fax +49 89 6283 50001 [email protected]

PR USA Derek Perez Ken Brown Director of PR PR Manager 408–486-2512 510-290-2603 [email protected] [email protected] 2701 San Tomas Expressway, Santa 2701 San Tomas Expressway, Clara CA 95050 USA Santa Clara CA 95050 USA

Bryan "BDR" Del Rizzo Senior PR Manager 2701 San Tomas Expressway Santa Clara, CA 95050 408-486-2772 [email protected]

PR Europe Luciano Alibrandi Jens Neuschaefer Director of Product PR EMEAI Central European Product PR +33 6 07405498 Manager [email protected] [email protected] +49 173 528 2912 NVIDIA Ltd 14, place Marie Jeanne Bassot NVIDIA GmbH 92593 Levallois Perret, FRANCE Rosenheimerstr.145b 81671 München, GERMANY

Adam Foat Stephane Quentin Northern European Product PR France and Belgium Manager [email protected] Product PR Manager +44 7855431501 +33 6 14308655 [email protected] NVIDIA Ltd NVIDIA Ltd 1310 Arlington Business Park 14, place Marie Jeanne Bassot Theale, Berkshire RG7 4SA United Kingdom 92593 Levallois Perret, FRANCE

Igor Stanek Product PR Manager Central Eastern Europe [email protected] +42 0602 135136 NVIDIA Ltd 1310 Arlington Business Park Theale, Berkshire RG7 4SA United Kingdom

PR Asia Sunny Lee Varun Dubey Korea Marketing Manager Product PR Manager, India [email protected] [email protected] +82 (2) 6000-8012 +91 9960 611 026 NVIDIA Korea NVIDIA Ltd #2101, COEX Trade Tower, 159-1 Level 4, Dynasty Samsung-dong Business Centre Kangnam-gu, Seoul 135-729 KOREA Andheri Kurla Road Andheri East Mumbai 400059 India

Peggy Chang Kaori Nakamura Marcom/Event Specialist, TASEA PR Manager, Japan [email protected] [email protected] +886 (2) 2175-5786 +81-3-6743-8712 NVIDIA Taiwan NVIDIA Japan Akasaka Tameike Tower 2F, 3F, #260, Tun Hwa N. Road, 2-17-7, Akasaka, Minatok-ku, Taipei 105, Taiwan R.O.C. Tokyo 107-0052

Peizhi (Perry) Deng Technical Marketing Manager, Asia Pacific

[email protected] +86-10 5866 1518 NVIDIA Corp. Unit 2901-2904, China World Tower 1 N0.1 Jian Guo Men Wai Avenue Beijing, P. R. China

Notice ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. Information furnished is believed to be accurate and reliable. However, NVIDIA Corporation assumes no responsibility for the consequences of use of such information or for any infringement of patents or other rights of third parties that may result from its use. No license is granted by implication or otherwise under any patent or patent rights of NVIDIA Corporation. Specifications mentioned in this publication are subject to change without notice. This publication supersedes and replaces all information previously supplied. NVIDIA Corporation products are not authorized for use as critical components in life support devices or systems without express written approval of NVIDIA Corporation. Trademarks NVIDIA, the NVIDIA logo, GeForce, nForce, PureVideo, SLI, HybridPower, CUDA, GigaThread, and Lumenex are trademarks or registered trademarks of NVIDIA Corporation. Other company and product names may be trademarks of the respective companies with which they are associated. Copyright © 2008 NVIDIA Corporation. All rights reserved.