The NVIDIA GeForce 9800 GX2, one of the 9th generation of the GeForce series of graphics processing units, is one of the fastest graphics cards available. The double-wide card measures 11", spanning from the built-in fan on one side to the two dual-link DVI ports (and one HDMI port) on the back. Its dual-GPU engine has 256 stream processors and a 1 GB frame buffer, enabling it to perform 30 to 50% faster than the GeForce 8800 Ultra GPU.

Key features of the NVIDIA GeForce 9800 GX2 include:

• 600 MHz core clock rate.
• 1 GB memory (512 MB for each GPU); 1000 MHz memory clock rate
• 128 GB/sec memory bandwidth.
• 1500 MHz shader clock rate.
• Texture fill rate 76.8 billion/sec.
• PCI Express 2.0 support, offering the highest data transfer speeds while maintaining backwards compatibility with existing PCI Express motherboards.
• NVIDIA PureVideo HD technology, which provides the combination of high-definition video decode acceleration and post-processing to deliver unprecedented picture clarity, smooth video, accurate color, and precise image scaling.
• HDMI Output Integrated HDMI connector enables sending both high-definition video and audio signals to an HDTV via a single cable.
• NVIDIA HybridPower technology enables you to switch from the GeForce 9800 GX2 graphics card to the motherboard GeForce GPU when running non-graphically intensive applications for optimal power savings.
• Quad NVIDIA SLI technology offers performance scaling by implementing 4-way AFR (alternate frame rendering) under Windows Vista with solid, state-of-the-art drivers.
• Supports DirectX 10, Shader Model 4.
• Delivers 2560 x 1600-pixel video with support for High-Definition Content Protection (HDCP) over both the card's DVI and HDMI outputs.
• 16X antialiasing.
• 128-bit floating point high dynamic-range lighting.
• Outputs include two DVI ports, an HDMI output, and S/PDIF in connector on-board for routing audio through the HDMI cable.
  

The graphics technology of the RADEON® HD 2900XT, from AMD ATI, is designed with DirectX 9 and 10 and Windows Vista Shader Model 4.0 support to deliver maximum graphical detail at the best possible frame rates. The HD 2900XT establishes a new level of performance with a next-generation unified shader engine with 320 unified stream processors, CrossFire, game physics, and built-in HDMI with multi-channel 5.1 surround audio.

The RADEON HD 2900XT comes with a 512-bit 8-channel GDDR3/4 memory interface 743MHz core clock and 2GHz memory.

AMD introduces 320 stream processors with this card, which can handle vector or pixel processing. The card also introduces a new type of anti-aliasing called Custom Filter Anti-Aliasing (CFAA) which offers several modes providing pixel sampling outside multi-sample box filtering in the form of wide and narrow tent filters; some of the new modes also feature an adaptive edge detection filter.

The key features of ATI RADEON HD 2900XT include:

• 743 MHz core clock rate.
• 512-bit 8-channel GDDR3/4 memory interface.
• HDMI with 5.1 surround sound audio; supports all display resolutions up to 1920x1080.
• Unified shader engine with 320 unified stream processors.
• Full support for Microsoft DirectX 10 including Shader Model 4.0.
• Custom Filter Anti-Aliasing (CFAA) up to 24x for improved quality; all anti-aliasing features compatible with HDR rendering.
• Memory clock rate: 825 MHz.
• Built-in HDMI with multi-channel 5.1 surround audio.
• 700 million transistors on 80nm HS fabrication process.
• 2x/4x/8x/16x high quality adaptive anisotropic filtering modes (up to 128 taps per pixel).
• Crossfire multi-GPU technology; scales up rendering performance and image quality with 2 or more GPUs. CrossFire is an open platform that supports multiple components and graphics cards that can be mixed and matched in a single system.
• Two integrated dual-link DVI display outputs.
• PCI Express x16 bus interface.
• ATI Avivo HD video and display platform for delivering high-definition image quality with stunning color and realism.
  

The NVIDIA GeForce GXT 280 is the new flagship of NVIDIA's GPU product line, taking over from last year's GeForce 9800 GTX. As the fastest single GPU card currently on the market, its performance compares to a GeForce 9800 GX2. The GeForce GXT 280 features 240 stream processors, a 512-bit memory interface and 1 GB of GDDR3 memory. With improved performance for DirectX 10 features such as geometry shading and stream out, the GXT 280 provides significantly enhanced computation ability for high-performance CUDA applications and GPU physics.

The key features of the NVIDIA GeForce GXT 280 include:

• 602 MHz core clock rate.
• 1 GB GDDR3 memory; 1107 MHz memory clock rate.
• Memory bandwidth 141.7 GB/sec. over a 512-bit memory interface.
• 1.4 billion transistors.
• PCI Express 2.0.
• Support for Microsoft DirectX 10.0.
• NVIDIA PureVideo HD technology, which provides the combination of high-definition video decode acceleration and post-processing to deliver unprecedented picture clarity, smooth video, accurate color, and precise image scaling.
• NVIDIA HybridPower technology.
• Texture fill rate 48.2 billion/sec.
• NVIDIA SLI technology, which enables dynamically scalable graphics performance, enhanced image quality, and expanded display real-estate.
• Maximum digital resolution 2560 x 1600.
• Maximum VGA resolution 2048 x 1536.
• Display outputs include 2 dual-link DVI and 1 HDTV-out.
• Display support for 2 monitors.
• Support for 10-bit color depth processing and output.
• NVIDIA PhysX support for GPU-accelerated physics effects in games.
• Delivers improved power management capability, including a substantial reduction in idle power.
  

The NVIDIA GeForce 9600 GT OC offers a powerfully immersive experience designed for high-definition gaming and video playback, with resolutions up to 2560x1600, at a lower cost.

With up to a 90% performance increase over the GeForce 8600 GTS and same size footprint as the GeForce 8800 GT, the GeForce 9600 GT OC supports DirectX 10, Shader Model 4, and Vista to deliver unparalleled levels of graphics realism and film-quality effects.

The key features of the NVIDIA GeForce 9600 GT OC include:

• 675 MHz core clock rate.
• 512 MB GDDR3 memory; memory clock rate 1.8GHz on a 256-bit bus.
• Memory data rate: 1800 MHz; memory bandwidth: 57.6 GB/sec.
• Shader clock: 1700MHz.
• PCI Express 2.0.
• Microsoft DirectX 10.
• NVIDIA PureVideo HD technology, which provides the combination of high-definition video decode acceleration and post-processing to deliver unprecedented picture clarity, smooth video, accurate color, and precise image scaling.
• Stream processors: 64
• Texture fill rate 21.6 billion/sec.
• NVIDIA SLI technology, which delivers up to 2 times the performance of a single graphics card configuration for unequaled gaming experiences by allowing two graphics cards to run in parallel.
• Maximum digital resolution 2560 x 1600.
• Display outputs: 2 dual-link DVI-I, HDTV + TV out.
• Premium user experience in Microsoft Windows Vista with dedicated graphics processor.

The NVIDIA Quadro FX5600 is the ideal solution for applications requiring challenging levels of immersive realism or interactive visualization by offering collaborative capabilities for advanced multi-system and multi-device visualization environments. Featuring frame lock and genlock support (via external hardware) the NVIDIA Quadro FX 5600 is built on a new NVIDIA unified architecture that dynamically allocates compute, geometry, and shader processing power to efficiently deliver optimized performance.

With the industry's first 1.5GB frame buffer with memory bandwidth up to 76.8GB/sec., the Quadro FX 5600 enables interactive visualization of the largest, 64-bit datasets. As the reference standard for Shader Model 4.0, the Quadro FX 5600 enables next generation ultra-realistic, real-time visualization applications with unprecedented image quality. With two dual-link DVI connectors, Quadro FX 5600 by offers the industry’s best image quality at resolutions up to 3840 x 2400.

The key features of NVIDIA Quadro FX5600 include:

• Core GPU clock 600 MHz.
• Memory clock 800 MHz.
• 1.5 GB GDDR3 -- industry’s first 1.5GB frame buffer; Industry's first 1.5GB frame buffer and massive memory bandwidth up to 76.8GB/sec. delivers high throughput for interactive visualization of large models and high-performance for real time processing of large textures and frames and enables the highest quality and resolution full-scene antialiasing (FSAA).
• Memory bandwidth up to 76.8 GB/sec; 384-bit memory interface.
• Dual slot with 1 open PCI Express x16 slot.
• Full support for Microsoft DirectX 10.0 features such as Shader Model 4.0; reference standard for shader model 4.0 enabling a higher level of performance and ultra-realistic effects for next generation DirectX 10 industry-leading professional applications.
• Genlock and framelock support via add-on NVIDIA Quadro G-Sync card which enables multiple systems to synchronize frame rendering across multiple displays, plus synchronize one or many systems to a common sync source.
• SLI technology; NVIDIA SLI technology enables dynamically scalable graphics performance, enhanced image quality, and expanded display real-estate.
• Maximum resolution over VGA port: 2048 × 1536 × 32 bpp at 85 Hz
• C programming environment for the GPU; the NVIDIA CUDA GPU computing software provides a C language environment and tool suite that unleashes new capabilities to solve complex, visualization challenges such as real-time ray tracing and interactive volume rendering.
• Fast 3D textures, jumbo 8K textures processing; Fast transfer and manipulation of 3D textures resulting in more interactive visualization of large volumetric dataset. Faster processing of very large textures resulting in higher performances when zooming and panning through high resolution images.