Announcing NVIDIA Maximus With complex add-on programs and long wait times it used to be that photorealistic rendering was reserved just for the styling and marketing departments. No longer. Realistic models are fast becoming a necessity for designers and engineers for more accurate and faster decisions throughout the entire process. Now, with rendering integrated directly within CATIA – powered by NVIDIA® Maximus™ technology – Live Rendering provides an intuitive and interactive means for creating images that rival photographs, up to 6x faster3. NVIDIA Maximus-powered workstations are the first to deliver a unified platform that combines the industry-leading professional 3D graphics capability of NVIDIA® Quadro® GPUs and the high performance computing power of NVIDIA® Tesla™ GPUs. The Tesla co-processors automatically take the heavy lifting of rendering, freeing the Quadro GPUs to do what they do best—enabling rich interactive graphics. For more information, visit our Maximus workstation page. Live Rendering Benchmark Results You can see how fast Live Rendering can be with the power of NVIDIA Maximus technology. No longer do you need to wait forever for beautiful, print-quality images. As you add Quadro and Tesla GPUs, performance gets exponentially faster. So you actually get more than you pay for. Which Quadro Solution is Right for me?   Quadro 4000 Maximus : Quadro 4000 + Tesla C2075 Maximus : Quadro 6000 + Tesla C2075 Users CAD Modelers Creative Designers Visualization Experts CATIA Live Rendering Usage Occasional Regular Intensive Benefit Create photorealistic images to quickly and visually communicate project progress and direction Interactively study model shape and material variations to optimize creative design workflow Review and change materials interactively to fine tune material library and produce high resolutions images CUDA Processing Cores 256 704 (256 + 448) 896 (448 per GPU) Total GPU Memory 2 GB GDDR5 8 GB GDDR5 (2GB + 6GB) 12 GB GDDR5 (6 GB/GPU) Memory Interface 256 bit 384-bit + 256-bit 384-bit per GPU Memory Bandwidth 90 GB/s 144 GB/s + 90 GB/s 144 GB/s per GPU Max Power Consumption 142 W 306 W 408 W Graphics Bus PCI Express 2.0 x16 PCI Express 2.0 x16 PCI Express 2.0 x16 Display Connectors³ 2 DisplayPort 1 Dual Link DVI-I 1 Stereo 3D 4 DisplayPort (2 per GPU) 2 Dual Link DVI-I (1 per GPU) 2 Stereo 3D (1 per GPU) 4 DisplayPort (2 per GPU) 2 Dual Link DVI-I (1 per GPU) 2 Stereo 3D (1 per GPU) THE PROVEN SOLUTION FOR CATIA For CATIA Live Rendering certified configurations and drivers visit Dassault Systemes certified system hardware page at http://www.3ds.com/support/certified-hardware/overview/ CATIA Live Rendering is available on all versions of CATIA V6 2011X and above.  For best experience, NVIDIA recommends running CATIA Live Rendering on a CATIA certified platform equipped with the latest generation Quadro and release 259.70 driver or above.  Quadro Plex 7000 is only supported on Driver release 270 and above.

CERTIFIED AND RECOMMENED GPUs Quadro Plex 7000 > 12 GB > 1024 CUDA Cores Quadro 6000 > 6 GB¹ > 448 CUDA Cores Quadro 5000 > 2.5 GB > 352 CUDA Cores Tesla C2075 > 6 GB > 448 CUDA Cores > To get the latest CATIA Performance    drivers for Quadro, click here. > To view Dassault certified hardware click    here. CATIA Live Rendering Optimization Guide Download the Optimization Guide to see the benefits of Live Rendering for: Styling and Marketing Industrial design Color and trim experts Master surfacers Design Review Engineering 3D modelers Perceived quality engineers Ergonomic engineers Light engineers Packaging engineers

¹ 6GB is supported on Win7 and Linux64 (4GB memory limit on Windows XP64) via Rel 256 driver.
² Two out of any three connectors can be active at a time
³The comparison presented for Catia Live Rendering is based on industry standard workstation: HP z800 with dual quad core Xeon W5580 CPU @3.2GHz and appropriate CPU or GPU   combination, 12GB RAM, Win7-64bit OS. 275.89 Dassault Certified Driver was used in the benchmark running Dassault Catia V6R2012.HF6 (6.211.6.0) at 1920x1200 resolution. The   comparison shows average frames rendered per second (on a total of 100 frames). CPU-only result is based on 8 CPU cores, and CPU+GPU results are based on 6 CPU cores.

© 2012 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, NVIDIA Quadro, and CUDA, are trademarks and/or registered trademarks of NVIDIA Corporation. All company and product names are trademarks or registered trademarks of the respective owners with which they are associated. Features, pricing, availability, and specifications are all subject to change without notice.