• Subscribe

Image of the week - Airbending with GPUs

Image of the Week - Airbending with GPUs


Zade Rosenthal/Industrial Light & Magic Nicola Peltz plays Katara in the Paramount Pictures/Nickelodeon Movies adventure, "The Last Airbender." Copyright (C) 2010 Paramount Pictures Corporation. All Rights Reserved.

What do whirling clouds, smoke, and ocean mist have in common with fire? They were all simulated using Plume, Industrial Light & Magic's proprietary fluid simulation system.

The artistic process is marked by repeated review and revision. But rendering graphics takes time, and if you have to wait hours to review and revise your work, it's hard to achieve artistic excellence.

To speed things up, ILM sped their computations up with GPUs. The NVIDIA press release states:

Central to achieving many of these sequences was the use of a tool created by ILM called Plume. Plume is both a fluid simulation system and a GPU-based renderer that utilizes the NVIDIA® CUDA™ parallel computing architecture. ILM realized dramatic speed increases with Plume by running simulation renders on a 12-machine GPU-based render farm powered by NVIDIA Quadro® FX 5800 professional graphics solutions.

"As with everything in high-end visual effects, iteration was essential," said Olivier Maury, research and development engineer, ILM. "By working within an NVIDIA GPU-based framework, we saw up to eight iterations each day of complex fire, dust and air simulations. That represents speed improvements of 10-15x over CPU-based simulations. Access to CUDA and NVIDIA GPUs has entirely changed the way we approach a variety of complex visual effects challenges."

Read more and see more pictures at the original press release.

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2019 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.

Republish

We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.