CoP 3: Case Study 1 'Battleship' 2012

I've decided to do two different case studies for my dissertation, the first one being about the water simulations produced by ILM for the action film 'Battleship' that came out in 2012. I chose to do a case study on this because of how complex and large the simulations were, and how ILM were going to face the challenges it draught.

For the film ILM formed an internal group called the "Battleship Water Project" that focused on the water simulations and helped improve the pipeline of the software being used for the film. Because the film had to produce large scale simulations the team had to come up with a new way of creating the simulations without destabilizing and loosing detail in the water simulation. Grady Cofer said that “even at a grid size that might resolve to real word scale of two-foot square, you are losing a lot of detail, there is so much fine detail in these complex water structures that you just don’t get from the based simulation.” (Cofer. G, 2009)

How they fixed this was that they placed on top a FLIP PIC solver for particle based simulations. This granted the GRID approach to simulate the extensive oceans but also granting, in the same scenes, over twenty million particles to be thrown up alongside the ships. Because the film used a lot of water sequences it was vital for ‘Battleship’ that the water breaks up into batches and then droplets and eventually scatters in to mist. Through the waters lifespan it gradually gets more influenced by wind currents and airfields. For ILM this was a new type of physics and rendering pipeline of water to figure whilst using new toolsets created by the CG supervisor Willi Geiger (Desowitz. B, 2012). Grady Cofer explained how they created this effect and how they managed to render this out for the film saying,
“we wanted to render everything together as a volume so that when we’re representing sunlight, say, that the sun is actually refracting through dense parts of water and then breaks up and illuminates in a scattered way.” (Cofer. G, 2012).

Not only did ILM create a new way in simulation large amounts of water with loads of detail but they also went that bit further and created a way the air was affected. Affecting the air in turn altered how the water comes off the waves and splashes as mist, and how the mist acted, and how the actions aerify the water and produced white water. This makes the simulation an amazing step up in realism and in an interview with the VFX supervisor Grady Cofer he says that “once we found the key to the air fields around out splashes, it really injected a lot of realistic.” (Cofer. G, 2009). ILM had another trick up their sleeve by developing a system called ‘ballistic particles’, that generates particles, which animators can implant in to the simulation, to erupt scientifically and add extra commotion to the simulation. This method of simulation brings a certain realism to the scene that adds haphazardness and unreliability in the water, it also creates character to the water due to the energy the water is simulating.

Relating this back to my practical, it taught me that I could use other forces to manipulate the particles to give the simulation some sort of character element. Adding an accelerator to a simulation or a force field will give the particles multiple directions in with they could move about, this is what I will try out in my next simulation.