One of the trickiest parts of any computer simulation is making water look realistic. So imagine the programmer’s task when it comes to simulating entire oceans.
Today, Emmanuelle Darles at the University of Limoges in France and a couple of friends survey the state of the art in ocean simulation and reveal the various tricks used to make the results look realistic. They also identify a number of outstanding challenges in this area.
The basic challenge in ocean simulation is to reproduce the shape of the water’s surface as well as the way it reflects light. There are essentially three approaches, say Darle and co.
The first is to measure the behaviour of the surface of real water using buoys or satellite measurements, develop a mathematical model that describes this behaviour and then use this to bound the behaviour of water “particles” in a computer model.
The second uses wave equations to compute the size and distribution of waves on a flexible surface.
The third is known as computational fluid dynamics and computes the behaviour of the entire body of fluid using the Navier-Stokes equations.
The first and second methods work best in the deep ocean where the surface has a clean oscillatory motion. But closer to the shore, the waves begin to interact with the ocean bottom, causing them to break. That kind of behaviour is much harder to reproduce and needs to be calculated from scratch using computational fluid dynamics. Even then, the calculations are hard so various approximations are needed.
And creating a seamless join between the deep and shallow water simulations is also tricky
Then there are the ocean’s complex optical properties, which are complicated by factors such as spray, foam and the presence of sand or organisms in the water.
That can be hard to reproduce. And Darle and co point to various gaps in the simulator’s tool box. “To or knowledge, there is no work dealing with realistic foam created by breaking waves,” they say.
All this makes oceans hard to simulate well.
Nevertheless, there are impressive simulations around, the best of which tend to exploit a variety of techniques to achieve their goals.
To show off some of these, here’s a gallery of images from the paper in order of the date they were generated. Clearly significant progress has been made since the first image above, which dates from 1986.
To find out more about the techniques used to create them, go to the paper:
A Survey Of Ocean Simulation And Rendering Techniques In Computer Graphics: arxiv.org/abs/1109.6494
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
How to fix the internet
If we want online discourse to improve, we need to move beyond the big platforms.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.