Archive for the ‘Genkii’ tag
secoNdlife problem
If you read our last post on the N-body problem and want to try your hand at playing around with simulations of this enduring problem, then this article from Wired Science may be music to your N=2 ears. A group of developers from virtual world developer Genkii and astrophysicists from the Meta Institute of Computational Astrophysics have put together a simulation tool using the open source OpenSim (an implementation of Linden Lab’s popular SecondLife virtual world engine) to run N-body problem simulations and aim to publish their results in the Journal of Virtual Worlds Research.
As to why the developers chose to use OpenSim (from their ArXiV pre-print):
“From the point of view of an astrophysicist dealing with gravitational N-body simulations, virtual worlds such as OpenSim are N-body simulators, with two extra features: a surprisingly elaborate graphics module, and a bug in the equations of motion. As to the latter: whereas objects should attract each other via Newton’s inverse-square law of gravity, objects in OpenSim fall straight down. However, that “bug” is easily fixed. We have done so, and we discuss our first results in this paper.”
The most interesting aspect of this work is not only that it was possible, but that, should virtual world administrators choose to allow it, these tools can be bundled into almost any virtual world running off an OpenSim compatible setup!
We have modified the standard physics engine of OpenSim using a plugin. Server administrators can select to replace the standard physics engine with our plugin at server-initialization time, region by region … Though [the example above] shows only one avatar in view on a remote “desert island,” a similar simulation could, in principle, take place anywhere on an OpenSim grid, and any user present could collaborate to construct the initial conditions, discuss the outcome with other avatars, save data from the simulation, etc.
Now, the current simulation has a limit: it can only simulate up to 50 bodies – but I’d like to think of this as just one powerful example of how virtual world technology might be used in the future to power new types of simulations and empower scientists to collaborate over them.