A massively-parallel computing device made from supersaturated solutions of sodium acetate? The basic idea is to use the wavefront of crystallization to perform calculations; the speed of the wavefront through the solution and the way it interacts with other wavefronts performs calculations.
Most experimental prototypes of unconventional computers either require a tailored hardware interface (analog computers, liquid crystals) and specialized equipment (memristors, gas-discharge systems), or they may have intrinsic limitations on the speed of computation (reaction-diffusion chemical processors). Physarum computer is the simplest to build but the most difficult to control, due to the sensitivity and somewhat unpredictable behaviour of the living creature.
We aim, therefore, to provide an example of a novel computing material which is cheap to build, requires minimal resources to operate, implements computational procedures relatively quickly and is capable of solving a wide range of computationally-hard tasks. We show that sodium acetate trihydrate (colloquially called ‘hot ice’ due to its resemblance to ice and its crystalline behaviour) perfectly fits our specification of an ‘ideal DIY unconventional computer’ because it solves a variety of tasks by traveling and interacting waves of crystallization in its supersaturated solution.
(Hot Ice computer video)
"A supersaturated solution of sodium acetate, commonly called 'hot ice', is a massively-parallel unconventional computer. In the hot ice computer data are represented by a spatial configuration of crystallization induction sites and physical obstacles immersed in the experimental container. Computation is implemented by propagation and interaction of growing crystals initiated at the data-sites. We discuss experimental prototypes of hot ice processors which compute planar Voronoi diagram, shortest collision-free paths and implement AND and OR logical gates.
In the video, sites where crystallization was induced by the pins are encircled. The planar Voronoi diagram of yields the same result as obtained by classical algorithm."
When I read about this, I thought about how it might be possible to create a very large crystalline computer system - and it would make a cool clubhouse as well.
(Superman's Fortress of Solitude)
Update: Roger Zelazny fans might also recall Speicus, the crystallized protein artificial intelligence, from his
remarkable 1976 book Doorways in the Sand. End update.