README.TXT
6883b645faa3efficient_neuronet

README.TXT
View Options

			Description:

			This is a simulation of pyramid-shaped objects resting on an immobile surface
			(resembling graphene). Each pyramid is built from spherical particles stacked
			like cannon-balls, or stacked fruit. Ordinarily, the stack does not move
			because the particles at the ground layer are immobilized. However,
			given an initial (small) perturbation the pyramids collapse in an avalanche.

			(In this example, the perturbation is due to shock because we (intentionally)
			did not minimize the system before starting the simulation. This shock
			causes an avalanche to occur approximately 5000 timesteps later.)

			The particles roll down the pyramid and bounce off the "ground". The bouncing
			is due to a repulsive external force which is added artificially.
			(See the "run.in" file.) The simulation looks weird without something
			to bounce off of. So I added a graphene surface at the bottom as scenery.
			(It does not exert any force on the atoms.)

			(Random comment: This could be a fun example to illustrate the Boltzmann
			distribution. Because there is no damping, in a small region, the particle
			heights should eventually approach the Boltzmann distribution for some
			suitable temperature.)