"neel" simulates the Ising model with exchange J, but with the addition of a long range dipolar interaction among the spins in addition to the nearest neighbor exchange. In this preset the Ising spins are aligned perpendicular to the plane of the array so that the dipolar interaction is negative for all pairs and equals D(a/r)^3.
In this first preset the dipolar interaction is set to zero and the exchange J = -1 to simulate the simplest antiferromagnet with nearest neighbor (nn) interactions only. Run to see the condensation into the antiferromagnetic state. As with the ferromagnetic transition, a short run leaves the system with a number of grains or domains separated by domain boundaries. What is the difference between the two kinds of antiferromagnetic domains? It is much less obvious than the "spin up" versus "spin down" in the ferromagnetic case.
To see the effect on the magnetic susceptibility of the negative exchange, change T to 3 and H to 2, reset the graph, and run. Increase exchange J to -0.5 and run again. Repeat for J's up to +1. Why does J influence the magnetization?