Figure 7
Variation of the mean cluster size (S) vs. the threshold parameter q for different numbers of nodes in a (a) triangular and (b) square lattice (Simulated steps \(t=20,000\), averaged over 100 runs). For \(q \gg q_c\) the typical sizes are much smaller than the system size and the cluster size does not depend on N. For \(q \approx q_c\) finite-size scaling appears and the cluster size depends on the system size. The similar behaviour of the two largest networks indicates that these systems did not finish growth after the 20, 000 time steps. (c)/(d) (upper) Relative cluster size of the largest cluster \(C_{max}\) vs. the threshold parameter q on an triangular (\(N=392\)) and square (\(N = 784\)) lattice (After \(t=20,000\) simulated steps, averaged over 100 runs). (lower) The fraction of the different tables depending on q. Here, \(T_X\) denotes the different type of possible translation tables as described in the full-text. The light area around the curves indicates the standard deviation. (After \(t=10,000\) simulated steps, averaged over 100 runs).
