Three real mean-square-displacement rows from a real physim Brownian-motion simulation (scenario brownian-motion-diffusion, D=0.5, N=2000 particles, seed=11), each fitting back a D close to, not exactly, the real input value. Exact arithmetic here means exact results for the stated model inputs; measured inputs still carry uncertainty and significant-figure limits.

highlighted = computed this step

Two real MSD rows each fit back a D close to the input

A real physim Brownian-motion simulation tracks 2000 independently diffusing particles with input diffusion coefficient D=0.5000 square metres per second. At real time t=2 seconds, the real mean-square displacement is 1.9487, fitting back D=0.4872. At t=5 seconds, real MSD=4.9167 fits back D=0.4917.

1.9487/(22)0.48721.9487/(2\cdot2)\approx0.4872
Real MSD fits back the input diffusion coefficientEach real MSD row recovers a D close to, not exactly, the real input D.input D=0.5000 m^2/st=2 s: msd=1.9487, D_fit=0.4872t=5 s: msd=4.9167, D_fit=0.4917

The third real row is still close, not exact

At real time t=10 seconds, real MSD=9.3242 fits back D=0.4662 — within a real 6.7580 percent of the input D, the largest gap of the three because a single real seeded run is one finite sample, not an infinite ensemble.

0.46620.50000.4662\approx0.5000
Real MSD fits back the input diffusion coefficientEach real MSD row recovers a D close to, not exactly, the real input D.input D=0.5000 m^2/st=2 s: msd=1.9487, D_fit=0.4872t=5 s: msd=4.9167, D_fit=0.4917t=10 s: msd=9.3242, D_fit=0.4662