latticegpu.jl/docs/src/flow.md

Gradient flow

The gradient flow equations can be integrated in two different ways:

1. Using a fixed step-size integrator. In this approach one fixes the step size \epsilon and the links are evolved from V_\mu(t) to V_\mu(t +\epsilon) using some integration scheme.
2. Using an adaptive step-size integrator. In this approach one fixes the tolerance h and the links are evolved for a time $t_{\rm end}$ (i.e. from V_\mu(t) to V_\mu(t +t_{\rm end})) with the condition that the maximum error while advancing is not larger than h.

In general adaptive step size integrators are much more efficient, but one loses the possibility to measure flow quantities at the intermediate times \epsilon, 2\epsilon, 3\epsilon,.... Adaptive step size integrators are ideal for finite size scaling studies, while a mix of both integrators is the most efficient approach in scale setting applications.

Integration schemes

FlowIntr
wfl_euler
zfl_euler
wfl_rk2
zfl_rk2
wfl_rk3
zfl_rk3

Integrating the flow equations

flw
flw_adapt

Observables

Eoft_plaq
Eoft_clover
Qtop