Docs/Build/Baselines: where it wins
CLIENT 0.4.3PYTHON 3.8+

Baselines: where it wins

The honest three-zone operating map.

SHARED: On static or slowly drifting plants, use a commissioned PI, Anderson-class method, adaptive RLS, or an exact model when its setup fits the budget. These methods can match or beat TrueLoop, and RLS catches the confirmed regulation cell when the deadline expands from six to twelve acquisitions. If your plant is stable enough to identify once and trust, you do not need TrueLoop, and we will tell you so.

EXCLUSIVE: TrueLoop separates when drift approaches or exceeds 0.1 rad per acquisition, coupling is global, the acquisition budget is tight, and per-channel behavior is unknown at power-on. The exclusive tested range begins at n ≥ 1,024. At n = 256, the measured response floor sits on the usability line and marginal passes are expected.

DECLINED: Do not use TrueLoop past the measured response floor, outside the actuator span, on scalar-only objectives, with infeasible or extremal targets, or when readout quality is insufficient. No controller wins past the qualified drift and coupling boundary.

For compatible optimization, compare against both vector-response controllers and scalar optimizers. Finite differences, SPSA, CMA-ES, coordinate search, and related scalar methods receive a different information interface when they receive only one score.

Need the answer on your plant?

Run the envelope, then race the runtime and your incumbent under the same measured budget.

Start evaluation