Feature #340

Optimize Romberg integrator precision

Added by Knödlseder Jürgen almost 12 years ago. Updated over 9 years ago.

Status:ClosedStart date:07/18/2012
Priority:NormalDue date:
Assigned To:Knödlseder Jürgen% Done:

100%

Category:-
Target version:1.0.0
Duration:

Description

The precision of the Romberg integrator has been set arbitrary to 1e-6. This was the original value taken from Numerical recipes.

Several studies have shown so far that a precision of 1e-5 seem to be sufficient, at least for CTA response computations (see for example GCTAResponse). We should investigate, based on extensive tests, what the default precision should be.

Recurrence

No recurrence.

History

#1 Updated by Knödlseder Jürgen almost 10 years ago

  • Target version set to 1.0.0

#2 Updated by Knödlseder Jürgen over 9 years ago

  • Status changed from New to Closed
  • Assigned To set to Knödlseder Jürgen
  • % Done changed from 0 to 100

The precision is no longer use for response computation, but rather the integration depth. This considerable stabilizes the convergence behavior. Consider as closed now.

Also available in: Atom PDF