I've had issues in the past with not managing to improve on final alignment steps, and found that switching to nearest star rather than n-star in the align options helped.

I'm not sure any of the above issues relate to the fault I am seeing in the Ekos/StellarSolver space, though. The behaviour is quite different and given that this occurs *without the EQMOD driver involved* in a simulator I think it unrelated. I've tried both the Stellarsolver internal SEP and internal solver as well as external solvers/SExtractor (my usual setup) and that all has the same issue.

Read More...