Test for need to recenter

With the speed of plate solving, I was wondering if it would be worthwhile including an option that platesolves every X images during a sequence to test if re-centering is necessary. There is already an option to re-center every X images however this process might be unnecessary if there has been no shift. In contrast if there has been a shift for some reason it would be worth interrupting image capture to re-center.

Hi Peter,

if you want to check … is better to do recentering. Realign is necessary if you not use autofocus with robostar each x sub that recenter the target automatically (for example if you not focus during sequence or using localfield).

All the best

Hi Leonardo,

For users that are using localfield I was thinking a plate solve could occur in parallel processing after image download while the next image is being acquired. If the platesolve solution indicated a pointing error greater than a specified tolerance then a re-centering routine could be triggered after the current image capture was complete.

I can see how robostar would address this issue however I understood localfield was a better focus choice for imaging systems that were not perfectly flat.


In this way can I think about … but not all of us have fast plate solving … and plate solving is something blocking (you can run one instance) so risk is to wait for plate solving if some necessary operations required it also if working in parallel.

All the best

I am finding ASTAP will resolve plate solving solution in 1-2 seconds on a 10 year old laptop so it would not introduce much delay in my workflow.