I'm trying to get some info from MaxIm because in one of their updates some time ago they changed the active star detection area, for lack of a better term, for the guide chip. It went from some unknown boundary to a new smaller boundary where the guide star will be detected by MaxIm. Seems it use to be small like maybe 5 pixels. By boundary I mean like an invisible rectangle that the guide star must be in before MaxIm will see and use it. I use to calibrate by picking a star at a PA0, east of the meridian, and as close to DEC 0 as possible. I'd position my guide star in the far upper left corner because the calibration for me would move the star from the upper left across to the right, back again , and then down, and back. I set both my move times (X & Y) to move the star as far as I could down (Y) without going off the chip. This was done to run out any possible backlash and usually gave a very accurate calibration that would not be changed for years short of a change in equipment.

Now that active guide window has shrunk and I can't seem to find the dimensions so I can draw a inner rectangle to be sure the star is usable. Of course center is ideal but not always possible when framing the target. The issue comes into play when your guiding is set to dither. I've had a series start and get multiple images but with the dither set to -1 in Scheduler targets I have no idea which way or how much that star may move and I'm suspecting that it may be enough at times to move it out of the active area of the usable guider chip. What I'm saying is I think it's possible to have multiple guiding failures of the same target that you may already have some images taken of and not have any idea why unless you stay up all night and watch the process.

Now maybe I need to get a better understanding on how dithering works so I can control this better and not accidently go outside the active detection area of the guide chip or not use the dithering option when setting targets up. I usually have large sets of data to process so the combine methods would take care of the outlier artifacts I would think. But it would be good to understand this better and use the best method.