Does the internal guiding module count the movement caused by dithering when calculating the RMS error? I've noticed that my RMS error will trend down while capturing an image, then jump up after the dither. Definitely not a major issue, just something I (think I) noticed.

I'm not sure if it does. But if it is a real physical motion being recorded, you may want to have a couple seconds set to settle after the dither completes.

I have noticed this as well, I have a minute settle after dithering and I still have this issue as well. It seems the guide star is off a little after dithering and the error goes into the RMS error no matter how long the guiding settle is.

Yes, I am also seeing the same thing.

Me too. This is one of the issues I noticed on a regular basis but for whatever reason I never reported.

I see it here too, settling time seems to have little if any effect.

Looking at gmath.cpp, calc_square_err is looping through all values of the drift array to calculate the RMS error, and I believe performProcessing is adding all movements (including dither movements) to the drift array, but I'm also not familiar enough with the guiding code to jump in and make any changes.

I suspect "Dither settle" is not what you're looking for. Dither settle is the time span between dithering being finished and guiding being resumed. I never quite understood what this is supposed to achieve. What you're probably looking for is "Guiding settle" which is (illogically, I might add) placed at Ekos Setup -> Options -> Ekos -> Capture -> Guiding settle. That's the time span between guiding being resumed and the next capture started. This (set to 13s here) provides enough time for guiding to get to a stable state before the next exposure is triggered. I've been advocating for moving this setting to the Capture Tab -> Options before where I think an "average user" would expect it to be.

Yes, Herrhausen, you are quite correct, however RMS values are still affected because of the shift in star position after dither. I do run dither settle for about 20 seconds or so, as my stars would appear egg shaped or rather Tadpole shaped because of the residual movement of the guiding action after dither. I think what may need to happen is RMS values should be taken AFTER GUIDING SETTLE time. This would at least eliminate the RMS values from showing incorrectly high values.

Totally agree!

The RMS error that is indicated does not affect the performance, even though its calculation is probably not correct as many of us have noticed by now.

What would be more important is to command the guide module to search for a new guide star following each dither event and then restart guiding. From the movements I am seeing on the guide plot, it looks to me like the module is moving the mount BACK to where it was before it started dithering, instead of resetting the position of the green square onto the shifted position of the guide star and then continuing from there. If what I am seeing is correct, that would negate the entire effect of dithering and explain why I still have these streaks in my OSC images, event though I am dithering like crazy.

Alfred, what do you think? Does that match what you are seeing?

Jo

Jo, the more I remember the guiding I've done before, the more I think you might be correct. I've noticed in processing that the offsets when aligning are pretty small even over several images, leading me to believe we aren't actually getting much out of the dither.