Can anyone confirm Duncs settings? An offset of 4 sounds strange for me, but if it works for you, Dunc, maybe i should give it a try? Or are we loosing too much dynamic range with this setting?The exposure recommended was 255s at gain 120 with an Offset of 4. The drivers default to 28 and 30 (ASCOM/direct) on my laptop when clean installed. With an offset of 4 I get lovely data.
I have rectified the artifact by generating artificial flats from my stacked images before applying any other post-processing, however it would be better to eliminate it at the capture stage if possible.
Don't face one's fate, can't be a real man;
Don't know manners, can't be gentle;
Don't know meaning behind the words, can't understand man's heart.
I am cautiously optimistic but have not been able to do conclusive testing so far thanks to the UK weather. A few nights ago I tried the same target (M51) but seeing was poor and the moon was bright, so my exposure length was very limited. Using the standard Unity Gain setting with Offset 30, 10 second exposures gave a histogram range of 2k to 65k (saturated highlights). Reducing the Offset to 14 allowed me to increase the exposure to 20 seconds, with a range of 800 to 65k, so better dynamic range than the standard setting.
I shot 150 subs and although the image quality was poor (due to bad seeing and insufficient target data), the purple/green artefact was definitely less noticeable than usual in the stacked image, but could still be seen by stretching and increasing saturation, so it was not entirely eliminated. I checked Dark frames shot at this setting and the bottom end was not clipped as I originally feared, so calibration was not impacted.
I'm definitely encouraged by this, but would like to repeat the test on a better night, first using standard unity gain setting, then an appropriate combination of low Offset, 120 Gain and exposure to suit. This should give a good comparison of the impact on the colour issue. Sadly the moon is full this week, so it will be a while before I can complete the experiment