M31 reprocessed

[JohnH]

This is a reprocessed version of M31, all data gathered on 3rd November.

ASI1600MM Pro.
Sharpstar 15028 HNT

LHaRGB

As an experiment I have thrown Pixinsight at it. To me, I am astonished by the amount of detail I have captured even though it is still fairly noisy and I suspect that some of the nebulosity is processing artefact.

To be fair:

1) There was a 95% illuminated Moon about 60 degrees away from M31,
2) there was a lot of blotching in the background which I have struggled to get rid of, and
2) I still do not really know what I am doing!

However, I do like to think I am getting better.

John

[NoelC]

Excellent scale and a huge amount of detail there John.
Very well done getting to grips with PI (something I've not managed).

Speaking as a non-expert; I suspect the sharpening has overcooked some of the stars making them a bit hollow, and there is a distinct fabric texture around the M32 and a sort of curdled look to some of the nebulosity as if it's been over sharpened. Is it possible to back off the deconvolution a few notches?

But the detail and colour you've captured makes a very pleasing image.  Well done.
Noel

[Roberto]

Looking very good John!  The dust lanes you have managed to extract look amazing 

Roberto

[JohnH]

Noel and Roberto, thanks for the encouragement.

I really did throw the Pixinsight equivalent of the kitchen sink at the image to see what could be done and so I am not surprised it looks a bit overdone.

From further reading I suspect that I have been taken in by the "Focal Ratio Myth". I had been working on the basis that f2.8 meant that I needed less time per channel but it really means more but shorter exposures. So my L filter was 100 x 20 second exposures and benefited enormously when turned into a "Superluminance" by adding 30 x 40 second reds, 20 x 90 second greens, 20 x 120 second blues and 10 x 300 second Ha. However, the colour filter frames should have been a much longer total exposure.

Also, the exposure times of the colour frames was arbitrary. I have now taken the transmission curves of the filters and the Qe curve of the camera and done a back of the envelope approximation which suggests that the overall transmission across the wavelength range of the RGB filters is consistent.

It is all part of the learning curve.

John

[Carole]

Now that is looking great, well done. 
Also well done for mastering Pixinsight.

Carole