DSS or Nebulosity for Stacking?

[BlueAstra]

Just for interest I ran some subs through DSS and Nebulosity to see how the stacking compares. I used only subs, flats and bias with each one, but in DSS allowed the 'cosmetics' to take out dead pixels, whereas in Nebulosity used a dead pixel map. My SXV-H16 is quite low noise, so I don't usually bother with darks. In each case I applied identical curves in PS to boost the image, and approximately the same black level adjustment, but no other processing. I think you can see the Nebulosity version seems to show more detail, but I think the cosmetic bad pixel elimination in DSS was better. Its the first time I've used a bad pixel map so I think I'll have to look into the parameters to improve the rejection ratio. Nebulosity also creates a huge number of files, with a set of images created at each stage of the process. I think there's enough of a difference to look a bit more closely at using Nebulosity rather than DSS.

Nebulosity Image

DSS Image

[Jessun]

Or try Astroart! One image in the end and that's it. No clicking on stars or waiting eons for the data to be processed... But I agree that Neb beats DSS.

[Sp@ce_d]

Yeah I tend to use Neb most of the time.. but I still like to check a comparison in DSS now and again. Must get around to trying out AA

[BlueAstra]

Everything is stalled at the moment while I try to figure out the best route for processing. I've gone down the full Nebulosity route, but found that even with a SD stack I always seemed to get a few dark pixels. I suppose it could be the bad pixel map, but not sure. (BPM, Pre-Proc with flats & bias, Normalisation, Align, Align SD1.5/2.0). I then tried the hybrid Nebulosity approach (Neb BPM, DSS with flats & bias) and this seemed to work a little better in that I didn't get any dark pixels (this could be due to the 'cosmetic' setting in DSS). The overall image quality seemed similar. I've included two sample images (full frame SXVH16 without FR), I went a bit mad with colour with the Nebulosity version so please ignore that aspect of it. So I think my future processing route will be to BPM the subs in Nebulosity, then submit them to DSS with just flats and bias, then use PS on the result.

Nebulosity only:

Nebulosity BPM & DSS:

[Jessun]

Looking promising - both routes I'd say judging by your samples. A dither between the subs will help too by moving those bad pixels around so that a sigma stack can kill most them off.