OSC NGC 300 Workflow part 2, PCC

Hi Adam,
I just watched the NGC 300 workflow. 30 minutes in part 2 (https://www.adamblockstudios.com/articles/ngc-300-part-2) you discuss difficulties with PCC producing white stars.  The dataset is from f2.2 RASA with 2-minute exposures and I wonder if the stars are over exposed. My C11 Hyperstar v4 with f1.9 and exposures > 90 seconds produce stars >.99 cores; and PCC produces white stars as well. I confirmed the NGC 300 workflow dataset also has many .99 star cores. Should the exposures be reduced to avoid star cores with very high values?

Best,
Jim

Comments

  • Hi James,

    I think the biggest issue is that the CMOS sensors tend to be only 12-bit or 14-bit sensors. This really limits options to control saturation of bright stars. These sensors have the combination of being both sensitive and "shallow" in terms of dynamic range. 

    So what to do? I am of the mind that what is most important are the processing steps that make the primary object (the subject of the image) look best. I think sometimes imagers, in a case like this with a faint galaxy, try to make stars look good at the expense of the galaxy. If the image was just about the stars...if they were the subject..it would matter more to me.

    However, these days it is possible to do as you are inclined to do- which is take shorter exposures for a separate set of star data and blend this (not brightening the stars as much) with the starless version of the galaxy. There are some downsides to doing this.. but I guess this is a way to do an end-run around the issue.

    -the Blockhead

  • Adam,

    Thanks for the expanded explanation! I use the Sony IMX455 16-bit ADC CMOS sensor with at zero gain for 14 stops DR. Hopefully, I am avoiding too shallow captures.
    I see your point on balancing tradeoffs, for example, the field of stars versus the galaxy as featured subject. I aim to visually optimize both. I will try separating stars and starless workflows and then blend together.

    Best,
    Jim
Sign In or Register to comment.