This before that or vice versa?
10/12/2015. Workflow isn't everything but it's plenty. Last night, I got several good frames of the Double Cluster (without an impinging dewshield) and, after some clouds, several hours of the Pleiades in decent but not spectacular skies (it's worth noting that the G11's extension allows the camera to clear the pier at this declination, so the mechanical power-off timer is nice but optional with this outfit south of about 35N).
Today, I have been refining the workflow in Pixinsight to get the most, best pixels. It's not easy on a relatively weak i3 CPU with 8GB under Windows 7. An SSD swap-disk helps; dual Xeons or an i7 and at least twice the RAM would help more. There are a lot of options to try and they all take time.
First, a pretty picture so you can ignore the techno-babble if you're of a mind to do that (No, I don't know where most of the color went; will try restacking these using lessons learned with the Pleiades).
23x180s @ ISO 1600
No calibration. Click the pic.
First thing this morning (and for the next five or so hours) I stacked the best 64 frames of the Pleiades from a run that extended all the way from clouds in the east to trees in the west. I got a mess for my trouble. The noise in the finished image looked as if I had not bothered with darks and bias frames (I had). In fact, it looked worse than I would expect based on my older, uncalibrated, half-size workflow. I had selected "pure raw" for the format conversion, a new option in this updated Pixinsight installation, and I thought that might have made my darks ineffective. I changed to VNG Debayering. I picked four subframes from the night's images and tried several variations.
The workflow is somewhat forgiving if you get some key elements right: the debayer step is crucial. Put it off as long as possible to minimize the amount of data you throw around. I may have thrown away considerable good data in by clipping a little too tightly in the pixel rejection parameters. The bright rejection map seemed to have a lot of useful signal [no, that made little difference, actually]. The dark was an average of 24 .CR2 files opened using the "Debayer, VNG" setting. They were shot in the refrigerator (I'm estimating 37F) so the dark probably left some noise behind (I was imaging at about 55F). The bias frame was produced similarly (though I am not sure about the original .CR2 opening method) and the "Super Bias" option applied.
64x180s ISO 1600
Clickable version on next page.
Best yet, but still not happy. Not quite as quiet in the fainter parts of the reflection nebula as I'd like, not quite as detailed in the brighter parts. Probably lost a lot of detail trying to clean it up. Time to learn some more about masking and much more about noise reduction in Pixinsight. While the files were loaded and all the integration parms were cached, I experimented with different pixel exclusion rules. "No rejection" is a non-starter; "Linear fit" tossed a lot of shadow detail; fairly radical changes to low and high sigma rejection made little difference to the Winsorized scheme and not that much to the standard sigma rubrics. All in all: use Winsor and take the defaults. And it might be different with a different target or with data collected on a darker night. Now back to the Double Cluster data. And the Veil. And Rho Oph. And the rest of the universe.
Solved! See next page.
Except where noted, deep-sky photos are made with an SBIG ST2000XM CCD behind a 10-inch Astro-Tech Ritchey-Chretien carried on an Astro-Physics Mach1GTO. The CCD is equipped with Baader wide- and narrow-band filters. The internal guide chip of the CCD most often keeps the OTA pointed in the right direction (I'll let you know when an OAG or guidescope takes its place). Camera control and guiding are handled by Maxim DL 5.12. The stock focuser on the AT10RC has been augmented with Robofocus 3.0.9 using adapters turned on the lathe downstairs. A Canon 6D and a modded 50D find themselves mounted on an Orion 10" F4 Newtonian or carrying widefield glass on an iOptron Skytracker. Beginning in May 2013, PixInsight has taken over more and more of the heavy lifting -- alignment, stacking, gradient removal, noise-reduction, transfer function modification, color calibration, and deconvolution. Photoshop CS4 et seq and the Focus Magic plugin get their licks in, too.
:: top ::