AIC 2012 has begun! Day one has been filled with some excellent sessions. The exhibitor hall is open and there will be a two-hour opening later this evening.

My first session today was Ron Wodaski’s introduction to PixInsight. He covered many of the basics (that’s a lot) and gave some glimpses of the advanced features in the platform. Even an experience PixInsight user like me found it a useful and informative presentation. I am looking forward to seeing the full deck with all the details Ron promised.

I then attended Tim Puckett’s presentation on supernovae searching. He has built some amazing telescopes and quite a robust process and team for finding supernovae. When it is clear, his team will image 2,000 galaxies to find potential supernova. I learned that fit order 3 or 4 is required in PinPoint for astrometric reporting, and how to access the reference magnitude function in Maxim DL.

After lunch, there was a great presentation by Steve Brady, co-developer of the great focusing program, FocusMax. FocusMax uses the idea that there is linear relationship of the half flux diameter or HFD to focus that can be used to quickly and accurately focus a telescope using a CCD. HFD is the circle that evenly splits the flux or total ADU of a star regardless of focus. It is expressed in pixels. Max flux for a system is best calculated by plotting max pixel and max flux vs. time and picking a max flux value below where the curve flattens out. Focus convergence was clearly shown to be the best way to get to focus, even though it takes longer than the default of 5 exposures.

There was a very interesting discussion on precision vs. accuracy in FocusMax, and the key point is the size of the critical focus zone. More reading will be needed on that later. Focus is driven by the Vcurve, which is a hyperbola. Fit of measurements in creating a Vcurve are fit to the hyperbola, not to a linear fit of each side of the v. One should pick the point just above where the Vcurve moves away from the line as the near focus HFD, with the start HFD being about five units higher. Get at least 12 Vcurves for best results.

A major new release is coming sometime in the relatively near future. For now, the beta version is quite stable and is the one to use.

Finally, there was a great presentation from Ken Crawford on using masks in Photoshop CS5. I think I understand alpha channels now — they are containers for masks. He will have a video tutorial for all AIC attendees.