This is my latest M31 Andromeda Galaxy attempt in LRGB with my FLI ML16200 on my Takahashi FSQ-106. For this image I planned a more careful choice of exposure times than in times past. I started with calculation of optimum exposure time given my typical sky fog conditions, targeting read noise contribution of about 5%. I think those numbers for the L filter came out to be on the order of just 30s, and I increased this to 60s as part of the balance of number of frames, managing accumulative download times, etc. For the color frames I set 1x1 binning as I wanted to examine the quality of straight RGB processing with L absent. For those channels I arrived at 120s exposure, thinking that 2 - 3 times the L exposure would be adequate and close to optimum for those filters.
As a result of using exposures shorter than I have in the past, I was pleased to see that the stars were less saturated, and the galaxy detail did not at all suffer compared with previous M31 images. This is a target I tend to visit annually as I am usually trying out new equipment or processing techniques as well as wishing to gauge how much my acquisition and processing skills have improved or deteriorated year to year. I will claim that I am still seeing marginal improvement on both counts, however by no means will I declare that I am done learning nor am I done improving. (And I hope you say the same about your skills!)
The straight RGB image seemed to me to be unfinished, likely due to the aggregate luminance of the three channels not receiving the same amount of integration time as the full luminance channel. To do the RGB justice, I would have needed to likely triple the total integration time of the three channels with that of the L. This is simple mathematics and comparison of the band passes of the filters. In actuality I about matched the total RGB integration time to that of L, following my usual recipe for LRGB imaging.
I'd like to mention a few processing points:
- First, I attempted Deconvolution in PixInsight on this data during the linear phase, and, not surprisingly, I saw little to no benefit, nor any appreciable deblurring effects due to the rather large image scale of 2.33 "/px. Therefore I did not include any Deconvolution in the final image. During the nonlinear phase of processing I attempted to do some minor sharpening, and here again I decided to leave it out as the result was unsatisfactory. In my experience, I have found each process to be much more effective and pleasing when the image scale is smaller, and not surprisingly this image was not a good candidate for that.
- Second, I began my color calibration with PI's PhotometricColorCalibration process using the G2V star parameter as this usually is a good starting point. But in this case I got a rather boring looking orange-red Andromeda Galaxy that held little interest to me. I played with the parameter and settled on the K2V star parameter, and that resulted in a better starting point with core and outer arm colors as you see here. In other words, it enhanced the blue to where I wanted it to be. With a liberal amount of color and saturation curve transformations, this is the final result. I'm willing to settle with this at least for now.
- Third, and this may come across as a hopefully rare gripe on my part, it is my humble and often unexpressed opinion that many of us imagers tend to overdo the de-emphasis of the core brightness of M31 in order to draw out what little detail there may be in the data. This can often result in a tiny, rather sharp blob in the middle surrounded by an otherwise flat and mostly featureless core region. Looking back at my Andromeda images from the past couple of years, I have been guilty of this. I find this to look unnatural and would recommend instead the goal of balancing a still bright but smoothly transitional core. The core region of Andromeda is mainly featureless, particularly at wide format scales such as mine, a testament to the fact that it has an estimated 1,000,000,000,000 (that's one trillion) stars, most of which are crammed into that bright core. In this attempt I sought a balance between de-emphasizing the core but maintaining this smoother transition that is still reminiscent of what one would see through a high quality visual scope. This is my opinion, of course, and you are just as free to ignore it as I am to state it!
- Fourth and final, it is on my list of things to do in the future to start incorporating Ha data into my LRGB galaxy images, and M31 would be a great candidate for that. I have no Ha data here, but I am pleased to see that with increase of saturation and color curve enhancement, I can see a number of prominent Ha regions around the arms. This would be nicely enhanced further with Ha data, and perhaps there will still be opportunity this season to accumulate some. Otherwise, I will no doubt visit Andromeda again next year.
Thank you for your attention to what turned out to be a much longer post than I intended, but I appreciate the format on this forum for presenting and discussing these topics.
If you are still with me: as always, your comments, critiques and suggestions are most welcome. Please feel free to counter my points and positions as I look forward to the discussion.
Please be sure to take a look at the images on AstroBin as well as the lengthier Description there.
Higher Resolution AstroBin Links:
- LRGB Image (lower resolution copy posted here)
- Luminance Image
- Inverted Luminance Image
- Annotated LRGB Image