I had an unexpected clear, but steamy hot, night recently in which I decided to try to catch a glimpse of M13, the Hercules Globular Cluster, discovered by Edmond Halley in 1714.  It was so warm that my camera could only cool down to -5 but the humidity was not too bad.  So, I took a chance and was able to collect about 4 hours of data.  What I really wanted to try was the Generalized Hyperbolic Stretch (GHS) to see if I could get a better star cluster vs. when I tried in the past using the traditional HT in PixInsight (PI).  LRGB imaging in my area is tough due to the sky conditions but I was hoping that this bright, large cluster would give me an opportunity to explore the differences to some extent. 

I have 3 images posted:

Original is using the GSH tool for the first time.  I made about 20 versions of the luminance and RGB while continuously watching the excellent video by David Payne and Mike Cranfield.  The challenge I had here was where to place the Symmetry Point.  Its placement seems straight forward when imaging a nebula but there really is no border region for a star cluster.     

Version B is using both MaskedStretched and ArcsinhStretch, which I have only used a few times in the past.  So, these were essentially another test approach for me.

Version C is using the more traditional HistogramTransformation and CurvesTransformation combination which I mostly use.

Versions E (wide field) and F (cropped) are using GHS but with a slight color punch added afterwards using Curves Transformation.

Version G is using version F but with a second run-through with GSH.  However, the SP location, B, and D+1 factors are significantly different than the first time it was used.  It could be my preference vs. what most expect to see, but I don't light brightening up the core any further and am trying to keep the colors from getting too cartoonist looking.   The most significant thing about doing a second run-through with GSH, in my opinion, was I could clearly see more how each parameter is doing what to this image.  I suspect it will act very differently on other, non-cluster, images.   

Version H Is essentially version G with a bit of core enhancement.  I am not sure I like this as well as version G but I think most folks prefer star clusters with the center more enhanced. 

In all three versions, I had a specific goal to capture a relatively clear image without oversaturating the stars, both in intensity and color.   In other words, I prefer a bright, yet a softer, more pastel looking image.

The first thing I noticed is that I did not see any dramatic difference between the three images.  Now, that could be due to me not using the GSH properly as it does have a learning curve.  Also, some of the stars are near saturation despite my efforts to avoid that from happening.  Some of my initial images from all three methods had significantly color saturated stars.  While they might look cool, they seem unrealistic to me - at least from my location.  Overall, I was somewhat disappointed in the GSH image but I realize I have a lot of testing and learning to do with it.  I am very anxious to try it on a nebula from my location.

I would really like feedback from anyone who is using the GSH approach, or for any approach for that matter. 

Finally, I would like to thank my friend, Mike Feigenbaum, for his wonderful help and suggestions for using the MaskedStretch/ArcsinhStretch and GSH tools.