This image is of NGC4395 which is a galaxy . I did it after NGC4244 and this 397secISO6400 RAWMonochrome Autosave image is done the way I usually do 'light' frames thru MaximDL Pro V6.16 . The next two images I did I turned on the DSLR's 'in-camera' 'Long Exposure Noise Reduction ON' to fully remove whatever is 'blurring' the images' edges .. if it will and it seemed to do it . With this image as a usual 'light' frame I I made RAWMonochrome 'darks' for DARKMaster frames and I left the 'bias' in them which means I didn't use a BiasMaster to 'calibrate' the 'darks . I did seventeen 'darks' for each exposure time Sequences I did at ISO6400 . I put each DARKMaster into the 'buffer' and I did "Convert Color - Scaling 100% for Red Green and Blue and Auto for Background" , then "Make Pixels Square" , and then "Kernal Filter - Median5x5" ... then "Save As..." with the word COLOR added to it's Header name so I know which one is Color . I put them all in a special Folder from their original Folders and I put them in the "Set Calibration" command . I put this image of NGC4395 in the 'buffer' as RAWMonochrome and I did a "Convert Color" and then "Make Pixels Square" and then "Kernal Filter - Median5x5" and then I did 'Calibrate' using 'Auto Optimize' and as a Medium 'screen-stretch' I watched the image it seemed like it sunk in and the edges seemed to do the same which seemed like a lot . Then I checked the FITS Header info and the 'calibration' selected the 315secISO6400-DARKMasterTRICOLOR . I put the 'screen-stretch' to 'Max-Val' and then I used "Digital Development-GaussianBlur: 1.0" , and then "Color Balance - Scaling Red-85 Green-110 and Blue-110 and Auto for Background . This single image shows the affect the DARKMaster had on it and the other image didn't have 'calibration' ....... you can queerily see the difference between the two images which are the same image done two different ways . Maybe 315secISO6400 for the DARKMaster might be OK for all exposures using ISO6400 . . xx 152mm A+M Refractor(OfficinaStellare-1200mmF/L) and CanonEOSII RebelT7i modified UV-IR @ Hutech thru a TeleVue 2XPowerMate with 35mm T-Ring for an effect of 2400mmFL and an IDAS VLC-74 77mm 'light green/yellow tint' filter from Hutech . I did MaximDL Pro V6.16 RAWMonochrome Autosave Sequences from 67sec to 397secISO6400 for twenty-four images and in "Settings" for 'Files' I had 'checked' 'Save Data in Unsigned Format' . For the last three images I did 397secISO6400 and the first one I did as I normally do 'light' frames with 67secISO100 flush-frames between each 'light' frame ; the last two 397secISO6400 'light' frames I turned 'ON' the 'in-camera Long Exposure Noise Reduction' which took another 397sec to process in the RebelT7i before it downloaded to the laptop ... luckily the 'guiding' still keeps going . There are two other JPG's and one is a regular and the other has the 'in-camera Noise Reduction Enabled' and there is a difference around the edges which looks like the blur is removed .x {{{{{{{ I read somewhere that the Canon DSLR 'in-camera' 'Long Exposure Noise Reduction Enabled' will try to figure out the odd stray statistics in the processor which could be off-center light going into a pixel which might be bouncing off the Lens or Telescope sides and accumulating a blur which doesn't seem to be coming into the 'sensor' straight in from the Lens . It might even use a 'Bad Pixel Map' for some of the evaluation. Maybe a 120sec or 180sec exposure might be enough for straight in arriving photons and with exposures from 300sec and more the processor might be able to detect the stray accumulating light at the image's edges and internally remove it with the 'Long Exposure Noise Reduction Enabled' fully which takes awhile but it might be worth it in the long run . I will try it one night with a Canon Lens and a Hydrogen Alpha filter and for a 'Bad Pixel Map' in the RebelT7i I will do 120secISO6400 , and then do two 457secISO6400 regular 'light' frames and then do four or more 457secISO6400 'in-camera' 'Long Exposure Noise Reduction Enabled' fully which then downloads thru MaximDL Pro as RawMono 'light' frames and see if there is a noticable difference between the two different types ... if there is ... then that is a cure for the edge blurring that is difficult to remove .x }}}}}}} .xx
