My first trip with the new rig out to dark skies - thanks for the hospitality Chris.
This is taken with my Tak FS-102, focal reduced to F6, unmodded 350D, 10 x 5mins at ISO800.
Processed in Maxim - on the steep learning curve at present. I used Michael Covington's Tutorial. He suggested boosting the Red channel to >200% when converting the files to colour - any comments on that???
I then did a Gamma Stretch & Linear Stretch, followed by some fiddles in photoshop - I reduced the red a bit as I thought it was too much.
I think I have blown out the core. Maybe some more shots with less exposure to merge into the image.
Also, the faint red "star" in the field - are they real or are they hot pixels??
I have another 4 runs to process, but that will have to wait until another night...
Thanks for looking
David T
Last edited by DavidTrap; 16-05-2010 at 01:33 PM.
Reason: added lots of images
Nice image David! The final colour appears to have turned out quite well, no matter how you got there! I'm not sure about the hot pixels - it would be good if you could post a full sized image on pbase or something. The core is a little overcooked as you say - I would suggest reading Wodaski's book "The AstroZone System". I've just read it, and he shows how to use curves in Photoshop to stretch the image without burning out the core; it's quite a good book but a bit expensive ($180 I recall).
An unmodded camera is way less sensitive to Ha and in the red channel. That is true.
I don't know that a rote formula of increasing red 200% is the way to go. I would suggest each image may have different requirements.
In your image here the red is overdone so that your background is too red.
Shadows/highlights is a good tool to use. Try moving the highlights sliders and you may find that overblown core may still have data there.
Also it can be a bit of challenge to bring out the star colours with a DSLR but it can be done. There are some real DSLR experts on this forum. RB, Humayun and others are very good with it. I'd be asking them for their processing workflow.
Also you should consider modding your 350D or picking one up second hand (they are quite inexpensive these days). The improvement in sensitivity will blow you away.
Greg.
Last edited by gregbradley; 16-05-2010 at 10:07 AM.
Greg - a mod is certainly on the agenda. I picked up this camera second-hand including an AC adaptor and DSUSB interface for a reasonable price. I'm hoping to use this as a stepping stone to a monochrome CCD in the future...
Im no DSLR expert but I started with a unmodded 350D.
My old workflow from memory was something like:
forget about neb colour to start with. Firstly I would duplicate the image and get the star colour approaching something that was pleasing. My favourite way was to select the stars either using Carbonis or my actions, expand (Select->modify->expand) the selection by 1 or 2 pixles (depends on star sizes) and adjust the star colour using Hue and Saturation slider (usually just saturation increase would be ok but red stars might still be weak. Also adjust background colour to taste on this image. Then flatten and save this as image 2.
Then go back to image 1 and boost the red by whatever means you decide to use - I use the colour channel enhancement from Starizona (I recorded an action for on my website). Then adjust neb colour to taste. Save this image 1.
Then copy and paste image 1 into image 2 and layer mask it and paint through the nebulosity. This may require several layer masks at varying opacities towards the edges to blend it in. Then flatten and save this as image 3. So image 3 should have the red neb with the correct coloured background and stars around it. however you will still have red biased stars in the neb.
So then paste image 2 behind image 3 as a layer mask and bring through just the correct coloured stars inside the neb using a small paintbrush. Which should give you consistent star colours across the image. flatten and save.
The trick is matching backgrounds and achieving a smooth transition between the neb and the background. If you play with it you can achieve a better result.
I'm following a couple of recipes that I've found on the web for Maxim DL. I'll try those and seek constructive criticism first, but will talk to you in due course.
Nice collection there David, particulary like your M16,nicely resolved core. My only suggestion is get rid of any colour bias before trying to colour balance. Thanks for sharing
Doug - I wanted to try the system out on a range of objects to see if I'm onto a winner or not - I'm pretty happy with the outcomes. I'll try to go for higher quality shots on the next outing. Still at the bottom of the learning curve!
I like them a lot David - in particular M16's Pillars. Very nicely done. Maybe the colour is a little psychedelic for my taste (I like muted, natural tones), but that's just processing. You're getting the drift pretty quickly...
I'm sending the camera away tomorrow to get it modded by Eric. My plan next time is to try 7-10 min exposures at 400ISO and see how that goes.
I take the point that concentrating on one object per night might improve quality, but I was shooting about an hour on each object. By the time I was properly polar aligned and the GOTO system aligned, it was about 8pm. I then shot constantly until 2:30 am when my toes were too cold. I parked the scope, shot some flats and then setup a run to do another ten dark frames whilst I went to bed. Ultimately, this new moon was a proving exercise for the new rig.
My plan next time is to lengthen the exposures, increase the number of exposures; and shoot some darks at the beginning, middle and end of the night. I want to spend about 2 hours per object. I hope this will be a compromise between quality and quantity.
2 hours at a minimum will give you splendid results. A lot of my early work was just that, 2 hours a pop at ISO-400, 5 minutes each. With the greater higher ISO capabilities and noise characteristics of sensors nowadays (Canon 40D and above), 10-minutes in cold weather is a breeze. Don't even bother in summertime, it's just plain horrible.
