Communal Night-Sky Photography Thread

[OM USer]

Nice and crisp Mike.

[MikeOxon]

After having found both the 17mm f/1.8 and 45mm f/1.8 to be very useful in low light situations, I decided to add the 75mm f/1.8 to my collection, especially since it has a very high reputation amongst Olympus lenses.


Noticing a clear sky yesterday evening, I decided to try the lens out by photographing the Pleiades cluster, which was almost overhead in the early evening.


Results at f/1.8 were very good although some coma was evident towards the corners. Stopping down to f/2.8 provided excellent results across the whole frame.


The whole frame coverage is shown below, followed by a crop of the Pleiades:

[MJ224]

Andromeda Galaxy by Mark Johnson, on Flickr

Mike, your photos inspired me to make a poor shot of the Andromeda galaxy. Work in progress I think. We do have fairly bad local street lights is my excuse...……

56mm, ISO 2500 F4, 4 seconds....

[MJ224]

Mirach, Mu And + Andromeda galaxy by Mark Johnson, on Flickr

Just a slightly better processed view of the three stars/galaxy. Processed in Faststone, will have to work out why its better than Lightroom (process that is)..

[Phill D]

Well got up at 6:30 to shoot the Moon with Venus and Jupiter. Manual focus, peaking and magnify set plus exposure set, iso adjusted, hand held but using the best IS in the business, sky was clear to start then clouds came in but then it cleared perfectly. So well prepared with the gods all aligned...or so I thought! images mainly pretty rubbish? guess what I'd done when I looked at the images...? 12-100 lens IS was switched off! so most of what I got are bright sharp squiggles Moral of the tale is get up 10 minutes earlier than you need to and have a coffee! Might go back to bed now!

[MikeOxon]

Commiserations, Phil. Those slide switches on the side of lenses seems to have a habit of switching themselves 'off' ! I have got into a routine of checking the switches on my 100-400mm after attaching it to the camera - except when I forget

[MikeOxon]

MJ224’s comment, above, set me thinking about the effects of using different Raw Converters so, being of an experimental turn of mind, I decided to put the raw data of my Pleiades shot, shown in my earlier post above, through various software packages that I happen to have.

To show what I could expect to see, I looked at the appropriate area of sky with the ‘Stellarium’ planetarium software, set for a high level of light pollution, appropriate to my urban environment.


Stellarium image

To start with, I processed my data with the new Olympus Workspace software. To allow comparison between the various converters, I displayed the image that each one produced on my screen at 200% magnification and took a screen grab. I then cropped the same area of sky from each image, as shown below:



Image processed with Olympus Workspace

The program ‘Raw Therapee’ is unusual in that it allows the user to select from several different conversion algorithms. These ‘de-mosaicing’ algorithms are all described in https://rawpedia.rawtherapee.com/Demosaicing

Astrophotography is probably not the main consideration when developing most de-mosaicing algorithms but the Rawpedia link suggests that RCD (Ratio Corrected Demosaicing) does an excellent job for round edges, for example stars in astrophotography. So, I tried that one next, using RawTherapee:


Image processed by RawTherapee - RCD algorithm

When working with very noisy, high ISO images, Rawpedia recommends using the LMMSE (linear minimum mean square-error estimation) or IGV demosaicing methods. These are said to prevent false maze patterns from appearing, and prevent the image from looking washed-out due to heavy noise reduction. I tried them both and the results looked extremely similar. The following image is made using LMMSE:


Image processed by RawTherapee - LMMSE algorithm

The default algorithm in RawTherapee is AMaZE (Aliasing Minimization and Zipper Elimination), as it claimed to yield the best results in most cases. The result on my image follows:


Image processed by RawTherapee - AMaZE algorithm

There is a fast algorithm that is not recommended in RawTherapee other than for batch processing, to compare several images. The ‘Fast’ method should not be used for the final conversion, as it is a low-quality algorithm for display purposes. The result on my image follows:


Image processed by RawTherapee - Fast algorithm

Finally, a couple of examples from commercial raw converters. I usually use DxO Optics Pro 11 because I find it provides a good result and has several useful tools for optimising an image:


Image processed using DxO Image Pro 11

I have also used Corel AfterShot 3, which is a very fast converter but I was never happy with the results:


Image processed by AfterShot Pro3

My opinion is that the Olympus Workspace produced a result that represents the stars very well, if a little soft. DxO Optics Pro has produced a grittier result, superficially sharper but with several artefacts, which has made some stars appear as doubles, when Stellarium confirms that they are not. The AMaZE algorithm has also produced some strange cup-shaped stars. The worst results are from the Fast algorithm and AfterShot Pro3, which have both introduced a lot of false colours and other artefacts.

Remember that the original image was far from perfect, being taken in an area with high light pollution and, of course, general atmospheric disturbance.

[wornish]

Quick go last night.
Hand Held. 300mm + 1.4TC , ISO400, 1/500th sec f/5.6

[Greytop]

Nice shot Dave, looks like there are some weird hatching artifacts on the surface there.
Maybe the 'Moon' film for real (Sam Rockwell etc)

[wornish]

Originally Posted by Greytop

Nice shot Dave, looks like there are some weird hatching artifacts on the surface there.
Maybe the 'Moon' film for real (Sam Rockwell etc)

You are not supposed to see them

I think I overdid the de-convolution trying to get more detail.
I tried to remove the artifacts by reducing noise but it didn't quite work.

I used an app called Astra Image which has some enhancing tools but as usual less is more. I can't resist pushing an image beyond its limit.

The only way to make handheld moon shots work for me is to use shutter speeds of a 1/1000th sec or even shorter.

Using a tracking mount makes a huge difference when at these long focal lengths.

[Phill D]

I thought you'd used some sort of canvas type effect ready for printing large

[MJ224]

Tonights Moon by Mark Johnson, on Flickr

About 20 photos stacked in Registax...

[Greytop]

Ha ha, I was at it too Mark...

One with the G9
Moon by Huw Prosser, on Flickr

And one with the E-M1 MkII
Moon by Huw Prosser, on Flickr

Both with with the Tokina 300 f/2.8 and three stacked converters.
The G9 has come out a bit better but that's no reflection on the camera body. It's basically down to atmospheric conditions and slight breeze etc

[MJ224]

Could say "SNAP" Huw,...……………..

Nice photos...…...

[wornish]

Both very nice and a lot better than my handheld moon effort.

I was out last night with my scope for the first time this year.
This is about 1 hour total worth of capture time using narrowband filters R, G and B to try and cut through the moonlight.
Can just about see The Running Man top centre.

Orion Nebula - M42