Astro 950

See my other Astrophoto 950 pages also!

Another more recent page of shots.     Two really nice shots of Orion

How I took these shots.

Time exposures of stars. (updated March 16, 2000)

In these first tests, the Nikon CP950 was able to photograph stars down to about 5th magnitude using the maximum 8 second exposure. I took these shots on the night of April 23, 1999. Their was a first quarter moon, and my sky is fairly light polluted. There was still a little twilight visible. I think that it might be possible to do better with better skys. These shots were processed in the following way...

* I used a tripod! (these are unguided, only 8 seconds)

* I forced the flash off, set Shutter priority in M-REC mode, set the exposure for 8 seconds, set the camera on infinity focus.

* I also took a "Dark field" at the same time, an 8 second exposure with my hand over the lens.

* Using Arcsoft PhotoStudio (Adobe Photoshop can do this too, but I don't have it so I can't tell you how to do it), I first subtracted the dark frame from each shot. This is needed because on my camera, I have about 20 noisey pixels that start to show up after about 2 seconds of exposure. (these are always the primary colors of the CCD, red, green or blue). I think that all 950s have this noise on long exposures.

* I boosted the highlights to make the stars more visible for the web at reduced size. In the original, this isn't necessary, this also made the background a little lighter. The originals are better looking, but don't show up well in reduced size for the web..

* I re-sized it and saved as a low quality JPEG to save space. On the original, the sky is a uniform black.

Scroll past the pictures for a few comments.



(Above) Brillliant Venus below Auriga. The bright star to the upper right is Capella. The streak at the bottom is from a house light out of the field. April 23, 1999 ISO 80, 8 second shot.


You know what this is! The big dipper. Ursa Major. Note the Alcor-Mizar pair!  This is a new ISO 320 shot that replaces the earlier shot. On the original, I think that I can see some 6th magnitude stars. April 24, 1999. ISO 320, 8 second shot.


The same shot of Ursa Major, but processed with a threshold filter. I picked a level that excluded what I felt was noise, but kept the valid star images. I looked at my copy of Tirion Sky Atlas 2000 while I did this. I also ran a blur filter to increase the size of the stars for the web. I think that there are some 6th mag stars in there, check your charts!



Here is a crop of the bowl stars in the big dipper. I've enhanced the star colors, hope they are visible on your monitor. A neat project would be to catalog stars by spectral type based on color.


On the originals, you can see colors on the stars better.

On a picture of Orion, which was setting, the M42 nebula was visible.

I was able to photograph an airplane trail.

Using the 3x zoom, there is just slight trailing on the stars in these unguided shots.

Hot Pixels and Dark Frames.

On long exposures (more than 2 seconds) in low light, you will see some "hot" pixels from the CCD. They look like colored stars, red, green, blue or purple. These are "normal" for the 950. They are caused by charge leakage on some of the individual cells on the CCD that make some pixels appear brighter than they should be. They do not affect exposures in normal lighting. To remove them from your star pictures, you need to take a "dark frame", which is a picture with no external light coming into the camera. You then subtract this frame from your actual exposure.

I've had some questions about how I subtracted the dark frames. Here are some sample dark frames below. They are all re-sized to 400x300 to fit on my site, so you can't see all of the "hot" pixels.


A "raw" dark frame, just re-sized. Dark, isn't it!  To take this shot, I set the camera to ISO 320, no flash, aperture priority, aperture wide open, put on the lens cap, went into a dark room, and took an 8 second exposure. You can't see much here, so from now on, I will "enhance" the highlights on the frames so they are more visible on the web. All were enhanced by exactly the same amount.


Dark Frame #1. The same shot as above, but with the highlights enhanced. Note three things. 1) the colored "hot" pixels. 2) The glow to the upper right (more on that later), 3) the wavy pattern. (Don't forget, I enhanced the highlights on these so they are visible on the web)


Dark frame #2. A separate exposure from #1. They sure look similar don't they? That is good, because that means that the pattern of hot pixels is reproducable. If the frames were truely identical, we should be able to subtract one from the other and get black.


Dark Frame #2 minus Dark frame #1. Enhanced as the previous shots, pretty close to black! Just a couple of random pixels. This proves that the noise is reproducable, which is good. Also notice that the background glow is gone. You can see that we can clean up camera noise by subtracting a dark frame, that is what I did for my star shots.


A sideline here. Here is a dark frame from a different CP950 camera (thanks Dan Lauring). Notice a couple of things! 1) It has the glow in the upper right mentioned before. I think that this is a "warm spot" on the CCD. Maybe there are some hot electronics near there. It could be that the extent of this glow might depend on how long the camera has been on, and how hot the parts have gotten. 2) the wavy line pattern is still there. 3) Dan's frame is not as "bright" as mine. What this means, we don't know. Is his camera just not as noisy? Is the gain set lower? Is it less sensitive to light? We don't know. More dark frame info soon.




Cold and Hot. Dan Lauring suggested testing the effect of temperature on the dark frame. (May 1999)  Here are crops from enhanced dark frames. On the left,. I took a dark frame shot after the camera had been sitting unused all night. On the right, a shot after the camera had been sitting for about 45 minutes turned on in play mode.  There is a little more noise on the "hot" shot, and the rogue pixels are brighter.  It is possible that the effect might be even greater if I had been using the camera in REC mode so those elecronics would warm up also, or if the camera had been on longer. The camera was still cool to the touch when I took the "hot" shots, not warm like it gets after you have been using it for a long time.

I have found that to get the maximum benifit from the dark frame subtraction that you should take a separate dark frame for each shot, under the same conditions. This is what people used to have to do for astronomical CCD cameras. Some of the newer ones have automatic dark frame. It even helps to subtract multiple different dark frames for star shots.


05/19/99 (above) The Alcor/Mizar pair in the handle of the Big Dipper. This was taken through the TC-E2 teleconverter, 8 sec at ISO 320. I subtracted multiple different dark frames to get the background so black. Note the star colors. Also compare to the excerpt from a star chart below... I think that you can see stars down to 8th magnitude.


Rotated to match the picure, roughly.


Overlayed shots. Here I simulated a longer exposure by overlaying four 8 second exposures in PhotoPaint 9. The following operations were used.

1) Took 4 identical shots.

2) Subtracted a dark frame from each

3) Added the shots together in PhotoPaint.

4) Adjusted the tone curve to discard some of the background noise and darken the background.

5) Cropped out the central part and resized for the web.

Most of the little dots are real stars, check your chart!



The Moon. No digital camera web site is complete without a picture of the moon! April 24, 1999. 8" SCT. 1/125 sec at f4 with ISO 100.

I used increased ISO so that I could get a shorter exposure and reduce the shake.  Untouched except to resize for the web. Note the famous "straight wall", follow the black tick marks on the side.

The ability to control the shutter speed is great for astronomical photos. Too bad you can't do it at the higher ISO on the CP950. I used IDO 100 for these shots to get faster shutter speed, but then I had to use spot metering to get the camera to expose properly. This was a real problem trying to take pictures of Mars, that same night. Because the planet was so small, the metering would not work, and it would have been great to have been able to control the shutter speed at the higher ISO, As it was, I was stuck with ISO 80 because the metering would not work. The Mars shots didn't really turn out. I took over 100 shots, but none are really good enough to post. Visually, I could see the polar ice cap, and some surface markings. None of the photos really show anything besides the ice cap. OK OK, here is a shot below. I'm not proud of this.


Mars. You can kind of see the polar cap just to the right of the top. Sort of. Well, you can see the color of the planet anyway.  The detail on Mars is really hard to capture. You can see why the early observers thought that they saw canals. April 24, 1999, 8" SCT.

More Astro 950 Material