Nighttime skies with the V1?
 

This may seem like an odd question to some, but has anyone out there used a V1 to shoot a star-filled night sky? From all the reviews and articles I've read, the V1 seems to have more than decent low-light performance. Living in Alaska, I'd like to try and get some shots of the spectacular skies we have up here during the long winter months- more specifically, the aurora borealis. Also, those skies where there are so many stars, they literally cast a glow on the mountains and surrounding landscape.

I know I'd definitely need one of those polar bear covers, since the best views happen when temps get below -15 degrees F (no distortion from heat rising from the ground), but all other things aside, is the V1 capable of doing something like this, and would it deliver a suitable image?

night time skies
 

I used my FX1 to shoot video of Comet McNaught (having no other camera with me). I used 1/3 shutter speed and 18dB gain, f/1.6 and got quite good images for my purpose. I recorded stars well and good images of the faint tail of the comet. However the video was noisy, although not horribly so. I used software to stack the frames and produce a good quality very low noise still. The V1 isn't quite as sensitive, but the thing is to give it a test. Certainly need a good tripod.

With those shutter settings, I can see how you got the video. Very cool. It's tough to capture stars on a small-form camera, trust me, I've tried. Aside from visual f/x, the only camera that can shoot the night sky well that I'm aware of, is the Sony HDW-F900. See Michael Mann's excellent COLLATERAL and MIAMI VICE, which also shot on the Viper and some 35mm film.

heath

Wouldn't it be easier to achieve this by stop-motion photography using a digital still camera instead?

Much higher resolution and much more control over exposure; low noise; etc.

Then simply cross-dissolve the stills to re-create the changing aurora or other changes in the sky.

Just an idea.

Interesting idea, actually. I might try it with a friend's camera and a good tripod. Only thing is, the earth moves just fast (or slow) enough that after about 30 minutes or so, the stars will change position.

heath

Why not use one of this "tripods" with motor ?

http://www.bhphotovideo.com/c/produc...quatorial.html


volki

This past winter I spent quite a bit of time outside with my digital still camera, taking various shots of the night sky. I would think that still photos of the stars themselves would be so much better than a video image, or even an HD image for that matter. Besides, most of the stars in the sky are so faint that I don't think anything but a dedicated CCD for astrophotography would work.

The aurora is a little different. It can cover a vast majority of the sky, and its motion is incredibly random. I've tried dissolving stills or running some AE filters on a still photo, but it just doesn't cut the mustard. I was able to build a cool 3D model and animate an aurora in 3DS Max, but I'd like to keep things like that limited to stylized projects. The aurora undulates at varying speeds, fades in and out sight, and at some points, it strobes like a flash bulb. It can change color from green to purple, to blue and white. I've seen pictures of a red aurora, but I've not seen it in person. If you haven't seen it, its kind of difficult to properly describe...

The motion of the stars across the sky isn't necessarily a bad thing either, I'd like to explore some potential for various timelapse sequences. That's actually one problem with still photography though. Unless you have a tripod designed for astronomical purposes with a tracking motor, you'll start to notice star trails in as little time as 30 seconds.

Definitely this is the answer to the above mentioned job. The technique also can be combined with a microprocessor controlled revolving support that turns the cam a pre-set degree per shot, creating pan effect in the stop motion sequence.
This technique is often used in BBC's Planet Earth series.
I have seen somewhere the device that gives the pan effect, its around USD1500.

barndoor tracker
 

The cheapest star tracking mount is called a barn door. Essentially 2 short planks joined by a hinge. The lower plank mounts on your tripod head while the camera is mounted on the other. You set a threaded rod through these so that by turning the thread you open the planks. You can set an electric motor and gear box to turn the screw, or you turn it so much every so often. Works quite well for wide sky photographs. Obviously you have to align the hinge parallel to the earth's rotational axis. And you can get fancy with your bits of wood. An example: http://www.astronomyboy.com/barndoor/

Just search "barndoor trackers"

Unless you get a tracking tripod, the overall sensitivity of your camera becomes an issue. Video cameras can actually have fairly good sensitivity compared to still cameras because the still imagers are often broken up into such small pieces to get that high resolution. I don't want to act as if I know much about reciprocity error, but it has been pointed out that extremely long exposures can be a problem for other reasons. Perhaps a slow shutter speed and a bit of gain can get enough exposure to do the job on a video camera? Someone around here tried the noise reduction from neatvideo.com and said it did a good job eliminating gain noise.

It might be interesting to do a long exposure with 15db of gain on the V1 and see if the stars show through after noise reduction is done. It's too bad the V1 only goes down to 1/4 second exposure. I can see the clouds and bright stars at 1/4 exposure, but it's all pretty faint. It might be enough for someone good at color correction and video enhancement to do something about, but a bit more exposure would really help.

Not sure if you're talking about digital still or film still cameras (reciprocity applies only to film). The problem for long exposures on digital devices is thermal noise. The devices collect photons and turn those into electrons, but electrons are also generated thermally at a rate which increases with temperature. While there are a lot of photons arriving then thermal noise isn't an issue, but in low light conditions then signal to noise decreases. So long exposure with high temperature can result in a very poor s/n ratio. So astronomical cameras are cooled (down to 30C below ambient or cooler) to reduce noise. Video cameras can produce a better low light image than an uncooled digital still camera by stacking frames. Software gathers 1000 or more frames, aligns them (to correct atmospheric and tracking problems), adds them together, reduces noise and sharpens. But you have a still image, not a video. Similarly video cameras designed for astronomy are cooled.

I knew there was a reason video cameras also can't do very long exposures effectively. I just couldn't remember the specifics. I do recall reading an article on how to DIY an astronomical camera by attaching a thermal electric cooler (peltier) to the CCD block of a sacrificed consumer camera. I wish I could provide a link, but I figured I don't have the courage to cut up my V1 to try to hot rod the CMOS censors and forgot about the article.

I tried to figure out if frames could be stacked with the V1, but the image is so faint that I'm afraid random noise would start to blur out the stars.

Nevermind! I found the link:

http://popsci.typepad.com/how20blog/...roject__2.html

I think this procedure would be far easier with a single large chip instead of a 3CCD/3CMOS camera. I still don't think I'm going to be the first person with a water-cooled V1. It looks like a fun project, but I hope the screws stay shut on my camera for at least five years.

If you want to try, have a look at http://www.astronomie.be/registax/