Video sensibility

[Jairo]

I can estimate whether a film can register stars by checking ASA, aperture, exposition, etc. But how should I translate a video camera's characteristics in these values? If it isn't practical, how else should I check this?

Thanks.

[dwight]

TV cameras have a much worse contrast ratio than film. When I was studying TV tech theory I was told it is around 50:1 which is still standard today.

[chew]

Oct 18, 2010 13:11:40 GMT -4 Jairo said:

I can estimate whether a film can register stars by checking ASA, aperture, exposition, etc. But how should I translate a video camera's characteristics in these values? If it isn't practical, how else should I check this?

Thanks.

How do you do it with film? The other day I took a picture of Polaris with my digital camera manually set to ISO 2000. The properties say focal length 6mm, f/2.8. How do you calculate how long of an exposure is required to get an image of, say, Sirius?

Polaris didn't show up just as I expected it wouldn't.

[dwight]

I do it via trial and error. You just wind up getting a feel for it. You probably can get a listing of recommended settings. I know a group in Hamburg make time exposures, but I cant recall their name.

[Kiwi]

Nov 2, 2010 17:03:11 GMT -4 chew said:

How do you do it with film? The other day I took a picture of Polaris with my digital camera... Polaris didn't show up just as I expected it wouldn't.

20 seconds at f2.8 or f3.5 is a good place to start.

That's the short story. The long one is that there are too many variables to give definite recommendations, so as Dwight says, trial and error are the best teacher, but keep notes of every trial exposure so you know what works and what doesn't.

The best exposure depends on a heap of things:

1. What you want -- just the brightest stars, or those plus the ones that are too dim to see with the naked eye, or nebulae, globular clusters, star trails, other universes... etc.

2. The state of your sky -- dark, clear, misty, cloudy, atmosphere-polluted, light-polluted.

3. Your camera gear -- ISO/ASA setting and maximum aperture of your lens. With film you're into reciprocity failure after about 4 seconds, so a theoretical exposure of ten seconds might actually require 20 or 30 seconds, depending on the film. I don't know whether electronic camera sensors have the same reciprocity problems.

My first attempt at star photography back about the mid-80s was of Scorpius rising above the southeast horizon and looking spectacular, so I set my camera on a tripod and framed the constellation in a Vivitar 70-210mm f3.5 zoom, which wasn't easy to do with few stars being visible in the viewfinder. I was using Fujichrome R100 slide film and guessed that 5 seconds at f5.6 should do the trick. Major fail! I found one very faint dot on the slides -- Antares.

In another attempt I didn't keep notes of what part of the sky I shot or for how long, but ended up with successful star trails. "What would give me two groups of three stars in a row with the centre of one group being red?" I asked someone who knew more about stars than I did. "Two lots of three in a row -- that's got to be Orion," he replied. Bingo!

So TAKE NOTES!

If you try shooting the moon, photographically it is just a rock in the sun, so under clear, dark sky you'll usually get a good exposure by just using the Sunny-16 rule, 1/ISO @ f16, or a little more if the moon is in the crescent phases and/or close to the horizon. Don't bother messing with a meter that is recording 99.99% dark sky and one small sunlit rock.

Other night-time shots to consider are satellite and meteor trails and landscapes taken by moonlight. I have a shot of Orion on its side, setting over the North Island's biggest mountain, Mount Ruapehu, in blue sky, with the few clouds and the snow on the mountain being white, the foreground lit up as if in sunlight, and the pink Orion nebula showing. 30 seconds @ f4, 200 ISO film, 24mm lens.

[gonetoplaid]

Nov 2, 2010 17:03:11 GMT -4 chew said:

Oct 18, 2010 13:11:40 GMT -4 Jairo said:

I can estimate whether a film can register stars by checking ASA, aperture, exposition, etc. But how should I translate a video camera's characteristics in these values? If it isn't practical, how else should I check this?

Thanks.

How do you do it with film? The other day I took a picture of Polaris with my digital camera manually set to ISO 2000. The properties say focal length 6mm, f/2.8. How do you calculate how long of an exposure is required to get an image of, say, Sirius?

Polaris didn't show up just as I expected it wouldn't.

For stars (pinpoint objects) it is aperture which matters, not F/ratio. Thus a 2" diameter lens, wide open (aperture not stopped down), will record stars 4X faster than a 1" diameter lens with its aperture wide open. As you can see, focal length doesn't even enter into the equation.

To photograph Polaris, zoom your camera lens to its maximum focal length. Assuming that you shoot at maximum aperture, this causes the zoom lens to utilize the full aperture of the front focus group at the front of the lens. At ISO 2000, a 5 to 10 second exposure should easily capture Sirius if the lens is at full aperture, zoomed to its longest focal length, and is precisely focused at infinity. Focusing on a distant street lamp will get the lens focused for infinity, given whatever the ambient temperature is. A 5 to 10 second exposure is short enough that you can shoot Sirius with a tripod with very little star trailing (due to the Earth's rotation) showing up.

Video though gets a lot more complicated. Simply assume that even 1/15 second frame rate video shot with a conventional video camera lens is not going to record any stars.

[chew]

Thanks a lot.