I Found the Light! (was it lost?)
I have heard so many photographers say, “there’s not enough light here for a shot.” WRONG! Even on the darkest of nights light is everywhere. Now is there enough light for a short exposure of 1/60th second or less? (That is about the limit of a hand held shot even with VR (vibration reduction) lenses give or take a bit.) But what about longer exposures?
In my last post I showed you some longer than average exposures of ¼ of a second and a full 1 second. But what if you want to shoot in the dark? Well for starters, you have to try to get into 100% darkness and that will require some effort as you will see below. In the first shot, I could barely see my hand in front of my face, but there was light there!
Hit the link to see how I found the light! (It was behind the sofa the whole time!)
This shot I took at about midnight from a remote shoreline just off the Sassafras River in Maryland (USA). The camera was on a tripod of course and set to ISO 100, focal length of 18mm, for 83 seconds at F/3.5. There were no lights on in the little cove and no lights near us and I really couldn’t see any reflections on the water or even a hint of a tree line. However, as you can see there was light coming from the horizon, lots of it! Enough to give the sky a wonderful warm glow (ain’t pollution grand) and fantastic reflections off the smooth water. Now even if the water was slightly rippled at that long of an exposure it would smooth out nicely. This image is not Photoshopped or manipulated at all. This is 100% straight out of the camera.
In the next shot I took a little longer to find the light, but I could already see it so I knew it was there.
This shot is of the Vermont night sky taken from my sister’s deck (Thank you Wendy!). The camera was on a tripod as well but the setup was a little different. ISO 400 this time, focal length of 18mm again and still at f/3.5 but this one was for a full 1826 seconds (that’s 30 minutes and 26 seconds). If you notice the bluish vignetting at the corners, that is the sensor warming a bit from such a long exposure and using just battery power. Which I might add, 30 minutes is about as long as you get out of a D200 battery. I might have gotten about 5 more minutes out of it but I couldn’t have re-shot with that battery. I did Photoshop this just a bit to turn the corners blue from there original purple. The lines are the wonderful star trails you hear about and that is what I was going for. I only wish I could have aimed a bit more to camera right so I could have caught the North Star not rotating. Perhaps this year’s vacation.
You can see these and more on my site at www.magickmichael.com anytime.
Until next time, Happy Shooting!
Gorgeous pictures Michael! Let me ask you something…. a question I’ve been wondering about. And in reading this article it reminded me of it again. Does digital also follow the law of reciprocity the way film did? What say thee?
Such difficult questions first thing in the morning… This is not going to be a short answer.
Yes and no.
As a digital sensor has a different sensitivity than film, so is the digital processing different from the emulsion process. Digital sensors are made to emulate the emulsion process as you know. However the idea that you shoot in different ways with film made for print compared to film made for transparency or slide also applies to a digital sensor. Sensors are often made to emulate the transparency process. This is why a shooter that shot mostly slide film usually had an easier time moving to digital than a person shooting for print.
Keeping this in mind, Digital has a similar but different reciprocity than film. At the same time it is based on the same reciprocity law. The difference is mainly that you are emulating a process and therefore trying to make one light sensitive material react like another. This happens within a range. Outside that range your digital algorithm sticks to a constant where Film varies based on film type, and exposure factors.
Because digital is exactly that, digital, I believe that the Schwarzschild effect (reciprocity law failure) is actually lessened. I don’t have any hard data to back this up, it is just my experience in computers and photography combined that cause me to believe this.
Reciprocity in film is accurate within a range of exposure and beyond that range is where the Schwarzschild effect comes into account. Because film is not a mathematical medium the effect will vary. Even the age of film can have an effect. However with digital sensors you can only emulate a certain specific film reaction via mathematical algorithm. You can’t account for age of the film, exposure to heat and humidity, etc.
So in essence the reciprocity is more constant. So does it follow the same law of reciprocity? It follows the same theory of the law but the physics are different so the factors involved change as well.
How’s that for a confusing answer?
What I want to know is what compelled you to find an area of total darkness, fumble around in in it with a bunch of expensive camera gear, to take a photo you couldn’t see!
Call me crazy. I had seen extreme long exposure shots and I felt like trying some. The darker it was, the longer the exposure needed to be. It was a challenge. Why climb the face of a mountain if there is a road? The challenge.
What you see is an artifact caused by IR light. Note that there is no cyan vignetting in the lower left corner because the sky (IR light source) is blocked by the tree.
see:
http://nemeng.com/leica/004f.shtml (scroll about half-way down)
http://www.ultrasomething.com/photography/2009/06/the-m8ing-ritual-part-3/ ( about 1/4 of the way down)
The IR filter before the sensor causes preferential vignetting of red light. In Bayer sensors which manifests itself as a cyan cast. The lens itself may also contribute depending on the glass’s IR transmission characteristics Red light vignetting happens at all focal lengths, but it is strongest with wide-angle lenses. Every photo you take with your 18mm lens had this artifact. Because the exposure is so long and the visible light (signal you want) is so low, the IR artifact is very strong in the example you show. The signal-to-artifact ratio is much higher than exposes that require only 10 to 1/1000 seconds.
There are two remedies. Use film. The Nikon F3 body light meter has a fantastic reputation for night-sky photography. The F3′s AE will keep the shutter open for for 8 hours or more. The other remedy is to create a Photoshop mask that subtracts the cyan artifact for each lens you use. There are several plug-ins designed for the Leica M8 digital cameras (this camera has a severe problem with wide-angle lenses). A Google search will lead you to more information about these plug-ins.
I think that everyone has provided some good answers. Having never tried long-term exposures when I owned a D200, I cannot speak exactly to the cause. Yes, I have read that sensors can cause the effect you have on your 30-minute exposure. The suggestion that the artifact is caused by IR light may have some merit. But why, then, would it only be occurring at the corners of the sensor?
I hate to say it, but you need to get rid of the D200 and get a D300. For one thing, the battery can last much longer. I have taken two-hour exposures on one battery and still have some juice left. And, with the long exposure NR set in the camera, you don’t get too much noise at that length of an exposure. Also, I would recommend against ISO 400 for long term exposures. I have done them at ISO 100 – the key is a wide open aperture. But make sure that you do not have any filter, even a UV filter, on the lens, or else it will create an effect in the center of the image; particularly if shooting the aurora borealis.
Another way to get good star trails shots without running into long term exposure issues is to set your intervalometer to take one 10 second exposure every fifteen seconds, ISO depending on whether full moon or not, and set the duration for as long as you want. Then you can have your camera taking snap shots over, say, a four hour period, then sandwich all the images together in one image, and, voila, you have a four hour star trails photo without the inherent problems with long term exposures.
I hope this helps. For some of my work in this area, visit my blog and click the “Nighttime” category.
Carl,
The cyan cast caused by low-frequency (red) light vignetting only occurs at the corners for the same reason visible light only looses intensity at the corners when a lens exhibits vignetting.
Wide angle lenses, in general, vignette more than longer focal lengths. The low-frequency (red) light vignettes more than the higher frequencies because of the characteristics of the glass (a weak IR filter) and/or the because of the IR filter in front of the sensor. This loss of low frequency (red) light intensity results in a cyan color shift during the Bayer filter de-mosaicing process. This is a well understood phenomenon due to all the problems Leica M8 camera users experienced using wide-angle lenses with IR filters. Read the links in my post for more information.
Michael … I see from your shots the value of going after such shots (especially on that top one; gorgeous warm tones!) when your target/subject is far away; you can use the infinity focus setting. This really isn’t possible, though, for any close-range subject where you have no light to help with focusing, though, right? Or am I missing something … or looking foolish for asking?
Joe,
If you are in need of focusing in a situation like this you can and should probably be in manual mode. At that long of an exposure I am not sure what you are going to be shooting at a close distance but I am sure it can be done. You can also use a flashlight to generate enough contrast to actually have something to focus on too.
PS – No such thing as a foolish question if you don’t know the answer!
Close focus situations in low light do have a non-flashlight solution as well, depending on the clubs in your bag. Even low-light auto-focus is possible with some camera/gear combinations such as Canon’s ST-E2 remote speedlite transmitter. Using it without a flash for it to trigger still generates a good strong infra-red focus assist and can allow AF in total darkness. Just my .02
http://www.luminous-landscape.com/reviews/st-e2.shtml