An image study for photographers.
Star trails, turning night to day, midnight landscapes. There is amazing image potential for the photographer willing to do their homework and pre-plan setups to take those grand steps into the dark of night.
This photo is lacking. What it shows me is not. I was in the New Mexico highlands. The sun has long set and the light is only from moon and stars in this one hour long exposure. This image is not fully up to my standards and hence may never make it into my Signature Collection. That said there’s something to be learned from it.
I’m trying to develop some baselines and ideas for working what I call ULE’s (Ultra Long Exposures) and I wanted to share some of those here. This image represents the first time I used “only” past experience to determine exposure time, running a single one hour exposure at 5.6, ISO160.
I was on a solid tripod of course and using bulb mode with a cable release. The exposure worked and the detail is amazing. Though in truth it was a stop lighter than I expected. The half moon really lighting land brightly. Is was so much like daylight in fact, that I had to work dial back exposure and work with tone values to bring it down.
Why only f5.6. The more open the aperture more light can be pulled in. I still needed depth of field, which I get with a larger f stop, but I find I’m stopping down for the sake of it sometimes. 5.6 was all I needed based on the near and far infinity. This is something I plan to do another study on soon.
What’s Wrong with this Scene? This image has a subject. A neat one at that. But in my opinion the focal point is not strong enough and that makes it only OK, which is not enough. I think we should note that just because we do something interesting or less common like star trails or long night scenes. Just because you take a long time to make something, does not mean it’s great. Regardless of how challenging the situation is, we still need to take into account the concepts of space, composition, line and tone. These are also topics we’ll be discussing further in my Lights & Shadows workshop.
High ISO Reference Tests: I often use these to get an idea of proper long exposures. Then convert the ISO down for the longer exposure. For example I might run a test at ISO 12,800 for 30 seconds minute and decide it was the right amount of light. Then calculating, I could determine the long exposure for say ISO100 at 64 minutes. This works pretty well and you can have a close idea of what you’re going to get.
More Thoughts: A few things should be noted. Digital is pretty straightforward in terms of exposure time. Film has reciprocity failure and will work differently (I hope to add some film tests in the future). It should also be noted that the downside of ultra long digital exposures is noise and artifacts. This gets worse as the sensor heats up, so the hotter it is outside the more this could be a problem.
So in cooler temperatures it seems we can get away with more. The bad noise and artifacts come from the sensor heating. So the cooler the temperature outside the longer I feel comfortable exposing. Spring weather with 40 degree nights seems to allow exposures well into the 1 hour range. I’m not sure I could pull that off on a 65 degree summer evening.
I have also found that using long exposure noise reduction really helps. Though it doubles the time the camera must remain active, as a dark exposure of equal length must be taken after the image. It does help.
Another approach is to make many shorter exposures. Say thirty 1 minute exposures. Then use special software to blend those. There’s nothing wrong with this idea, but it’s really not an ultra long exposure and I prefer the challenge and beauty of film inspired night exposures in a single frame.
I’m just musing here and sharing experience. Below are some other articles I’ve written that relate to this and will give your further notes. Also share your own experience in the comments. I’ll also be talking about all this further in my up-coming EXposed video workshop.
Things to Consider:
- Ambient Light. From the moon, lights etc (the high ISO test can help with this).
- Temperature. How fast will your sensor heat up and make more noise.
- Length. Too short produce trails too small. Too long can cause more noise.
- Visualization. You still need to plan a great image. It’s best to start in the daylight, planning what you want the image to be.
- Stability. A solid tripod is a must, but also consider wind, moving trees etc.
Favorite successful images and some of my related articles
- 140 Minutes of Night – Arizona – One of my best. Lots of notes.
- Night Watcher – New Mexico – A new one with more notes coming.
- The River Fork – A look at senor noise and heat issues.
A Digital Night Exposure Baseline at f8 – ISO100
I’m trying to set a baseline idea, not a rule. Consider how changes in settings will effects the result. For example, increasing to ISO200 would add 1 stop or light. Changing aperture to f5.6 would do the same and increase light by a full stop. f11 would cut it by a stop. Using different settings you could alter exposure time to keep overall exposure as desired.
These baselines are for actual night scenes where it’s mostly dark and stars are visible. Of course this would all differ with a cityscape or any brightly lit scene. Those have enough light to expose with more normal techniques. Bear in mind these are just starting guidelines based on exposures I’ve made and I’m still experimenting. You can use the high ISO test above to make you own conclusions in each situation as everything can shift depending on ambient light intensity, reflectiveness of foreground elements etc.
- Scenes with little or no moon / ambient light – 1-2 hours.
- Scenes with a small amount of moon or ambient light – 45 min.
- Scenes with bright or full moonlight, but still dark outside – 30 min.
I’ll be back to add updates to this article and refine it as I research more. It’s not a paper yet. Just trying together some ideas and experience 😉
Gavin