Metering moonlight exposures

[john w revie]

The story behind this photograph is I have been trying to work out the correct exposure for moonlight shots. The method that I settled for was to take a test shot at f2.8 iso 3200
This gave a base exposure time to work from for the sake of simplicity lets say a 1 second exposure. The photograph was then taken at f11 iso 100. To workout the exposure time I multiplied the base exposure by 32 to compensate for the change in iso then multiply by 4 to compensate for the change in f-stop. This method only works if the photograph is taken at least one and a half hours after sunset or one and a half hours before sunrise, as in those periods the light levels are still changing. If anyone knows of an easier way of doing this and is inclined to share it with the community it would be much appreciated



John

[Colin Southern]

Hi John,

I do a fair amount of this kind of stuff. In essence, what you describe is one easy way - another is to spot-meter with a hand help light meter - identify the brightest part of the scene that you want exposed correctly - and then extend the exposure by two stops because the lightmeter will give you a reading that will expose the brightest part (that you measured) as a mid tone.

This technique works well for dusk / dawn transistions, but light levels are possibly too low outside these periods (haven't tried). For long exposures in these transitional periods you need to allow a "fudge factor" for changing light levels, but generally you don't have to allow as much as you might think, as with dawns most cameras can handle tones 3 to 4 stops above the mid-tone reference (4 with highlight tone priority on for Canon shooters) and for dusk shots it's usually close enough that you can make any adjustments from what's contained in a RAW capture (ie if it under-exposes a bit then "who cares") (there's usually quite a lot of exposure latitude with low-light multi-minute exposures).

The other approach is the "suck it and see" approach - shoot something for a few minutes - see how bad the histogram looks - and re-shoot from there - once you're through the transitional times then the readings don't change - for me I find that it's more of an issue deciding how far you want things like street lights to blow to be able to recover regular mid-tone detail without having to try and dig it out of shadow areas will all of the associated noise.

Biggest dilema I face is that the smaller the aperture, the nicer the star patterns eminating from street lights - but it can mean upwards of a 40 minute exposure - and I just get too bored waiting! (although I usually get set the times and leave the camera to it while I sit in the car and keep warm!).

Does this help?