[milesy]

HI All

any ideas on how i can estimate the amount of time needed for a long exposure.what i am trying to do is recreate some of those photos taken at night that look like they are during the day (im sure you know what i mean)

obviously the in camera metering is pretty useless while its dark, so how can i estimate the time needed (by the way using k-7)

[Nass]

Experience, trial and error. One way you might be able to shortcut this is do some searches on Flickr of the type you want to take and see if there's EXIF data available for those. Also what I quite often do myself is take one at 6400 just to get an idea of the light on a shorter test exposure then bring it down to 100 and adjust the time accordingly.

[Mike Cash]

HI All

any ideas on how i can estimate the amount of time needed for a long exposure.what i am trying to do is recreate some of those photos taken at night that look like they are during the day (im sure you know what i mean)

Like this?

Pentax K20D
Cosina MF 20/3.8

obviously the in camera metering is pretty useless while its dark, so how can i estimate the time needed (by the way using k-7)

Nass has hit the nail on the head. Trial and error will do the trick, but as each shot can take several minutes it is an incredible waste of time and battery to start your guesswork at your shooting ISO. Crank it way up. Also open up your aperture for the guesswork phase. At this point you don't care what the DOF is or whether the shot is in focus or how grainy it comes out. All you want to do is get your exposure time figured out. Take a short shot, review on the rear screen, take a look at the histogram, and try to make a guess how many stops your need to adjust. Usually a couple or three (relatively) short shots are enough to do the trick. Then it is just a matter of doing math in your head to compensate for adjustments to arrive at your shooting ISO and aperture. Just remember that each stop up or down results in a doubling or halving of the shutter speed.

For example, if your test shots show at ISO1600 you need 10 seconds, then 800 would be 20, 400 would be 40, 200 would be 80, and 100 would be 160. If you tested at f4 but want to shoot at f16, then that 160 secs at f4 goes to 320 at f5.6, 640 at f8, 1280 at f11, and 2560 seconds at f16. Of course, it gets a bit ridiculous doing that sort of math, so at some point you may want to do some rounding in your head and switch off to minutes and fractions of minutes. 640 seconds at f8....may as well call it 11 minutes....f11 is 22 minutes, f16 is 44 minutes.

As you can see, the further you go along those lines, the greater the amount of time between each stop of exposure. And there is a correspondingly decreasing need for neat precision in timing your shots. Cutting off a minute or two too soon or too late is not all that great a difference in exposure and isn't worth worrying about if you're talking about something like a 44 minute exposure as in the example above.

Naturally, you will need to have with you some sort of locking cable release (or locking remote) for doing exposures over 30 seconds. It helps tremendously to have a watch with an alarm function. I use the timer on my iPhone. I've had some luck doing night shots up to about 10 minutes long just by counting off the seconds out loud when I forgot to bring something to time it with.

It also helps to have something to do while you're standing around waiting for the time to go by. Or to at least be the type who can appreciate being able to stand there and enjoy the quiet and solitude without a damned thing to do.

[Nesster]

On the other hand, I've taken some photos of Manhattan with a K100D - or a ProgramPlus, or an ES-II - metering straight - remember that the camera doesn't know it is night and wants to see everything as ordinary light.

This for example - 6 sec @ f/5.6 ISO 400 +.7 EV DA 16-45 K100D

[milesy]

thanks Mike and Nesster for your replies

Mike i guess that was what i was trying to get into my head the othe night and i did try taking a photo at 3200 and working out the equivalent time for iso 100 but id didnt seem to work that well, i think mainly becuase the sky was starting to blow out due to city lights, and there was nothing to light the forground (like in your shot above it looks like there was a full moon and a dark sky) ..... ill give it another shot though

Nesster yes your right of course, the camera doesnt know its night, however the meter only will meter up to a certain amount (like to make a 30s exposure) so there are limits in how usefull this is

[tuco]

Nass and Mike covered the technique to use for sure. Like Mike said, the longer the exposure, the less critical your manual timing has to be.

Here I used two exposures to make one shot because I couldn't get a solid-filled ferry when it was sitting still no matter what I tried. In one shot I tried 90% of the time it sitting still and 10% the Ferry moving and still got a ghost effect. I was able to do it with film in one shot.

Anyway the shot was not night but still pretty dark. It was metered for 1 minute. I had to estimate when the ferry took off and get at the distance of light streak within that minute. Well, the Ferry did not depart soon enough for that one minute and I just let the exposure go for around 90 seconds. That was only a 1/2 stop and really didn't hurt much.

I think the misalignment of the Ferry lights was due to the first shot did not have all the cars loaded and it settled down in the water more for the second shot of it moving. The K-7 doesn't turn off DFS in bulb so I took the first 1 minute shot pretty early to make sure it was ready for the second and the camera was on manual with manual focus and I didn't want to adjust it for a shorter first shot time in case I moved it. I was in a precarious position on some rocks and the tripod was not that firmly planted.