11

I would like to take some long-time exposure images from rivers and lakes, but I do not own a grey filter. That results in overexposed images, up to complete white images. Thus two questions: How can I still take those pictures without a grey filter, and how do I calculate the needed time for smooth water without having a grey filter?

  • 1
    The question is different, but the answer is the same: photo.stackexchange.com/questions/11730/… – Michael C Apr 9 '16 at 12:37
  • 3
    I tried the same thing, there is just not much more to do. Get a 2-$ ND filter for testing or use a welder's glass (1 $ in the home depot) and if you think it is your thing, buy a high quality one. – Aganju Apr 9 '16 at 16:01
  • Yeap. Try to invest a cuple of dolars on one or two filters. – Rafael Apr 9 '16 at 19:37
20

You can merge multiple short exposure photos into a single long exposure image. There are a lot of tutorials on the net, for example:

https://www.youtube.com/watch?v=nAuQWfS3pLg

Basically, he opens the sequence of photos in photoshop as layers in a single picture, then "auto-align layers", "convert to smart object" and "stack mode" - "mean".

Image alignment (the most time consuming task) can be avoided if the camera is perfectly stable between shots. If your photos are already aligned, basic stacking can be performed very quickly, even without sophisticated tools, for example this is how the free and open source ImageMagick does that:

convert  photo1.jpg photo2.jpg photo3.jpg -evaluate-sequence mean  result.jpg
  • 1
    Stacking photos could give a similar effect, but it may not look the same as a proper long exposure. – vclaw Apr 9 '16 at 13:26
  • For the best results one should take a large number of pictures, I would say at least 25, otherwise the water surface won't appear to be smoothed out in a similar way as in a long exposure picture. – Count Iblis Apr 9 '16 at 18:41
  • 1
    Actually, averaging over many images may often even give better results than having the same effective exposure time by means of ND filters: For sufficiently man input images, the result is like having a much lower ISO. – Hagen von Eitzen Apr 10 '16 at 11:37
  • 1
    @vclaw The main difference between averaging and a real ND filter is that in the first case, you have gaps between the exposures; it's ok if the movement you are blurring is random (like water, for example) but not ok if it's for example, human movement in a ballet or light streaks --- you'll have gaps. And remember that a ND10, for example, reduces the light 1000 thousand time --- you need the same number of "fast" shots to have the same effect. See blog.patdavid.net/2013/09/… – Rmano Apr 10 '16 at 16:42
10

Assuming you're doing the obvious - setting ISO to the minimum and using the smallest aperture you can - then there's nothing else you can do without an ND filter. They're not that expensive :-)

  • 6
    And go out on a cloudy / overcast day. That can make a big difference to exposure times. – vclaw Apr 9 '16 at 12:05
  • @vclaw: The tests I did where already quite in the morning, in shadows. – arc_lupus Apr 9 '16 at 13:21
1

Wait for the sun to go down...or at least lower in the sky. You need less light so you can decrease the shutter speed, assuming you've already reduced your ISO and closed down your aperture.

0

You can limit the amount of light received by moving a black sock, piece of dark cardboard etc. in front of the lens. If you have a 10 second exposure and repeatedly cover the lens for a total of 8 seconds during this time, the net exposure will be 2 seconds, but the blur effect in the image will be something between a 2 second and a 10 second exposure because you didn't expose for 2 seconds continuously. This will obviously take some practice, but can be applied very creatively in different ways (e.g. to simulate a graduated ND filter by masking only parts of the image etc.)

Alternatively, you can also reduce light with transparent media other than an ND filter, such as black stockings or mesh.

  • If you cover 8 of the 10 seconds up, it would be the same as only exposing for two seconds. – Aganju Apr 9 '16 at 16:00
  • 1
    Using stockings or mesh over the lens is a classic way to take soft-focus photographs, often with interesting effects around any specular highlights. So, sure, it'll decrease the amount of light but it will have other effects, too. – David Richerby Apr 9 '16 at 16:29
  • 2
    @Aganju It's not exactly the same. You're right that the total exposure is 2 seconds, but it's really n exposures at 2/*n* seconds each, so the effect is more like image stacking than it is a single exposure. – Caleb Apr 9 '16 at 16:32
  • 1
    @Aganju It's not quite the same because we're talking about a moving subject. An effective exposure of 2s spread over 10s of real time will show more movement than a "straight" 2s exposure. However, if you cover for e.g. 2s then expose for 0.5s and do that four times, you'll probably end up with rather harsh-looking movement because the scene "jumps" between the parts of the exposure. To get this looking OK, I think you'd have to continuously move the card over the lens, covering and uncovering repeatedly, many times. That would be hard to control but not impossible. – David Richerby Apr 9 '16 at 16:32
0

How can I still take those pictures without a grey filter

One option is to try to DIY a neutral density filter. You won't get the best quality, of course, but you can get some useful ND-like effect by stretching a piece of Mylar film over your lens and securing it with a rubber band or two. Stretch it tight so that there are no wrinkles. If you need a darker filter, use two layers.

how do I calculate the needed time for smooth water without having a grey filter?

The longer the exposure, the more time you have for differences in the water surface to average out and the smoother it will appear. I'm a big fan of calculating things, but the empirical approach definitely wins here -- there are just too many variables and too much complexity to come up with a useful mathematical model.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.