Long exposure for digital camera

Discussion in 'Mirrorless Digital Cameras' started by daniel_b., Dec 16, 2004.

  1. Hi All,

    First I'd like to say what a great forum you all have here. I've been
    a "lurker" for quite some time and post just a little bit on some
    other photography boards.

    I am a beginner photographer/hobbyist. I have a Canon Rebel Ti and am
    thinking about upgrading to either a Canon Digital Rebel or 20D.

    My question is this: With the digital cameras coming out now, when
    you do longer night time exposures (20, 30, 60 minutes or longer),
    how is the overall image quality? Is there a significant difference
    between film and digital under these circumstances?

    Many thanks in advance for any guidance you kind folks can give. I
    apologize if this has been asked before. I tried searching and didn't
    find anything after 20 minutes of reading.

    Take care and thanks!
  2. I haven't tried exposures that long yet but I hope to try my hand at astrophotography with my 20D soon as I can counterbalance the weight on my telescope. I understand that long exposures increase the noise in the image over time. The 20D has a long exposure noise filter you can enable for that and I have read some other posts discussing combining multiple shorter exposures or subtracting in photoshop to clear up the image. The Rebel doesn't have such a filter.
  3. http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=007s3Y

    The last post has some promise. At least it's more fodder.
  4. Thank you Andrew and Rob!

    I appreciate your thoughts and the link. If anyone else would care to chime in ... please do so.

    Be good,
  5. Last night and today I did a fairly extensive noise characterization of my personal 'walking around' camera - a Canon 10D. As part of this I took a four hour exposure at room temperature of the inside of my lens cap at ISO 100. I was astonished to find that the noise hadn't yet crept to a third of the way up the histogram. I can't say my other (far more expensive) camera is nearly that good.
    Anyway, regarding quality.... Long exposures with digital tend to be visibly noisy, particularly in dark parts of the photo. The noise appears similar to snow on a television set superimposed on the image. The longer the exposure or the higher the ISO setting used, the more of this noise you will see in the shot.
    I find that film is different, since film tends to remain very smooth and clean in underexposed areas. But the whole frame can take on a kind of weird color cast, especially if you are pointing the camera at the sky for that length of time.
    Both the color cast of film, and the noise of digital, can be attacked using various tools and techniques. For digital, noise reduction software such as noise ninja does well. I've just written a draft article about applying dark and bias frames to digital photos which may be helpful to some. And sometimes you can clip off noise by adjusting levels.
    Anyway, yes, there is a significant difference. They are different mediums, and there are some new things to learn. But most of what you know from film will be transferable, and the learning curve shouldn't be very steep.
  6. In general, the longer the sensor is active (i.e., shutter is open), the more it generates heat. That in turn creates noise in the image. You would want to keep the camera as cool as possible, though not so cold that it freezes the batteries. I think the best thing is to simply experiment and see if you like the results.

    I've tried a couple with lower level digicams and predictably the images were quite noisy to the point of being unacceptable.

    Film does have an advantage in very long exposures provided your camera can keep the shutter open that long (not a problem for manual cameras, of course).

    Several years ago (might have been 10 years or more), a photographer put together a volume of work that comprised photos taken at night. Each shot was generally a six- or seven-hour exposure. The results were not daylight and not night -- very different. But none of the photo displayed any heavy grain. I don't recall any of the technical specs other than exposures were measured in hours and not minutes or seconds.
  7. A thread running right now in the Canon EOS forum

  8. I don't have an example handy but when I first got my D60 I tried some 30 minute exposures. You definitely get the added noise but after about 15 minutes you also (at least on the D60) get a magenta color cast creeping in from the right side of the frame. I often do long exposures with the D60 but keep it to 10 minutes and below to get results I'm happy with.

    I know this doesn't help with with 20d or DRebel. I'm sure the noise is a little better than the D60 but I'd be curious about the magenta cast which I'm assuming is heat related.
  9. I remembered that I had posted about this a while back. Check this out for an example of the magenta cast

  10. I'd be curious about the magenta cast which I'm assuming is heat related.
    What you are referring to is something called amplifier glow. Contra the responses in the other thread you pointed to, it isn't unfathomable or mysterious; it is very common and I've seen it on every digital imaging device I've ever used. The cause is indeed thermal - the higher temperatures on one side of the sensor are caused by off-sensor or off-imaging area readout amplifiers, and their power sources and other support circuitry; there is variation in the exact details depending on the design of the camera and the sensor type. In some cameras, these amplifiers can be (or are) turned off during imaging to help reduce its contribution to dark current.
    I believe that one big difference between the Canon 10D and the Canon D60 was the introduction of techniques to significantly reduce amplifier glow. You may be interested in checking out a comparison between the 10D and 20D that isolates the amplifier glow component of the dark frame (scroll down about halfway to the "amplifier glow" section); and in a comparison between the 10D and Nikon D70 that does the same. As you see, technical and scientific imagers are very interested in this phenomenon.

Share This Page