Site Sponsors
Results 1 to 6 of 6

Thread: Poor man's multishot?

  1. #1
    Member
    Join Date
    Apr 2008
    Location
    Lancs, UK
    Posts
    181
    Post Thanks / Like

    Poor man's multishot?

    Reading through the P65+ threads sparked a few synapses, and I thought I'd do the right thing and start another instead of hijacking those like I normally do

    In sensor+ mode (as well as speed) an advantage is meant to be increased DR and IQ over a normal 15MP sensor.
    With a normal back, if you were to take (say) a 4-shot stitch and down-sample it to be the equivalent size of a single shot from the same back, is there an obvious difference between the results? Is this something that is worth doing as a habit when time and subject allow, or is the resulting difference not worth the extra effort?

  2. #2
    Super Duper
    Senior Member

    Join Date
    Nov 2007
    Location
    Royal Oak, MI and Palm Harbor, FL
    Posts
    8,498
    Post Thanks / Like
    Images
    44

    Re: Poor man's multishot?

    Multishots don't work that way. So, it wouldn't be a "poor man's multi-shot."

  3. #3
    Member
    Join Date
    Apr 2008
    Location
    Lancs, UK
    Posts
    181
    Post Thanks / Like

    Re: Poor man's multishot?

    Quote Originally Posted by fotografz View Post
    Multishots don't work that way. So, it wouldn't be a "poor man's multi-shot."
    Whilst you're not going to get results as good as a true multishot back, I wondered if this technique might serve as a half-way house.

  4. #4
    Super Duper
    Senior Member
    carstenw's Avatar
    Join Date
    Nov 2007
    Location
    Berlin, Germany
    Posts
    2,530
    Post Thanks / Like

    Re: Poor man's multishot?

    First of all, 4-shot mode on a multi-shot back moves the sensor around so that every pixel gets a real reading of every colour. The normal pattern is like this:

    RG
    GB

    repeated to cover the whole image. The 4-shot mode moves the R pixel to the first G pixel, makes another shot, to the B pixel, makes another shot, and then to the second G pixel, and makes the last shot. Then full colour information exists for every pixel. This avoids Bayer interpolation and thus a number of artifacts.

    If you do several shots yourself, you can improve the quality of the final result, but each image still goes through de-Bayering, so the artifacts will be present in each image, and adding more images only makes it smaller, it doesn't remove it.

    Note also that the success depends completely on how accurately you could move the back around. You might move the camera in such a way that you end up recording exactly the same pattern on top of where it was before. Taking 4 shots would not result in the same quality as multi-shot, but it might improve the image a little.

    Try it out, and if you can see the difference, and if it seems worth it to you, then I guess it makes sense for you.
    Carsten - Website

  5. #5
    Member
    Join Date
    Apr 2008
    Location
    Lancs, UK
    Posts
    181
    Post Thanks / Like

    Re: Poor man's multishot?

    Quote Originally Posted by carstenw View Post
    Note also that the success depends completely on how accurately you could move the back around. You might move the camera in such a way that you end up recording exactly the same pattern on top of where it was before.
    I think you're misunderstanding me. I don't mean take the same shot 4 times; I mean take a stitched image, of say twice or three times the file size and downsize it.

    Quote Originally Posted by carstenw View Post
    This avoids Bayer interpolation and thus a number of artifacts.

    If you do several shots yourself, you can improve the quality of the final result, but each image still goes through de-Bayering, so the artifacts will be present in each image, and adding more images only makes it smaller, it doesn't remove it.
    Granted. If your large image is 150% bigger, the artifacts are 2/3 the size. The increased number of samples (admittedly interpolated) should give a more accurate colour rendtition, too.

    Quote Originally Posted by carstenw View Post
    Try it out, and if you can see the difference, and if it seems worth it to you, then I guess it makes sense for you.
    I don't have any stitching capability yet, and unfortunately I'm a few thousand miles from my camera kit, and can't try it for some time; hence I thought I'd ask. I'm lazy, but not THAT lazy!

  6. #6
    Super Duper
    Senior Member
    carstenw's Avatar
    Join Date
    Nov 2007
    Location
    Berlin, Germany
    Posts
    2,530
    Post Thanks / Like

    Re: Poor man's multishot?

    Quote Originally Posted by shakeshuck View Post
    I think you're misunderstanding me. I don't mean take the same shot 4 times; I mean take a stitched image, of say twice or three times the file size and downsize it.
    Ah, I see. Interesting. Well, as long as the de-Bayering doesn't have an effect too far outside of the original 2x2 square of samples, this probably has a better chance at reducing artifacts than just attempting a multi-shot workflow.

    I don't have any stitching capability yet, and unfortunately I'm a few thousand miles from my camera kit, and can't try it for some time; hence I thought I'd ask. I'm lazy, but not THAT lazy!
    How does one end up several thousand miles from his camera kit??? I didn't think it possible!

    Anyway, it seems like it would work, but the added work would be considerable, and you are at the mercy of the stitching software for quality.
    Carsten - Website

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •