Manufacturers short-changing us, we could gain +1 stop for free

Discussion in 'Digital SLR' started by RichA, Sep 22, 2008.

  1. RichA

    RichA Guest

    After converting a Nikon D100 to IR (removing the IR filter in the
    camera) I realized the mfg's are cheating us by using what they use
    for the IR filter. The filter works by using a cyan tint to remove IR
    light. This works well, but the problem is the cyan-tinted filter
    causes a stop loss when it comes to light coming through. If the
    mfg's used dichroic hot filters (no colour tint in them) they could
    block the IR and save a stop's worth of light, thereby increasing the
    camera's light throughput. This is no doubt done for cost reasons, as
    a tinted filter is much cheaper than a multi-layer dielectric
    filter.
    Problem is, you have to (if you want to use the converted camera for
    normal photography) have to use a cyan filter as the sensor and
    processing are balanced for it's effect on the visible as well as the
    IR spectrum so you can't replace the cyan IR filter with a dichroic.
     
    RichA, Sep 22, 2008
    #1
    1. Advertisements

  2. RichA

    Me Guest

    Is there a new D100 model - or are you talking about a model that was
    made before hysteria about high iso performance became the most talked
    about feature of digital cameras?
     
    Me, Sep 22, 2008
    #2
    1. Advertisements

  3. RichA

    RichA Guest

    That's the one. They still go for $400-$500 on Ebay, I figured it
    would make a good IR camera and it does.
     
    RichA, Sep 22, 2008
    #3
  4. RichA

    RichA Guest

    You can get really good, accurate cutoffs, but the cost is very high.
     
    RichA, Sep 22, 2008
    #4
  5. RichA

    RichA Guest

    Colour response is shifted so much in certain areas that the best you
    can do is balance a few colours, others are way off. It's not
    critical, I bought the camera for IR only, I've got three others for
    normal photography.
     
    RichA, Sep 22, 2008
    #5
  6. RichA

    Scott W Guest

    Dielectric filters have viewing angle problems, not a problem with a
    long lens at f/8, but go to a wide angle lens at f/1.4 and the image
    is going to suffer.

    The loss with the cyan filter is mostly red, which the sensor is verry
    sensitive to so lossing some is not a big problem. If you remove the
    cyan filter you are going to blow out the red pixels way before the
    others in most light.

    Scott
     
    Scott W, Sep 22, 2008
    #6
  7. RichA

    RichA Guest

    Yes, but I'm wondering if the whole Bayer pattern of sensor filtering
    is based on using such a cyan filter? Would the sensor filter pattern
    be completely different if a cyan hot filter wasn't used?
    Based on what it looks like, I'd say the cyan filter is some kind of
    gel sandwiched between two thin plates of clear glass. Cheap to
    make. I'd like to get a look at what Leica used in the M8 and what
    the extra filters needed looked like, to see how deep the blue was in
    them.
    But I've got a Tiffen dielectric hot filter (type that screws on a
    lens) and it makes almost no difference to the image therefore much
    more is being done by the internal filter than just screening out IR.
     
    RichA, Sep 23, 2008
    #7
  8. RichA

    Scott W Guest

    A filter that goes on the front of a lens only deals with a small
    range of angles, at least if the lens is not too wide angle. But a
    filter in front of the sensor will have to deal with a larger range of
    angles, for fast wide lenses.

    People are finding that the add on IR blocking filter that are needed
    in front of lenses on the M8 start to have problems if the lens is
    very wide angle, and I believe Leica is doing some correction in the
    camera firmware to try and reduce this effect.

    The filter in front of the sensor is in fact a dyed glass, both Hoya
    and Schott make them.

    Here are the curves for BG39 and BG40, for green and blue there is
    much less then a stop of loss, and red is normally not a problem for a
    sensor anyway.
    http://www.noao.edu/kpno/filters/sgt/123b_sm.gif

    I have used both for the glass filters and interference for cameras
    and in the end I keep coming back to the glass ones.

    Scott
     
    Scott W, Sep 23, 2008
    #8
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.