Thread: How about the "new OLPF" color filter... as a 4x5.65 or screw-on?

Reply to Thread
Page 1 of 2 12 LastLast
Results 1 to 10 of 18
  1. #1 How about the "new OLPF" color filter... as a 4x5.65 or screw-on? 
    EDIT: I am only talking about the color filter component of the OLPF. You'd still have an OLPF that removed the moire in front of the sensor. It just wouldn't have the IR-cut / spectral color filtering.

    So the new OLPF filters out IR better (and also filters out a little visible light as a side effect) - and this is what improves the color, right? Combined with different color processing...

    Could RED (or another 3rd party) make a filter with similar spectral characteristics but as a 4x5.65 or screw-on filter?

    Then folks could change between "low light but IR contamination" and "good colors but less sensitivity" by plopping a filter on the front of the camera without having to perform mid-shoot OLPF surgery. We could have a "clear OLPF" on the camera sensor and just change the spectral sensitivity characteristics as needed for the shot.

    Anyone with greater light physics knowledge than me think this is impossible (I like to learn so would love a technical explanation)?

    For the record, I PERSONALLY like the new OLPF much better and think that RED should just go with the new OLPF... and focus maximum effort on an improved Dragon sensor with better signal to noise ratio and a design that doesn't need lots of black-shading work.

    But I know that my opinion is not everything so I thought I'd throw this alternate thought out there. Crazy?

    Bruce Allen
    www.boacinema.com
    Reply With Quote  
     

  2. #2  
    Senior Member
    Join Date
    Nov 2008
    Location
    Dallas, TX
    Posts
    2,788
    Interesting....
    JAKE WILGANOWSKI
    Director of Photography / Filmmaker
    CINE-AUTOMATIC.COM
    Reply With Quote  
     

  3. #3  
    Agree, if it's possible to split it up I would rather have the olpf2 part of the equation infront of the lens then behind, just like an IR or ND filter. But I assume it's not that easy...
    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se
    Flame / VFX / Motion capture / Monstro
    Reply With Quote  
     

  4. #4  
    Senior Member Bob Gundu's Avatar
    Join Date
    Mar 2011
    Location
    Toronto, ON Canada
    Posts
    9,756
    Doesn't the OLPF have to be as close as possible to the sensor?
    ___________________________

    VFX, Cinematographer, Photographer
    10 frame handles
    Vimeo
    Instagram
    Reply With Quote  
     

  5. #5  
    Senior Member
    Join Date
    Sep 2007
    Location
    NYC
    Posts
    817
    Quote Originally Posted by Bob Gundu View Post
    Doesn't the OLPF have to be as close as possible to the sensor?
    That is correct, the OLPF has to be specifically positioned relative to the sensor. It has a pattern that corresponds to the pixel array of the sensor that eliminates aliasing/moire patterns.

    Optical elements of any kind between the OLPF and the sensor will interfere with the function of the OLPF.
    - Scarlet 996 -
    Reply With Quote  
     

  6. #6  
    Quote Originally Posted by Todd T View Post
    That is correct, the OLPF has to be specifically positioned relative to the sensor. It has a pattern that corresponds to the pixel array of the sensor that eliminates aliasing/moire patterns.

    Optical elements of any kind between the OLPF and the sensor will interfere with the function of the OLPF.
    Yes but the IR cut, I assume could be any where in the chain. So old olpf behind lens infront of sensor and then the difference in between olpf1 and two infront of lens in shape of a normal filter.....

    Don't know if thats possible but as I understand thats what Bruce asks for. If that would work I think that would be excellent.
    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se
    Flame / VFX / Motion capture / Monstro
    Reply With Quote  
     

  7. #7  
    Yep. that's what I'm asking for. Thank you, Björn!

    The OLPF currently does multiple things: eliminate aliasing / moire AND filter IR (and some visible light, obvious because... well, it's blue if you look at it!).

    You'd still have either a v1 OLPF or a "don't filter color OLPF" (eg it would still eliminate a little aliasing / moire) on the sensor.

    Then add a filter in front of your lens that does the IR cut. This is obviously possible to some degree, if you look at all of the different IR-cut filters out there.

    The question is: are RED doing something more than that, eg using a very expensive / fancy piece of glass with extra properties that would be prohibitively expensive to get in a large size... or something about the way it works means it can't work in front of its lens.

    If anyone has a good spectrometer and some courage, I'd love to see what happens if you:
    1. point a spectro at a light
    2. point a spectro at a light, with an OLPF held in front

    Oh yeah... and would it work for MX too? Although I think some of Dragon's better color comes from less crosstalk in the color filters of the Bayer filter itself, which can't be helped.

    Bruce Allen
    www.boacinema.com
    Reply With Quote  
     

  8. #8  
    And just how would the spectral color filtering become mapped to the sensor through a vast array of arbitrary optics and low-precision lens and filter mounts?

    As for IR cut, if it were removed unto external filtration, then we would be right back where we were with the MX and especially the old M sensor. Too many people not understanding or compensating for it, complaining that "such and such camera doesn't need IR filtration, why does RED?"

    The OLPF is tuned to work in conjunction with the photo sites or pixels of the sensor and must be placed at a precisely engineered distance to create the proper effect. This is not anything that can be done through an optical pathway that is amorphous or capable of endless permutations and variations.
    - Jeff Kilgroe
    - Applied Visual Technologies, LLC | RojoMojo
    - Just me and my 8K Monstro VV kicking ass.
    Reply With Quote  
     

  9. #9  
    REDuser Sponsor Gunleik Groven's Avatar
    Join Date
    Dec 2006
    Location
    Norway
    Posts
    13,342
    Quote Originally Posted by Jeff Kilgroe View Post
    And just how would the spectral color filtering become mapped to the sensor through a vast array of arbitrary optics and low-precision lens and filter mounts?

    As for IR cut, if it were removed unto external filtration, then we would be right back where we were with the MX and especially the old M sensor. Too many people not understanding or compensating for it, complaining that "such and such camera doesn't need IR filtration, why does RED?"
    This...
    Life is good. So is RED...


    http://quine.no
    Reply With Quote  
     

  10. #10  
    Quote Originally Posted by Jeff Kilgroe View Post
    And just how would the spectral color filtering become mapped to the sensor through a vast array of arbitrary optics and low-precision lens and filter mounts?
    I agree... but if you put a 80A filter in front of a lens, it generally works the same, regardless of lens.

    RE: IR cut, certain filters have a vignette effect, others do not so much.

    Which is why I ask if anyone has pointed a PR-650 or whatever at the OLPF and seen what it is actually doing spectrally? And what it's like at different incident angles?

    Quote Originally Posted by Jeff Kilgroe View Post
    As for IR cut, if it were removed unto external filtration, then we would be right back where we were with the MX and especially the old M sensor. Too many people not understanding or compensating for it, complaining that "such and such camera doesn't need IR filtration, why does RED?"
    Very true - you see that all the time when people use the Blackmagic cameras. And I do keep criticizing RED for being over-complex, difficult to specify on a rental, etc. Very good point. But the alternative (swappable OLPFs where half the market has v1 and half has v2) is also messy. People will be saying "why don't these two RED DRAGONS match?!" when the A cam is OLPF v2 and the B cam is OLPF v1.

    Quote Originally Posted by Jeff Kilgroe View Post
    The OLPF is tuned to work in conjunction with the photo sites or pixels of the sensor and must be placed at a precisely engineered distance to create the proper effect. This is not anything that can be done through an optical pathway that is amorphous or capable of endless permutations and variations.
    I agree of course... except for the color filtering. I would love to understand why order or distance is so important there - if you cut out certain wavelengths and it's not angle-of-incidence critical, you cut them out, right? Regardless of whether they are before or after the lens. The lens may cut out its own wavelengths too, but I'm not sure why the effect would change depending on the order?

    Anyway, thank you for responding and listening as always.

    Bruce Allen
    www.boacinema.com
    Reply With Quote  
     

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