Panasonic's new color splitting method increases sensor efficiency

[Aglet]

here's an interesting story.
Instead of the color filters used on typical Bayer or other patterned image sensors, this is a color-splitting method based on diffraction within the sensor surface.
This could utilize significantly more incoming light to create electrons for readout signal.
Overall increase in effective QE and SNR possible.

IMO, This may be a method better suited to tiny pixels in small sensors rather than larger sensors and larger pixels in higher end cameras. Either way, an interesting and promising development.

http://photorumors.com/2013/02/04/panasonic-to-double-sensors-sensitivity-with-unique-micro-color-splitters/#more-38804

 

[Mt Spokane Photography]

While I see how it works, I'm doubting the claim of better low light performance of a actual camera (which does not exist). The light loss seems to me to merely be moved to a different place. That very tiny diffraction grating is 100% efficient, but light that does not strike it is reflected back and lost while all of the light that strikes a color filter is kept, even if its 50% efficient. How good will the top micro lenses be at focusing all the light on the diffraction grating??

 

[bchernicoff]

I feel like they wouldn't have presented it to a scientific journal if their claims weren't accurate.

I'm always very excited to read about these kinds of breakthroughs.