The Sigma site does say 12 bit and it's supposed to be the same sensor. I got my number from the LL article.
On Wed, Sep 5, 2012 at 11:02 AM, Toralf Lund <tor...@toralf.net> wrote: > On 9/5/12 8:54, John Francis wrote: >> >> On Tue, Sep 04, 2012 at 10:26:07PM -0400, John Sessoms wrote: >>> >>> I really like the concept behind the Foveon sensor - every >>> photo-site captures all 3 colors like film did& no need for the >>> >>> anti-aliasing, but what's the deal with the poor image quality above >>> ISO 400? >> >> Physics. If you don't have separate sensor sites for the three colo(u)rs, >> then the photons for two of the colours have to pass through at least one >> of the layers that contain the sensors for the other colours. >> >> To oversimplify the situation, you can't create a sensor that captures all >> the blue photons, and yet lets 100% of the red or green photons pass >> through. >> As a result, the second and third layers don't receive 100% of the photons >> that they would respond to if they were the frontmost layer. This leads to >> a degradation in sensitivity and in the signal-to-noise ratio. > > Quite. Like I sort of said elsewhere, though, it seems to me that you could > pass through a lot less than 100% and still be better off than you are when > filtering at every site. > > Also, maybe there is actually more noise, too, and not just less signal - as > a result of more complex read-out circuitry or whatever... > > - Toralf > >> > > > -- > PDML Pentax-Discuss Mail List > PDML@pdml.net > http://pdml.net/mailman/listinfo/pdml_pdml.net > to UNSUBSCRIBE from the PDML, please visit the link directly above and > follow the directions. -- Steve Desjardins -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.