[deejjjaaaa]

.

The only promising approach to solve this I know about is to replace a sensel's microlens by a trichroic microlens assembly and the underlying detector by three of them. But this approach is closer to the Bayer approach as it focusses on detecting as many photons as possible w/o caring about their property.

how this is close to Bayer approach where you by definition have CFA that filters out a lot of photos based on exactly their properties ?

[falconeye]

how this is close to Bayer approach where you by definition have CFA that filters out a lot of photos based on exactly their properties ?

IMHO, it is because the silicon part of things wouldn't change (except that it needs a pixel pitch of 2.9µm when Bayer needs 5.0µm). You only need to replace the microlens/filter layer by something else (the trichroic microlens layer) which is pure optics/micro mechanics.

The Foveon approach however, requires a complete reimplementation of the photo-electric effect in CMOS.

[deejjjaaaa]

IMHO, it is because the silicon part of things wouldn't change (except that it needs a pixel pitch of 2.9µm when Bayer needs 5.0µm). You only need to replace the microlens/filter layer by something else (the trichroic microlens layer) which is pure optics/micro mechanics.

The Foveon approach however, requires a complete reimplementation of the photo-electric effect in CMOS.

yes, but still it is wrong to call it close ("spiritually") to the solution which was by its core nature based on filtering out the light... it is actually close to the original Foveon which started w/ prism based cameras (their first product was prism-based, three-sensor, one-shot studio camera) and by nature based on separating the light spectrum to absorb all parts (instead of filtering).

So "bayer" approach is to filter out and absorb only what was left, "foveon" approach is split and absorb everything after all... technically if you are talking about replacing CFA layer w/ prism layer it is close to how we make "bayer" sensors, yes... but idea is totally "foveon".

[falconeye]

technically if you are talking about replacing CFA layer w/ prism layer it is close to how we make "bayer" sensors, yes... but idea is totally "foveon".

I totally agree.

In some rare cases however, implementation beats algorithm. This may become such a rare case.

[Digitalis]

the only problem I see with using a light splitting Dichroic prism system Falk is suggesting that if someone wants to use a completely different spectrum of light e.g IR due to the nature of dichroic coatings UV and IR frequencies would be impossible to capture, the contributing factor to why we can presently image scenes in IR light is due to the Law of forbidden transmissions Dichro prisms designed for visible light simply wouldn't work with IR light.

The law of forbidden transmissions at work:

[Pentaxor]

although there is still some further improvements to be done with the sensor and some unresolved issues, I'm pretty much impressed with what I'm seeing with the SD1, especially with regards to studio and landscape/architecture use. I'm pretty much sold on the color resolution and how much detail the sensor could render. I would be a hypocrite to say that it doesn't look great for such uses and could be really nearing MF resolution territory (atleast near the highest average which is around 40MP). 30-35MP just my guess. but still the price kinda killed off it's thunder. although it is somehow cheaper than the actual MF cameras but still quite ridiculous for most people. maybe if Sigma dropped it at $4,000, they could generate some very good sales. at $2,000-$,2,500, Sigma could had created some major turmoil in the camera market or got everyone on board the Sigma boat. I guess, I'm just really disappointed that Sigma got my hopes high with their initial price assessment.