What would the FFF have to do in order to be competative?

Scottelly wrote:

abera wrote:

D Cox wrote:

The FFF should have the same image quality as the Merrill sensor, but with double the area. Simple.

Images from the Merrill are already extremely good, as we all know.

They have good properties and bad properties compared to contemporary alternatives.

The main goodness is lack (or at least reduced) amount of false color artifacts.

Resolution from the image point of view would be relatively competetive with todays cameras if the pixel pitch were to remain the same (as the bottom photodiodes were already made bigger than the top ones for SNR reasons, it's not likely the pixels would get much smaller). But being "relatively competetive" is not enough for market breakthrough. Especially since conventional sensor resoltions are growing - I guess the next generation top models have 100MP or so and 40+ MP will be the normal (to allow for 8K video).

The price for this is reduction of performance in pretty much every other department.

I don't expect images from wide angle vintage lenses adapted to L mount to be any good, because they are already poor on the fp. But I do look forward to seeing what longer lenses such as the 105mm Nikkor can do on Foveon.

If they can improve the QE so that one can get good images at a setting of ISO 800, that would help.

It's not really so much the QE - that's not actually that bad. There's no CFA, thus the top layer will capture photons at pretty good efficiency. The other two ones aren't the greatest performers though. The dead spaces can't really be removed. So this means issues for creating colors (in addition to the poor color separation) and this increases noise.

The read noise is one more component - it's not competetive and likely never will be. Even if they somehow managed to create a proper correlated double sampling they'd still have more photodiodes per pixel and more data to be converted by the ADCs at acceptable rate creating more noise.

Anyhow, I think one can get perfectly fine ISO 800 images already on Foveon - collecting more than twice the signal with a larger sensor would make things better.

I hope FFF sees the light of day just for the sake of it being different and interesting. But even if that happens, I don't think there will be a (Foveon sensor based) successor for it.

What I think would be an interesting version would be a chip with four layers with the bottom one being a very thick capturing mainly IR. It could be made perform quite a bit better in IR than current conventional cameras even when IR converted (as the photodiodes don't go deep enough to the silicon). Though I doubt it's worth the effort from economical point of view, but hey, Foveon is a hobby proyect anyhow the way I see it

Call it what you like, but Sigma's latest lines of cameras with Foveon sensors in them were very competitive in price and low ISO image quality.

Some metrics of low ISO quality are good, not all.

Why do you think high ISO isn't the greatest? Why would the same reason not influence lower ISOs?

In fact, even to this day there is no camera sold new under $1,000 that can compete for image quality at low ISO.

All that offer raw offer potentially better color accuracy.

Also probably all offer larger DR (even at low ISO) and at least as large or larger SNR throughout the DR (though with the caveat of B&W photography - I don't know what the saturation capacity of today's Foveon pixels are for all the layers - there the higher end of DR might be competerive.)

The APS-C Quattros were all sold for well under $1,000.

I wonder if Sigma made a penny of profit from any of them? Even if Foveon is a hobby, it can't lose too much money.