Does any1 know what it would take to improve Foveon high ISO performance?

Started Aug 31, 2012 | Discussions thread
Aku Ankka
Aku Ankka Contributing Member • Posts: 591
The correct answer: no CDS, thus high read noise

jl123 wrote:

The Foveon really does seem to be an amazing sensor.

Not really. It has excellent resolution for the pixel count, but pretty much all the other properties of the sensor are inferior to similar sized competetive designs.

Of course the million dollar question about this sensor is, how will Sigma be able to improve the higher ISO performance of it?

The key for improving the low light capability and dynamic range over all kinds of situations is to reduce the read noise. The pre-SD1 Foveon sensor had upto 70 e- read noise (under 70e- is an official figure, I just found some old measurements of mine which indicate 60-65e-, depending on the channel). The competition has in the ballpark of 3-4e- read noise (low ISO may be somewhat higher on some cameras, like Canons). The current sensor is likely to have somewhat lower read noise, but it's still way up there an order of magnitude or more above the competiition.

The reason for the very high read noise is the lack of so called correlated double sampling (CDS) - I am not sure if it could be implemented in any way to a design like Foveon of today - a radical new design might be required and regardless there would be side effecs like lower QE and increased complexity of the design.

-- hide signature --


Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow