ISO 25 on phones: "real"?

Started 2 months ago | Discussions thread
ForumParentFirstPrevious
Flat view
J A C S
J A C S Forum Pro • Posts: 19,836
ISO 25 on phones: "real"?
2

The iPhone XS max is estimated to have 8,980 e- FWC at ISO 25 here. Pixel size = 1.4µm (link), so this makes it approx. 4,580 e- per µm^2.

The Canon R6's figures are: 104,854 e-, 6.56µm, so we get around 2,436 per µm^2.

Nikon Z7II: around 3,100 e- per µm^2.

A7IV R: 2,476 e- per µm^2.

So, the iPhone has a FWC a bit less of 2x higher, per area, than the Canon and Sony, and about 1.5x of that of the Nikon. At ISO 25 vs. 100, one would expect 4x but 2x is still something.

Which bring me to the question: how "real" is ISO 25 on phones? I do not want to get into the lengthy discussions of what ISO is - my question is about FWC per area. If we want a lower sensitivity, we can implement the equivalent of an ND filter somehow but the figures above are in e-, not photons.

Bill's site shows detected NR at ISO 200 or below (ISOs above that look scaled). So another question is how reliable are his figures about the FWC if we cannot measure the photon noise reliably. Now, a good analysis of the noise under the assumption that the NR was a convolution could, in principle, recover the noise before the NR but I doubt that he did that.

This is related to a thread I started in the past - whether ISO 100 (in terms of FWC) just happened to be what one can achieve with today's sensors or it is done by design. If the latter, for smaller sensors, we may want to design it to be much lower at the expense of lowering the high end, I guess, but that would be OK. Is Apple or whoever is making those sensors doing that, at least halfway?

ForumParentFirstPrevious
Flat view
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow