More interesting to me than Ctein's theoretical endpoint, though admittedly much more prosaic, is: where is the point at which the buying public will have gotten its fill of higher ISOs (sensitivity) in digital cameras?
I've argued many times that a useful way of looking at the history of photographic technology is that, for any given property, first a "problem" is highlighted, then there's a "race" as various manufacturers vie to see who can outdo the others—encouraged, supported and rewarded at every step by the buying public—and finally there's a stage at which the market loses interest, because a state of sufficiency has been reached. That is (to quote myself), the point is reached at which further improvements no longer inspire additional sales.
This has happened with all sorts of things. Of course, with each parameter, manufacturers do go beyond the sufficient point, and a small speciality market embraces the extremes, but the public in the main has had enough. I've written about all that before.
I speculated a year or two ago that we might be seeing that point reached with the "megapixel wars," and I think that's bearing out. The two 24-MP DSLRs (three, if you want to consider the A850 and A900 as different cameras) haven't gone over very well with buyers, and image sizes from 16 MP to 21 MP don't seem to be impeding sales of the cameras that offer those pixel counts. Personally, I didn't even pay much attention to the pixel count of the last camera I bought. It had enough, that much I knew; actually more than enough, by definition, since it had more than my previous camera and my previous camera had enough. I decided on the newer camera based on other considerations.
I have no doubt that manufacturers will exceed 24 MP in mainstream cameras, and specialty cameras for speciality uses will find their niches. But there are signs that the mainstream camera buyer might already think that 24 MP is overkill.
The next development will probably obliterate the present context: that is, some technical development in sensors will come along that elevates some new method of measurement, some new kind of spec, to a position of central importance, and traditional pixel counts will cease to matter.
We'll get to the sufficient point with low-light sensitivity far sooner than we reach the theoretical technical limit, if history is our friend. In fact, we might even have arrived at the threshhold of that zone with the sensitivity of the D7000 / K-5 / D3s—which is mighty good, happy talk of fireflies and handheld astrophotography aside.
And speaking of the next development that will obliterate the present context: Anyone want to place bets on the end of the lifespan of the Bayer array?
My call is 7–11 years, and, since it has seemed more persistent already than I thought it would be, I'm going to go with the far side of that estimate: I'll say the Bayer array will be essentially history by 2022.
My friend Oren will remember I said that, so no doubt I'll get called on it 11 years from now.
Note: Links in this post may be to our affiliates; sales through affiliate links may benefit this site. More...
Original contents copyright 2011 by Michael C. Johnston and/or the bylined author. All Rights Reserved.