Monday, February 25, 2008

Point of sufficiency

Mike Johnston writes another interesting article, this one about historical and future developments of various parameters of digital cameras.

I particularly like his "point of sufficiency" argument. When cameras hit a point where almost nobody in the market can find any use for higher image quality or sensor sensitivity, then the sensors will become smaller and cheaper. And so small sensors may not seem like the dead end they seem to some in 2008.

It's not unlikely that a thumbnail-sized sensor in a few years may give the same quality as today's sensor in a Hasselblad costing $30,000 and weighing three kilos. And when a pocket camera can make pin-sharp prints at super-sized posters, what will happen to the market which funded research and production of bigger cameras?

6 comments:

Jon said...

It is always the glass that makes the difference. That article well might be true, and for many many years film was at its zenith, but that did not stop the high end cameras from being made. I personally bought a tiny pentax which was a SLR and I forget the cassette that you dropped into it,and it took great photos, but nothing like those that a full size camera took, and it all comes down to the glass. Jon Barry

Eolake Stobblehouse said...

That would be the Pentax 110.
Which used 110 casettes. They had a film frame which was just way too small. You couldn't even make an 8x10 print without big grain.

Today's lenses are amazing. For example the lens on my pocket Canon or indeed the Fuji. At low ISO settings, they make big prints scarcely distinguishable from those made with my biggest cameras.

Anonymous said...

I agree with most of what is being said on smaller sensors technology, but the real problem lies with depth of field...

From the same point of view and to keep the same angle of vision, you must divide the focal length by 1.6 when moving from FF to APS-C format.

Then, to keep the same depth of field, you must divide the aperture by 1.6 when moving to APS-C (-1 1/3 EV).

Here are a few examples for FF to APS-C lens equivalences :
17mm f/4 -> 10mm f/2.5
24mm f/2.8 -> 15mm f/1.8
50mm f/1.4 -> 30mm f/1
100mm f/2 -> 60mm f/1.2
135mm f/2 -> 85mm f/1.2
200mm f/2.8 -> 125mm f/1.8
300mm f/4 -> 180mm f/2.5
400mm f/5.6 -> 250mm f/3.5

The real difference is of course on the shallow depth of field side :
- For tele-lenses, you get the same result for a similar weight.
- On the wide angle side, one is generally not looking after shallower depth of field anyway.
- But between 50mm to 100mm, it doesn't do it for the APS-C format... And it is even worse with smaller sensors !

So if your are not too keen on shallower depth of field, then smaller sensors are the way to go... On the other hand, nice blurry backgrounds are for the full frame or even better medium format guys !

Eolake Stobblehouse said...

True, there is that.

I like a deep DoF, most of the time. But for portraits and such, a very narrow one is wonderful.

Anonymous said...

"point of sufficiency"

What a horrible concept. Humans are not wired for this kind of thinking.

Mike Johnston needs to go listen to the Yello album You Gotta Say Yes to Another Excess.

Anonymous said...

"What a horrible concept. Humans are not wired for this kind of thinking."

When you reach a point where no tangible advantage can be gained by increasing specification X, the discussion won't die, you are right.

But that discussion will rapidly narrow down to a small circle, the so aptly named measurebators. Just like the so-called "audiophiles" who can hear electrons moving in their speaker cables still argue on the merits of right vs. left twisted cable stranding.

Most of us will have grown tired of the then pointless arguments and moved on, and improvements on specification X will stop reflecting in increased sales. That's the point of sufficiency, and it has little to do with a conscious decision, imo.