I know I promised a column this week on Raw files and how they're not quite what you think, but the confusion that has popped up over DxOMark's measurements of ISO has made me realize that I need to explain what ISO is really all about before I can explain this Raw business.
Photographers have the idea that ISO (whether analog or digital) is a fundamental physical measurement, a way of describing the "photon capture efficiency" of a film or sensor. It's not. Ultimately, what it's really about is what exposure will produce the best-looking photograph for a particular subject luminance, according to the collective judgment of viewers. That is a pretty wobbly criterion.
Even in the film days, this was an arbitrary matter. Those of you with long photographic memories will recall that in the 1950s, negative film speeds (then called ASA) suddenly doubled. The films hadn't changed, but film use had.
Black-and-white negative films get grainier and less sharp with increasing exposure. When large format was king that was not very important, and ASA was biased in favor of producing a long range of shadow detail. As the dominant film formats shrunk, grain and sharpness became more pressing issues. The sweet spot for the best looking photograph shifted towards sacrificing some shadow detail in exchange for getting finer grain and more sharpness.
This was essentially a voting process, not an objective determination of physical fact. Some photographers chose to continue to "over"-expose their film by one stop to get the extra shadow range and they were happy to live with a bit less sharpness and more grain. [Present! —Ed.] As always, personal Exposure Indices (E.I.) count for more than manufacturers' rated speeds.
Even film "shadow speed" is arbitrary. It's pegged at an exposure that produces a density 0.1 density unit (d.u.) above film base plus fog for a certain kind of standard development. Why that development? Why 0.1 d.u. vs. 0.2 d.u. or 0.05 d.u.? We're not talking about unchanging physics here, we're talking about photographers' preferences, even with as rigid a medium as film.
Digital captures are a lot more flexible. There are two different measurements of ISO in the industry and one of them is objective-physical-based.
1. Sensor ISO is based on the amount of light it takes to saturate the sensor to pure white. This is what DxOMark measures and reports. Sometimes that yields counterintuitive values. Double the collection efficiency of your sensor so it only takes half as much light to saturate the sensor, and you've doubled the ISO. That makes sense.
Suppose, though, you just make larger pixels that can hold more photoelectrons. What happens then? Well, it takes more light to saturate the sensor, so the sensor ISO drops even though the sensor still has the same sensitivity to light.
This is one reason why many large format digital backs, with their huge pixels and huge dynamic range (which means a huge exposure range) come in with such low DxOMark ISO values. Their pixels aren't insensitive, but they are designed to be huge light buckets, so it takes a lot of light to fill them.
This measure of ISO, while useful for avoiding clipping and correct for a site that is measuring sensor performance, isn't of prime importance to photographers. It doesn't tell you what exposure will produce the best looking photograph. So, we come to...
2. Camera ISO. There are, god help us, five different industry-standard ways to determine camera ISO. Fortunately, only one applies to any cameras made in the past five years and works for both JPEG and Raw files and sRGB and Adobe color space. It's the one that makes the most intuitive sense to photographers, but it's the least physically-based:
Make a bunch of exposures. Process the Raw files into viewable RGB images and decide which exposure produces the best looking results. Once you know the best exposure for a particular subject luminance, you've got the ISO.
Notice that this doesn't directly connect to what exposure saturates the sensor (sensor ISO). That is just one factor. For example, less exposure (which equates to a higher ISO) leaves you with more highlight headroom before you start getting clipping, but the picture will look noisier. Conversely, more exposure (lower ISO) produces a cleaner photograph but your highlights will block up sooner.
Different camera makers will decide on different camera ISOs for the very same sensor, depending on what they think is the optimum balance for image quality. Almost always, they choose a camera ISO that is somewhat greater than the sensor ISO because blown-out highlights are unpopular.
What we care about
Camera design affects camera ISO, too. The noise in the picture is the result of the entire electronic chain, the amplifiers and all the rest. Different designs produce different amounts of picture noise for the same exposure with the same sensor. That shifts the balance between what's an acceptable level of noise and an acceptable range of highlight detail. Still, the sensor ISO hasn't changed.
And, let's not forget the different camera makers (and cameras) will have different hardware and software to convert the Raw data into viewable RGB photographs. Unlike film ISO, there is no "standard developer" for digital files; each camera maker decides what it considers the best "development."
The bottom line
That's why DxOMark gives you a plot of camera ISO vs. sensor ISO. It doesn't reveal that a camera manufacturer is "fudging" or "lying." It tells you how the two different kinds of ISO, which are legitimately determined in very different ways, compare. That's all. It's not an exposé. It's just technical information. So far as real photography practices go, we care a lot more about camera ISO.
In two weeks I'll tell you how camera ISO ties into Raw and why Raw isn't as fixed into objective physics as you'd think. Next week, though, I've got a more pressing matter to write about. See you then.
Note: Links in this post may be to our affiliates; sales through affiliate links may benefit this site. More...
Original contents copyright 2012 by Michael C. Johnston and/or the bylined author. All Rights Reserved.