Wallen wrote:
It seems many readers fail to see the gist of the question. Many argue of just shooting it whatever the consequence while some speak of their very good & capable equipment. It is not about missing a shot or specific equipment/set-up. The question is about a standard to which we can actually compare capabilities.
As an example, we start with a value of 100lux and we shoot a test panel at 2 meters with a lens of t-stop value 2. Let us assume this value as ISO 100. Then we lessen the lux by 10 and shoot again. At what point (lux) will the test camera have 50% noise compared to the clean image. Then that should be the standard rated optimal limit. Beyond that point would be called extended ISO.
Then other cameras will be set up the same and their output rated as per the standard done above. Hence if all cameras will be tested and rated this way, all ISO values stated in cameras will all be comparable.
The main thing here is that ISO is supposed to be a standard. The truth is it is not. every manufacturer goes their own way. If it is a standard being followed, all cameras will have the same image quality when compared at any ISO setting-within their optimal limit as they are rated to their "ACTUAL WORKING ISO".
It seems many readers fail to see the gist of the ... (
show quote)
Film speed isn't a physical quantity derived from science, like velocity, mass, or temperature.
It has connections with two engineering concepts: sensitivity and gain. But it's really just a
convenience number, like shoe size or dress size. If you were a size 10 shoe, the question
"10 what?" has no answer.
Obviously the inside of a shoe is an irregular shape, and it would take a lot of measurements
to adequately describe it. But we need one number (sometimes augmented by a width).
"ISO 100" seems so real, it's hard to remember that the term has no meaning except
with respect to some particular standard, which sets forth the method used to measure it
and how the number is dervived from the test data. And we forget that ISO is actually
an acronym for a standards organization.
The definition of "ISO 100" for film and digital are so different, that they are basically
two completely different measures. For film, it is baesd on the Characteristic Curve
(also called Hurter-Driffield Curve: negative density plotted against log of exposure,
all this assuming some "standard" development). But for digital, there is no negative,
no density and no development.
In both film and digital, speed atttempts to reduce a empirically dervived curve to a single
number. Obviously, this is somewhat arbitrary. Moreover, the standards have changed many
times over the years, and so have the standardsd bodies:
ASA -- American Standards Association (later renamed ANSI),
NAPM --- National Association of Photographic Manufacuters
ISO -- International Standards Organization
B&W Negative film:
US:
1943 ASA Z38.2.1
1946 -- revised
1947 -- revised
1954 -- superseded by ASA PH2.5-1954
1960 -- superseded by ASA PH2.5-1960
1979 -- revised
1986 -- adopated in the US as NAPM IT2.5-1986
1993 -- revised
International:
1974 ISO 6-1974
Both:
1993 ISO 6-1993
Before 2006, there was no speed standard for digital cameras: manufactuers each came
up with it's own way of arriving at "ISO" numbers (which had nothing to do with the
International Standards Organization).
Then came ISO 12232:2006 "Photography -- Digital still cameras -- Determination of exposure
index, ISO speed ratings, standard output sensitivity, and recommended exposure index"
This was superseded by ISO 12232:2019 -- the current standard.
Unfortunately, the text of ISO standards have to be purchased and are quite expensive.
So I have to rely on secondary sources.
According to Wikipedia:
"The ISO standard ISO 12232:2006[63] gives digital still camera manufacturers a choice of five different
techniques for determining the exposure index rating at each sensitivity setting provided by a particular
camera model. [ . . . ] Depending on the technique selected, the exposure index rating can depend on the
sensor sensitivity, the sensor noise, and the appearance of the resulting image. The standard specifies the
measurement of light sensitivity of the entire digital camera system and not of individual components such
as digital sensors, although Kodak has reported[65] using a variation to characterize the sensitivity of two
of their sensors in 2001. "
So ISO 12232:2006 is a good example of a "non-standardizing standard": the standard just codifies
the lack of standardization. (Something similar was done in the ANSI standard for light meter
calibration--which is why light meters from different manufactuer's can give different readings,
even when correctly calibrated.)
In fact, manfacturers generally don't want it to be easy to compare their cameras with those of
their compeitor, or to test advertising claims.
Worse, there is no way to know whether a particular manufacture is or isn't following the standard
the standard on a particular camera, since compliance is voluntary, and there is no independent
laboratory testing of cameras.
Finally, there are two engineering notions that may be relevant:
Generally in electronics, sensitivity is defined as the smallest amplitude input that will produce
a measurable output. With a camera we are not just concerned with very weak signals, but with
the entire range of luminances that may be in the frame.
Noise is generally expressed as a signal-to-noise ratio. For a camera, the basic idea is that a given
pixel represents eithe signal (the image) or noise. Exactly how to figure it is another question.