There was some discussion in "Post Your Photography" about why digital camera ISO range usually doesn't include lower numbers, like ISO 25.
So, here's a bit about speed ratings.
Shortly after the beginning of practical photography (daguerreotypes, tintypes, cyanotypes), which required quite long exposure times and produced one-off photos, silver halide negative/positive processes became common, and facilitated by retail sales of dry plates and roll film. It was then necessary to supply film of standardized sensitivity so that usable negatives would result from fixed setting snapshot cameras under sunny conditions or from using an adjustable camera under variable conditions.
It was always desirable to obtain the highest speed rating possible, so that moving objects would not be blurred, camera shake would not be a problem, and great enough depth of field could be obtained for acceptably sharp images to be made with fixed focus lenses, all while having enough exposure that shadow details could be seen in a print. So, the question became, how can we determine the minimum exposure necessary to meet these criteria. (With standardized development also, of course.)
The optical societies, heavily influenced by Kodak scientists, determined the minimum negative density (or slope of the density = shadow detail contrast) that should be reached to produce a negative that yielded a quality print. As the desire was to specify the minimum required exposure, there was little concern with highlight overexposure latitude, as the exposure being sought would leave all the latitude on the bright end.
The next step was to determine what fixed exposure settings would produce the highest percentage of desirable negatives in daylight. To do this, research (again, heavily supported by Kodak) was done to measure with a spot meter the darkest and brightest parts of hundreds of outdoor scenes, and to distill the data down to a fixed number that would result in 95% of shots having adequate exposure in the shadows. The speed rating set this way meant that 95% of outdoor snapshots would result in excellent negatives. Sophisticated photographers with an adjustable camera could increase this percentage especially for shots not taken in sunlight; could determine if their subject was likely to fool the meter (snowscapes, people on the beach, black cats on dark backgrounds) and still use standardized processing. Really sophisticated photographers (Ansel Adams) could just ignore all this and make their own determination of how to expose for the shadows and develop for the highlights.
A part of these investigations was the determination of the average reflection of the hundreds of scenes (about 12.5 percent) so that average-reading reflected light exposure meters could be calibrated (with the smart photographer compensating for meter-fooling bright or dark scenes) (Ansel Adams again went his own way and insisted on using a printer's medium gray of 18% as an average reference, coercing Kodak to publish 18% gray exposure reference cards with a warning to increase meter indicated exposure by 1/2 stop.)
By the way, around 1950 or 1951, the film makers determined that a higher percentage of negatives had acceptable shadow detail than they thought, and the ISO ratings of negative films were doubled (increased by one stop) simply by changing the number on the box - no change in film or processing whatsoever. All the non-adjustable cameras out there got the same density negatives they always did, but photographers with adjustable cameras not got better motion stopping or depth of field.
Color negative film followed the same general rating process of exposing for acceptable shadow density.
A note here about "disposable" cameras. These usually had/have aperture and shutter settings that corresponded to ISO 100, but the film in them was usually actually ISO 800. This meant that 3 stops of its overexposure latitude was used in bright sunlight, and the exposure was "correct" (minimum shadow detail for a good print) in open shade.
Color reversal (slide) film ratings had to be determined differently. In negative/positve processes, the negative determines shadow compression and the print determines highlight compression. Reversal film has both compressions built in - the entire "S curve" is determined in the original piece of film and therefore the great overexposure latitude of a negative is not available. Therefore, transparency film ISO rating is determined by the mid tones, trying different ratings with experts judging what ISO produces the greatest number of good pictures.
Now to digital cameras:
Digital sensors have photosites that accumulate charge proportional to the amount of light that hits them during exposure. They have a maximum level above which they simply won't accumulate any more - highlight clipping. So, the camera maker determines what level of charge will be middle gray (~level 127 in a jpg file), whether this is generally 12.5% of diffuse white or 18% or some other number, and therefore how much overexposure latitude is available in the raw image (and highlight recovery in post processing). Digital sensors do not have a toe (loss of contrast in the shadows) so that does not give any clue to setting the ISO rating using the method for film. They do, however have noise that is more visible in the shadows than the highlights, so that prevents the manufacturer from using unlimited overexposure latitude and unlimited ISO.
Summary for digital camera ISO: manufacturer picks the maximum charge level (lowest possible ISO) that will correspond to middle gray in a rendered image. Higher ISO settings simply declare a lower charge to be mid gray, and then either analog amplify the signal or multiply its digital value, or (often) both.
Since both prints and monitors have a certain limited contrast range, digital processes always apply an S-curve that mimics film's gradual shadow and highlight compression somewhere along the way. The S-curve makes much better looking pictures than if the shadows or highlights suddenly clipped at a certain level. Some digital processing experts have provided linear camera profiles for use in Lightroom or other raw image processors so that the default S-curve can be avoided and more or less shadow and highlight contrast can be available as the photographer desires.
So, here's a bit about speed ratings.
Shortly after the beginning of practical photography (daguerreotypes, tintypes, cyanotypes), which required quite long exposure times and produced one-off photos, silver halide negative/positive processes became common, and facilitated by retail sales of dry plates and roll film. It was then necessary to supply film of standardized sensitivity so that usable negatives would result from fixed setting snapshot cameras under sunny conditions or from using an adjustable camera under variable conditions.
It was always desirable to obtain the highest speed rating possible, so that moving objects would not be blurred, camera shake would not be a problem, and great enough depth of field could be obtained for acceptably sharp images to be made with fixed focus lenses, all while having enough exposure that shadow details could be seen in a print. So, the question became, how can we determine the minimum exposure necessary to meet these criteria. (With standardized development also, of course.)
The optical societies, heavily influenced by Kodak scientists, determined the minimum negative density (or slope of the density = shadow detail contrast) that should be reached to produce a negative that yielded a quality print. As the desire was to specify the minimum required exposure, there was little concern with highlight overexposure latitude, as the exposure being sought would leave all the latitude on the bright end.
The next step was to determine what fixed exposure settings would produce the highest percentage of desirable negatives in daylight. To do this, research (again, heavily supported by Kodak) was done to measure with a spot meter the darkest and brightest parts of hundreds of outdoor scenes, and to distill the data down to a fixed number that would result in 95% of shots having adequate exposure in the shadows. The speed rating set this way meant that 95% of outdoor snapshots would result in excellent negatives. Sophisticated photographers with an adjustable camera could increase this percentage especially for shots not taken in sunlight; could determine if their subject was likely to fool the meter (snowscapes, people on the beach, black cats on dark backgrounds) and still use standardized processing. Really sophisticated photographers (Ansel Adams) could just ignore all this and make their own determination of how to expose for the shadows and develop for the highlights.
A part of these investigations was the determination of the average reflection of the hundreds of scenes (about 12.5 percent) so that average-reading reflected light exposure meters could be calibrated (with the smart photographer compensating for meter-fooling bright or dark scenes) (Ansel Adams again went his own way and insisted on using a printer's medium gray of 18% as an average reference, coercing Kodak to publish 18% gray exposure reference cards with a warning to increase meter indicated exposure by 1/2 stop.)
By the way, around 1950 or 1951, the film makers determined that a higher percentage of negatives had acceptable shadow detail than they thought, and the ISO ratings of negative films were doubled (increased by one stop) simply by changing the number on the box - no change in film or processing whatsoever. All the non-adjustable cameras out there got the same density negatives they always did, but photographers with adjustable cameras not got better motion stopping or depth of field.
Color negative film followed the same general rating process of exposing for acceptable shadow density.
A note here about "disposable" cameras. These usually had/have aperture and shutter settings that corresponded to ISO 100, but the film in them was usually actually ISO 800. This meant that 3 stops of its overexposure latitude was used in bright sunlight, and the exposure was "correct" (minimum shadow detail for a good print) in open shade.
Color reversal (slide) film ratings had to be determined differently. In negative/positve processes, the negative determines shadow compression and the print determines highlight compression. Reversal film has both compressions built in - the entire "S curve" is determined in the original piece of film and therefore the great overexposure latitude of a negative is not available. Therefore, transparency film ISO rating is determined by the mid tones, trying different ratings with experts judging what ISO produces the greatest number of good pictures.
Now to digital cameras:
Digital sensors have photosites that accumulate charge proportional to the amount of light that hits them during exposure. They have a maximum level above which they simply won't accumulate any more - highlight clipping. So, the camera maker determines what level of charge will be middle gray (~level 127 in a jpg file), whether this is generally 12.5% of diffuse white or 18% or some other number, and therefore how much overexposure latitude is available in the raw image (and highlight recovery in post processing). Digital sensors do not have a toe (loss of contrast in the shadows) so that does not give any clue to setting the ISO rating using the method for film. They do, however have noise that is more visible in the shadows than the highlights, so that prevents the manufacturer from using unlimited overexposure latitude and unlimited ISO.
Summary for digital camera ISO: manufacturer picks the maximum charge level (lowest possible ISO) that will correspond to middle gray in a rendered image. Higher ISO settings simply declare a lower charge to be mid gray, and then either analog amplify the signal or multiply its digital value, or (often) both.
Since both prints and monitors have a certain limited contrast range, digital processes always apply an S-curve that mimics film's gradual shadow and highlight compression somewhere along the way. The S-curve makes much better looking pictures than if the shadows or highlights suddenly clipped at a certain level. Some digital processing experts have provided linear camera profiles for use in Lightroom or other raw image processors so that the default S-curve can be avoided and more or less shadow and highlight contrast can be available as the photographer desires.
Last edited:
