With regard to exposure (not white balance), Chuck appears to use the textured white target as his highlight exposure reference and depends on how it is actually read by the sensor to determine whether that highlight is being recorded at the level he intends.
At least that's how I understand that portion of his method, and it's what I've adopted. I'm an old Zone System user since the early 70s, and I'd been working hard to make digital make sense within some portion of Zone System methodology until Chuck led me to realize that recording data collected by a digital sensor is just fundamentally different from recording data on film.
Recording light intensity data collected by a digital sensor has much more in common with recording audio data collected by a microphone. I've done a bit of that, and I realized that's really where Chuck has gone with his exposure methodology. For exposure, the gray card is actually part of the wrong methodology. It was great for film, but this is a fundamentally different thing.
Another part of what I have to call the paradigm shift from film exposure to digital exposure is realizing that the superiority of learning to truly understand the histogram lies in the fact that a light meter can only provide a prediction of exposure conditions. The histogram provides a report of what the sensor actually collected.
It's like the difference between a weatherman's prediction of rain next week and a weather report of the rain this morning--a report of what actually happened will always have the potential of greater accuracy than a report of what might happen. Presuming we interpret the report properly.
I think we need to further explore how to properly interpret the histogram, and I think Chuck's concept of identifying the top end of the sensor's range and using the histogram to ensure it's placed properly makes good use of the sensor itself to set exposure rather than a light meter that is far, far removed from the actual "ground truth."
As the song says, "You don't need a weatherman to tell you which way the wind is blowing."
RDKirk wrote:
Recording light intensity data collected by a digital sensor has much more in common with recording audio data collected by a microphone. I've done a bit of that, and I realized that's really where Chuck has gone with his exposure methodology. For exposure, the gray card is actually part of the wrong methodology. It was great for film, but this is a fundamentally different thing.
Good points and an excellent analogy to draw from.
What my reproduction background taught me is the importance of getting the perceptual reference tones on both ends of the tonal scale correct: A good solid blacks and shadow detail, and a clean specular whites and highlight detail.
The photographic process is engineered to be linear, which means if we get the ends correct all the other steps of the tonal scale will have a normal progression: a gray scale will look the same in a photo as is does by eye perceptually.
After learning to shoot halftones National Geographic I realize that the Zone System which I'd learned a few years earlier really wasn't moving Zone V around. The middle of the tonal scale on the straight line part of the D vs E plot rises and falls with the highlights due to the D=log(E) natural of the photographic response.
The halftone process is what Levels was modeled on. It allows changing the middle without affecting the ends, something that can't be done on film. With digital if we shoot and get the BOTH points correct -- the reason I preach about fill and using flash outdoors -- we can change the "internal" contrast by moving the middle slider in Levels, or raising / lowering the middle in curves.
Also not realized I think is how sensor range affects tonal rendering. A camera histogram represents the range of any sensor. Perceptual middle gray falls in the middle between black and white.
If a camera sensor has a 4 stop range from black to white then the middle is 2 stops from either end. Thus the difference in exposure needed to render something that is white as middle gray would be 2 stops.
If a camera has sensor has a 8 stop range from black to white then the middle is 4 stops from either end. Thus the difference in exposure to render something that is white as middle gray would be 4 stops.
That that means in practical terms is that if a file is exposed for detail at both ends where the mid-tones like shadows on a face will fall in terms of f/stop difference from the sensor will be determined by the range of the sensor and will fall around DR/2.
Even in the Zone System we'd manipulate contrast at capture and dodge and burn. What those things did is alter the middle tones beyond what the limits of the linear photographic process was able to record. We really have no idea whether the deep shadows in Penn's photo were the result of lighting, or lighting combined with burning in the shadows hand-puppet style on the enlarging easel. Being in the reproduction business I know what photographers would do if making many copies of a darkroom manipulated print would be to make a high quality copy negative of the edited print then make the copy prints from that neg. No darkroom manipulation was needed that way.
Today when we shoot digital and expose for the end points the middle tones will wind up DR/2 stops from the highlights. If we want to make the mid-tones lighter or darker if we lower or rise the fill it will affect the tone of the darker shadows and we risk making them lighter the natural which will make the photo look flat, or darker and loose detail. The better approach, from a technical point of view is to:
1) "Dodge / burn" the lighting while shooting using gobos (go betweens) between the fill light and face in areas we want darker than the sensor response puts them when deep shadows are exposed with detail, and add a supplemental fill source (reflector, second gridded fill light) aimed just at the mid-tones on the shadow side of the face we want darker. Cinematographers use this method as a staple out of necessity but few still photographers except the very savvy would think to use a gobo or reflector to manipulate the fill source nowadays.
2) Dodge / burn the print. That's a generalization for all the different ways mid-tones can be manipulated independently from the end points in photo editors: levels, curves, adjustment layers, cloning, etc.
I think a few like RDKirk who did zone system work will understand what I'm saying here: We can get both detail on both ends of the scale and make the shadows on the face light or dark at the same time, but we can't rely on just the response of the sensor to put the mid-tones where we want them. Those tricks were part of the secret to the buttery smooth tonality of the studio work done with hard lights back in the 1930s and 40s characterized by a full rich tonal range and magical control if the lighting. It wasn't magic it was knowing how to use a gobo, a skill learned from the guys next door making the movies the stars were acting in.
Here is an image by Penn, BTW, that I personally regard as superior to the Picasso portrait. This is of the French novelist Colette, who was 80 years old and bedridden when he took her photograph in her bedroom by window light.
Rolliflex with 75mm f/3.5 lens, Plux-X film, 1/2 second at f/11.
