Photography Exposure and Image Quality Syllabus

What is exposure bracketing (AEB), and what is a typical bracketing sequence?

Bracketing captures multiple frames of the same scene at different exposures to guarantee one is correct or to merge them later. A typical auto exposure bracketing set is 3 frames at $-2$, $0$, and $+2$ EV (or 5 frames at $\pm 1$ EV steps).

Give two main purposes of exposure bracketing.

(1) Insurance in tricky or changing light — pick the best-exposed frame afterward. (2) Source frames for HDR merging, where dark frames preserve highlights and bright frames preserve shadow detail of a high-contrast scene.

What does 'Expose to the Right' (ETTR) mean, and what is its limit?

ETTR means giving the greatest exposure possible so the histogram is pushed toward the right edge without clipping highlights. This maximizes signal-to-noise ratio; the limit is that any highlight actually clipped is unrecoverable, so you stop just short of clipping.

Why does ETTR improve image quality in raw files (the tonal-levels argument)?

A linear sensor allocates half of all recordable tonal levels to the brightest stop: in a 12-bit file with $2^{12} = 4096$ levels, the top stop holds $2048$ levels, the next $1024$, and so on. Exposing brighter places data in level-rich stops, so shadows show less noise and banding after normalization in post.

On an image histogram, what do the horizontal and vertical axes represent?

The horizontal axis is tonal value (luminance) from pure black at the left ($0$) to pure white at the right ($255$ in 8-bit terms); the vertical axis is the number of pixels at each tonal value.

How do the histograms of a low-key and a high-key image differ, and is either 'wrong'?

A low-key image's histogram is massed toward the left (shadows); a high-key image's is massed toward the right (highlights). Neither is wrong — there is no single 'correct' histogram shape; it must match the intended tonal character of the scene.

What does 'clipping' mean on a histogram, and how does it appear?

Clipping means tones have been pushed beyond the sensor's recordable range, so detail is lost. It appears as a spike pressed against the histogram's right edge (blown highlights recorded as pure white) or left edge (blocked/crushed shadows recorded as pure black).

In digital photography, which is generally harder to recover in post: clipped highlights or underexposed shadows, and why?

Clipped highlights. Once the sensor saturates, no data exists to recover, whereas underexposed shadows usually retain some data that can be lifted (at the cost of noise). Hence the common digital rule: protect the highlights.

What are 'blinkies' (highlight alert) on a camera, and what are they used for?

Blinkies are a playback warning that flashes areas of the image where highlights are clipped to pure white. They let the photographer spot blown regions instantly and reduce exposure to preserve highlight detail.

Why can the RGB (per-channel) histogram reveal clipping that the luminance histogram hides?

The luminance histogram is a weighted combination of channels, so a single saturated channel — e.g., the red channel in a vivid red flower — can clip while overall luminance still looks safe. The RGB histogram shows each channel separately, exposing single-channel saturation that causes lost texture and color distortion.

In an RGB histogram, one channel is clipped at the right edge while the other two are not. What is the practical consequence in the image?

The clipped channel loses tonal detail, so saturated areas of that color render as flat, textureless patches with shifted hue (e.g., blown reds in flowers or stage lighting). Reducing exposure or saturation until that channel comes off the edge restores detail.

Define dynamic range in photography and state the unit used to measure it.

Dynamic range is the ratio between the brightest and darkest tones a scene contains or a device can capture/display simultaneously with detail. It is measured in stops (EV), where each stop is a factor of 2 in luminance.

Roughly compare the dynamic range of a modern full-frame sensor, a JPEG file, and a typical print.

Modern full-frame raw sensors capture about $12$–$15$ stops; a standard JPEG holds roughly $8$–$9$ stops; a photographic print reproduces only about $6$–$7$ stops. High-contrast scenes can exceed all of these, forcing a choice of what to sacrifice or a technique like HDR.

What happens when a scene's dynamic range exceeds the camera sensor's dynamic range in a single exposure?

The photographer must sacrifice one end of the tonal scale: expose for the highlights and let shadows block up, or expose for the shadows and blow the highlights. Detail outside the sensor's range is clipped and unrecoverable in that frame.

Describe the standard HDR workflow for a high-contrast scene.

Shoot a bracketed series of the same composition (e.g., $-2$, $0$, $+2$ EV) from a tripod, keeping aperture constant and varying shutter speed; merge the frames in software into a single high-dynamic-range image; then tone-map the result to compress its range for display.

What is tone mapping in HDR photography, and why is it needed?

Tone mapping compresses the wide tonal range of a merged HDR image into the much narrower range a monitor or print can show, while preserving local contrast and detail. It is needed because display media hold far fewer stops than the merged data.

Name two classic artifacts of HDR photography and their causes.

(1) Ghosting — moving elements (people, foliage, water) appear semi-transparent or doubled because they shifted between bracketed frames. (2) Halos and an unnatural flat 'HDR look' — caused by aggressive local tone mapping that exaggerates local contrast around edges.

When shooting HDR brackets, why should exposure be varied with shutter speed rather than aperture?

Changing aperture alters depth of field between frames, causing focus/blur mismatches that ruin the merge. Varying shutter speed changes only brightness, keeping sharpness and depth of field identical across the bracket.

What is a graduated ND (GND) filter, and for what classic situation is it used?

A graduated neutral density filter is dark at the top and transitions to clear at the bottom, reducing light in part of the frame without changing color. Classic use: darkening a bright sky to balance it with a darker foreground in landscape photography, capturing both in one exposure.

Distinguish hard-edge, soft-edge, and reverse graduated ND filters.

Hard-edge GND: abrupt transition, for flat horizons (seascapes). Soft-edge GND: gradual transition, for broken horizons (mountains, trees). Reverse GND: darkest just above the center line, fading upward — for sunrise/sunset when the brightest zone sits right at the horizon.

How many stops of light does an ND filter with optical density 0.9 block? State the general relationship.

Density $0.3$ corresponds to 1 stop, since each stop is a factor $2 \approx 10^{0.3}$. Therefore density $0.9$ blocks $\frac{0.9}{0.3} = 3$ stops, transmitting $\frac{1}{2^{3}} = \frac{1}{8}$ of the light (also sold as 'ND8' or '3-stop').

Compare phase-detection and contrast-detection autofocus.

Phase detection splits incoming light and compares the two images' phase, so it knows both direction and distance of the focus error — fast, great for tracking motion. Contrast detection maximizes edge contrast on the sensor by hunting back and forth — historically slower but very accurate; modern hybrid systems combine both on-sensor.

Compare AF-S (single-servo) and AF-C (continuous-servo) autofocus modes and when to use each.

AF-S (One-Shot) locks focus once when half-pressing the shutter — best for still subjects. AF-C (AI Servo) continuously re-adjusts focus while the button is held, predicting subject movement — best for sports, wildlife, and anything in motion.

What is back-button focus and why do photographers use it?

Back-button focus assigns autofocus to a rear button (e.g., AF-ON), decoupling it from the shutter release. This lets the photographer focus once and recompose or shoot repeatedly without the camera refocusing, and switch instantly between locked and continuous focusing behavior.

What is focus peaking, and in what situations is manual focus preferred over autofocus?

Focus peaking is an electronic viewfinder/live-view aid that overlays a colored highlight on the edges of highest contrast, i.e. the in-focus areas. Manual focus is preferred in low light, low-contrast scenes, macro work, astrophotography, video (to avoid AF hunting), and when shooting through obstacles like glass or fences.

Define hyperfocal distance and state what depth of field you get when focused at it.

The hyperfocal distance $H$ is the nearest focus distance at which infinity remains acceptably sharp. Focusing at $H$ yields acceptable sharpness from $\frac{H}{2}$ to $\infty$ — the maximum possible depth of field for that aperture and focal length.

Give the hyperfocal distance formula and define each symbol.

$$H = \frac{f^{2}}{N\,c} + f$$ where $f$ is the focal length, $N$ is the f-number, and $c$ is the circle of confusion — the largest blur spot still perceived as sharp (about $0.03\,\text{mm}$ for full frame). The $+f$ term is often negligible.

Compute the hyperfocal distance for a 24mm lens at $f/8$ on full frame ($c = 0.03\,\text{mm}$), and state the resulting zone of sharpness.

$$H = \frac{24^{2}}{8 \times 0.03} + 24 = \frac{576}{0.24} + 24 = 2424\,\text{mm} \approx 2.4\,\text{m}$$ Focusing at $2.4\,\text{m}$ gives acceptable sharpness from about $1.2\,\text{m}$ ($\frac{H}{2}$) to infinity.

What is focus stacking, and in which genres is it most used?

Focus stacking captures a series of frames of the same composition with the focus point stepped from nearest to farthest, then blends only the sharp regions of each frame in software into one image with extended depth of field. It is most used in macro photography (where DOF is millimeters) and landscape photography with close foregrounds.

Why is focus stacking often preferred over simply stopping down to $f/22$ for maximum depth of field?

At very small apertures diffraction softens the entire image, so $f/22$ trades depth of field for overall sharpness. Stacking lets each frame be shot at the lens's sharpest aperture (around $f/5.6$–$f/8$), combining diffraction-free sharpness with front-to-back focus; requirements are a static scene, a tripod, and consistent exposure across frames.