Camera Settings — Position cameras and set focal length and sensor settings - Blender

SECTION 1

Historical Context & Motivation

The act of framing a scene through a camera—whether physical or virtual—rests on centuries of optical science and artistic convention. Long before Blender or any 3D software existed, painters and photographers wrestled with the same fundamental question: how does one translate a three-dimensional world onto a two-dimensional surface? The camera obscura, an ancestor of the modern camera, demonstrated that light passing through a small aperture could project an inverted image onto a surface, establishing the geometric principles that underpin every virtual camera in Blender today. Understanding this lineage is essential for visual artists because the decisions you make in Blender's camera properties panel—focal length, sensor size, position, and rotation—are direct descendants of physical optics and photographic craft.

~1000 CE • Ibn al-Haytham's Optics

The Arab polymath Ibn al-Haytham (Alhazen) published the Book of Optics, rigorously describing the camera obscura effect and laying the groundwork for perspective projection used in every 3D render engine.

1839 • Birth of Photography

Daguerre and Niépce introduced practical photographic processes, linking lens focal length and plate size (the 'sensor') to field of view—a relationship Blender replicates precisely in its camera object.

1995 • Blender's Genesis

Ton Roosendaal created Blender at NeoGeo, incorporating a virtual camera system modeled on real photographic conventions including focal length, sensor dimensions, and depth of field.

2018–Present • Physically Based Cameras in Blender 2.8+

With the EEVEE and Cycles overhaul, Blender adopted physically based camera models supporting real-world sensor presets, anamorphic bokeh, and cinematic depth-of-field simulations—bringing virtual cinematography closer than ever to live-action production.

This historical progression reveals a persistent question that digital artists must answer every time they set up a shot in Blender: how do camera position, focal length, and sensor size interact to determine what the viewer sees? The rest of this lesson provides the conceptual, mathematical, and practical tools to answer that question with confidence.

SECTION 2

Core Principles & Definitions

Before adjusting any slider or typing any value in Blender's properties panel, you need a clear mental model of the four interrelated concepts governing virtual cameras. Each concept maps directly to a physical camera property, so understanding them here transfers seamlessly to real-world cinematography and studio photography.

Camera Position & Orientation

The camera's location (X, Y, Z) and rotation (pitch, yaw, roll) in world space determine the viewpoint from which the scene is rendered. In Blender, these are standard object transforms accessible via N-panel → Item or the Object Properties tab.

Focal Length

Measured in millimeters, focal length controls how 'zoomed in' the camera appears. Short focal lengths (e.g., 18 mm) produce wide-angle views with pronounced perspective distortion; long focal lengths (e.g., 200 mm) compress depth and narrow the field of view.

Sensor Size

The sensor defines the virtual imaging plane's physical dimensions (width and height in mm). Blender defaults to 36 mm × 24 mm, matching a full-frame 35 mm still camera. Changing sensor size while keeping focal length constant alters the field of view.

Field of View (FOV)

The angular extent of the scene visible through the camera. Field of view is a derived quantity: it depends on both focal length and sensor size. Blender can display or accept FOV directly as an alternative to focal length in the camera data panel.

Projection Type

Blender offers perspective (objects diminish with distance) and orthographic (no foreshortening) projection. Perspective is the default and mimics how the human eye and real lenses perceive the world.

✦

KEY TAKEAWAY

Think of the camera as a window in a wall. Position and rotation decide where the wall sits and which direction it faces. Sensor size determines how large the window is. Focal length is like stepping closer to or farther from that window—a shorter focal length is like pressing your face to the glass (wide view), while a longer one is like viewing through a telescope mounted at the window (narrow, compressed view).

SECTION 3

Visual Explanation — Camera Geometry

The following diagram illustrates how a perspective camera in Blender projects a 3D scene onto a 2D image plane. The camera is positioned at a single point in space, and light rays converge toward this point through the rectangular sensor (image plane). The angular spread of these rays defines the field of view, which is controlled jointly by focal length and sensor dimensions.

Top-down schematic of a perspective camera. The camera point (purple) sits at the apex of the viewing frustum. The image plane / sensor (pink line) is positioned one focal length away. The cyan cone represents the field of view.

In the diagram above, notice that the focal length (yellow arrow) is the perpendicular distance from the camera's optical center to the image plane. If you shorten this distance, the sensor subtends a larger angle—widening the FOV and capturing more of the scene, albeit with increased perspective distortion. If you lengthen the focal length, the frustum narrows, compressing depth and isolating subjects. This geometric relationship is the single most important visual concept in virtual cinematography. In Blender, you can observe this frustum in real-time by selecting the camera and enabling Display → Limits in the Object Data Properties panel.

SECTION 4

Mathematical Framework

Blender's camera system is grounded in straightforward trigonometry. The central equation relates the horizontal field of view (FOV) to the focal length (f) and the sensor width (S). Because the sensor is centered on the optical axis, each half of the sensor subtends an angle whose tangent equals half the sensor width divided by the focal length.

HORIZONTAL FIELD OF VIEW

FOV_h = 2 × arctan( S_w / (2 × f) )

Where FOV_h is the horizontal field of view in degrees, S_w is the sensor width in mm, and f is the focal length in mm.

VERTICAL FIELD OF VIEW

FOV_v = 2 × arctan( S_h / (2 × f) )

Where S_h is the sensor height in mm. For a standard full-frame sensor (36 × 24 mm), the vertical FOV is always narrower than the horizontal.

SENSOR ASPECT RATIO & RESOLUTION

Aspect Ratio = S_w / S_h = R_x / R_y

Where R_x and R_y are the render resolution in pixels. Blender computes the sensor 'fit' mode (Auto, Horizontal, Vertical) to ensure the FOV aligns with the chosen render resolution.

These equations have a practical corollary: doubling the focal length halves the apparent angular size of the scene (approximately, for small angles). Likewise, if you switch from a full-frame sensor (36 mm wide) to a Micro Four-Thirds sensor (17.3 mm wide) without changing focal length, you effectively crop the image, producing a narrower FOV equivalent to a longer lens. This is the well-known crop factor concept. In Blender, you can exploit this by adjusting sensor size to emulate specific camera bodies (RED, ARRI Alexa, Canon 5D, etc.) and lenses simultaneously.

CROP FACTOR

Crop Factor = 36 / S_w(actual)

A crop factor of 1.0 corresponds to full-frame. An APS-C sensor (~23.5 mm wide) yields a crop factor of ≈ 1.53, meaning a 35 mm lens behaves like a ≈ 53.6 mm lens on full-frame.

SECTION 5

Focal Length Classification & Visual Impact

Cinematographers and photographers classify lenses into broad categories based on focal length, and Blender artists benefit from internalizing these categories because they carry strong aesthetic connotations. A 24 mm lens communicates vastness and immediacy; a 135 mm lens suggests voyeuristic distance or intimate portraiture with creamy background separation. Choosing a focal length is as much an artistic decision as a technical one.

Side-by-side comparison of wide-angle, normal, and telephoto focal length ranges, showing how the viewing cone narrows as focal length increases. The bottom bar maps focal length ranges onto the full-frame (36 mm sensor) spectrum.

Focal Length	Horizontal FOV (36 mm sensor)	Common Use in CG / Film
14 mm	≈ 104°	Extreme wide / environments, virtual tours
24 mm	≈ 73°	Architecture, establishing shots, action sequences
35 mm	≈ 54°	Documentary feel, dialogue scenes
50 mm	≈ 39°	Human-eye equivalent, 'nifty fifty', general purpose
85 mm	≈ 24°	Character close-ups, portraiture
135 mm	≈ 15°	Beauty shots, product rendering, compressed depth
200 mm	≈ 10°	Extreme compression, surveillance aesthetics

Common focal lengths and their approximate horizontal FOV on a 36 mm sensor

💡

Blender Tip

In the Camera Object Data Properties panel (camera icon), you can type a focal length directly or switch the 'Lens Unit' dropdown from Millimeters to Field of View if you prefer thinking in degrees. The conversion uses the FOV equation from Section 4 automatically.

SECTION 6

Worked Example — Setting Up a Cinematic Character Shot

Suppose you are creating a character portrait render in Blender. Your character stands at the world origin, and you want a medium close-up with a natural perspective similar to what a director of photography would achieve on set with a Super 35 mm cinema camera and an 85 mm lens. Your render resolution is 1920 × 1080. Let's walk through the full setup.

Cinematic Character Portrait Setup

1
Step 1 — Set the Sensor to Super 35

In the Camera Object Data Properties panel, expand the 'Camera' section. Set the Sensor Fit to Horizontal. Set Sensor Size to 24.89 mm (the Super 35 mm width standard used by ARRI Alexa in 4-perf mode). Blender will automatically compute the sensor height from the render aspect ratio.

Sensor Width = 24.89 mm, Sensor Height ≈ 14.0 mm
2
Step 2 — Set the Focal Length

In the same panel, set the Focal Length to 85 mm. This will yield a horizontal FOV of 2 × arctan(24.89 / (2 × 85)) = 2 × arctan(0.1464) ≈ 2 × 8.33° ≈ 16.66°. Note that this is considerably narrower than the ≈ 24° horizontal FOV that an 85 mm lens produces on a full-frame (36 mm) sensor, because the smaller Super 35 sensor captures a tighter portion of the image circle. This narrow angle is ideal for portraiture—it flatters facial proportions and provides gentle background compression.

Focal Length = 85 mm → Horizontal FOV ≈ 16.7° (Super 35 sensor); equivalent to ≈ 24° on full-frame
3
Step 3 — Position the Camera

For a medium close-up, you want the framing to run from roughly mid-chest to slightly above the head. If the character model is approximately 1.8 m tall, mid-chest is around Z = 1.2 m. Position the camera at X = 0, Y = −3.0, Z = 1.4 m. To aim the camera at the character's face near the origin, apply a Track To constraint targeting an Empty placed at the character's nose—this is the most reliable method. If you prefer setting a manual X rotation instead, the required angle is approximately 90° − arctan(1.4 / 3.0) ≈ 90° − 25° = 65°, since the camera is elevated 1.4 m above the origin and 3.0 m along Y. Note that a flat, horizontal camera would be 90° in Blender's convention; tilting down to the target reduces the X rotation to approximately 65°.

Camera Location = (0, −3.0, 1.4) with Track To constraint (or X rotation ≈ 65°)
4
Step 4 — Verify Through Camera View

Press Numpad 0 to enter camera view. The character's head should be centered in the frame with comfortable headroom. If the framing is too tight, increase the Y distance (move the camera back); if too loose, decrease it or increase the focal length. Fine-tune by pressing G then Middle Mouse to dolly the camera along its local axis while in camera view.

Confirmed: Medium close-up framing with natural perspective
5
Step 5 — Compare Full-Frame Equivalent

The crop factor for Super 35 relative to full-frame is 36 / 24.89 ≈ 1.446. The full-frame equivalent focal length is 85 × 1.446 ≈ 122.9 mm. This means your shot will have the same field of view as a 123 mm lens on a full-frame camera. As a cross-check, the horizontal FOV of a 123 mm lens on a 36 mm sensor is 2 × arctan(36 / (2 × 123)) ≈ 2 × arctan(0.1463) ≈ 16.7°, which matches the 16.7° computed in Step 2—confirming that the Super 35 / 85 mm setup and the full-frame / 123 mm setup are equivalent. Note that the table in Section 5 lists an 85 mm full-frame FOV of ≈ 24°; your Super 35 sensor narrows this significantly, which is why the full-frame equivalent is the longer 123 mm focal length.

Full-Frame Equivalent ≈ 123 mm (FOV ≈ 16.7°) — classic portrait compression

SECTION 7

Perspective vs. Orthographic & Sensor Presets

Blender offers two fundamental projection types, and the choice between them profoundly affects both the aesthetic and informational quality of your render. Additionally, the sensor preset you select determines how your chosen focal length maps to a specific field of view. Understanding these comparisons empowers you to match your virtual camera to the look of specific real-world camera systems or to make deliberate departures from photorealism.

Property	Perspective Projection	Orthographic Projection
Foreshortening	Yes — distant objects appear smaller, creating depth	No — all objects rendered at true proportional size regardless of distance
Parallel Lines	Converge toward vanishing points	Remain parallel in the render
Focal Length	Controls FOV and perspective distortion	Not applicable — 'Orthographic Scale' controls visible area
Depth of Field	Available — mimics real lens blur	Technically possible but rarely used; lacks realism
Typical Use	Cinematic renders, character art, product viz	Isometric game art, technical illustration, UI mockups

Perspective vs. Orthographic projection in Blender

Sensor Preset	Sensor Width (mm)	Crop Factor	50 mm Equiv. FOV
Full-Frame (35 mm)	36.0	1.0×	≈ 39.6°
Super 35 (ARRI)	24.89	1.45×	≈ 28.0°
APS-C (Canon)	22.3	1.61×	≈ 25.1°
Micro Four-Thirds	17.3	2.08×	≈ 19.6°

Common sensor presets and their effect on a 50 mm lens

✦

KEY TAKEAWAY

Sensor size and focal length are like partners in a dance: changing either one affects the field of view. It is analogous to how a recording studio's microphone placement (camera position) and its polar pattern (sensor/focal-length combo) together determine what the listener hears. In a VFX pipeline, matching your virtual sensor preset to the real camera used on set is critical for compositing CG elements seamlessly into live-action plates.

SECTION 8

Connection to Advanced Camera Techniques

The foundational camera settings covered in this lesson—position, focal length, and sensor size—form the base layer upon which more advanced cinematic techniques are built. As you progress, you will encounter features that extend this foundation: depth of field (controlled by f-stop and focus distance), motion blur (shutter angle and speed), lens distortion (barrel and pincushion effects via compositing nodes), and camera tracking (solving a real camera's motion path from footage to replicate it virtually). Each of these depends on having the correct baseline camera settings established first.

Concept	This Lesson (Foundation)	Advanced Extension
Position / Orientation	Manual placement via transforms or constraints	Camera tracking from solved footage, animated dolly/crane rigs, follow-path constraints
Focal Length	Static value controlling FOV	Animated focal length (dolly zoom / Vertigo effect), anamorphic lens simulation
Sensor Size	Preset selection matching a physical camera	Anamorphic squeeze ratios (2× for CinemaScope), custom sensor for tilt-shift effects
Projection	Perspective vs. orthographic	Panoramic (equirectangular, fisheye) for VR/360° renders in Cycles

Foundation concepts vs. their advanced extensions

As you move into depth-of-field work, remember that the circle of confusion—the optical basis for bokeh—is directly influenced by focal length and sensor size. A longer focal length at the same f-stop produces a shallower depth of field, and a larger sensor produces a shallower depth of field for the same framing. These relationships build directly on the geometry explored in Sections 3 and 4 of this lesson, reinforcing why mastering the basics of camera settings is essential before attempting cinematic depth-of-field effects.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

A Blender artist switches the camera's focal length from 50 mm to 24 mm without moving the camera. Describe qualitatively what will happen to the rendered image in terms of field of view, perspective distortion, and the relative apparent size of foreground vs. background objects.

Show answer

Switching from 50 mm to 24 mm roughly doubles the horizontal field of view (from ≈ 39° to ≈ 73° on a 36 mm sensor), so much more of the scene becomes visible in the frame. Perspective distortion increases significantly: objects close to the camera will appear disproportionately large compared to objects farther away, exaggerating the sense of depth. Background elements will appear to shrink and recede, while foreground elements will dominate the composition. Straight lines near the edges of the frame may appear to bow outward slightly, especially if the subject has strong architectural geometry. This is why wide-angle lenses are avoided for portraiture—they can make noses appear bulbous and ears seem to recede unnaturally.

Question 2: BASIC CALCULATION

Calculate the horizontal field of view for a camera in Blender with a focal length of 35 mm and a sensor width of 36 mm (full-frame). Express your answer in degrees, rounded to one decimal place.

Show answer

Using the FOV equation: FOV_h = 2 × arctan(S_w / (2 × f)) = 2 × arctan(36 / (2 × 35)) = 2 × arctan(36 / 70) = 2 × arctan(0.5143). The arctan of 0.5143 is approximately 27.22°. Therefore, FOV_h = 2 × 27.22° ≈ 54.4°. This places the 35 mm lens at the wide end of the 'normal' range, making it a versatile choice for documentary-style scenes in Blender. You can verify this by setting these values in Blender and switching the lens unit to 'Field of View' to see the readout.

Question 3: INTERMEDIATE

You are compositing a CG robot into live-action footage shot on a RED Komodo camera (sensor width 26.21 mm) with a 50 mm lens. What sensor width and focal length should you set in Blender so that your virtual camera matches the real camera's field of view? Alternatively, if you keep Blender's default 36 mm sensor, what focal length would produce the equivalent FOV?

Show answer

The most accurate approach is to match the real camera exactly: set the Blender sensor width to 26.21 mm and the focal length to 50 mm. This produces FOV_h = 2 × arctan(26.21 / 100) = 2 × arctan(0.2621) ≈ 2 × 14.69° ≈ 29.4°. If you prefer to keep the default 36 mm sensor, you need a focal length that yields the same FOV: 29.4° = 2 × arctan(36 / (2 × f)), so arctan(18 / f) = 14.69°, meaning 18 / f = tan(14.69°) ≈ 0.2621, thus f = 18 / 0.2621 ≈ 68.7 mm. Setting 36 mm sensor with ≈ 68.7 mm focal length will match the RED Komodo's 50 mm perspective. This demonstrates the crop factor: 36 / 26.21 ≈ 1.374, and 50 × 1.374 ≈ 68.7 mm.

Question 4: APPLIED

You are creating an animated architectural walkthrough in Blender. The client wants the camera to start outside the building with a dramatic wide-angle establishing shot (FOV ≈ 90° horizontal), then smoothly transition as the camera pushes through the front door to a comfortable interior view (FOV ≈ 50° horizontal). Assuming a 36 mm sensor, calculate the starting and ending focal lengths. Then describe a Blender workflow to animate this transition.

Show answer

For the starting FOV of 90°: 90° = 2 × arctan(18 / f), so arctan(18 / f) = 45°, meaning 18 / f = tan(45°) = 1.0, thus f = 18 mm. For the ending FOV of 50°: 50° = 2 × arctan(18 / f), so arctan(18 / f) = 25°, meaning 18 / f = tan(25°) ≈ 0.4663, thus f = 18 / 0.4663 ≈ 38.6 mm. In Blender, you would keyframe the camera's focal length: on the first frame, hover over the focal length field (set to 18 mm) and press I to insert a keyframe. Move to the frame where the camera enters the interior, set the focal length to approximately 38.6 mm, and insert another keyframe. Use the Graph Editor to adjust the interpolation curve—a smooth ease-in/ease-out avoids jarring focal-length shifts. Simultaneously, animate the camera's position along a path using a Follow Path constraint for smooth spatial movement.

Question 5: CRITICAL THINKING

The 'dolly zoom' (Vertigo effect) maintains a subject at the same apparent size in the frame while dramatically changing the background's compression. Explain, using the FOV equation and the relationship between camera distance and focal length, how this effect works. Then outline how you would implement it in Blender, identifying which parameters must be animated and how they must covary.

Show answer

The dolly zoom works by simultaneously moving the camera toward (or away from) the subject while adjusting the focal length so that the subject maintains the same angular size in the frame. As the camera moves closer, the focal length increases (zooming in) to keep the subject the same size; however, the narrower FOV compresses the background, making it appear to rush forward. Conversely, moving away while decreasing focal length makes the background appear to expand. Mathematically, for a subject of height h at distance d, its angular size is approximately 2 × arctan(h / (2 × d)). To keep this constant while changing d, you must change f proportionally: if you halve d, you roughly double f (for small angles, angular size ≈ h/d and framing size ≈ f × h/d / sensor, so f must scale inversely with d). In Blender, animate the camera's position along the Y-axis (or a path) and simultaneously keyframe the focal length. At each keyframe, compute f_new = f_original × (d_new / d_original) × correction_factor. Use the Graph Editor to ensure both curves are synchronized. For extra precision, use a driver: set the focal length property to be driven by the distance between the camera and an Empty placed at the subject, using the expression f_base × (distance / d_base). This ensures mathematically locked covariation and produces the unsettling spatial distortion that defines the Vertigo effect.

SUMMARY

Lesson Summary

Summary

This lesson established the foundations of virtual cinematography in Blender. You learned that a camera's position and rotation define the viewpoint, its focal length (measured in mm) controls how much of the scene is captured and how perspective is perceived, and its sensor size determines the imaging plane's physical dimensions. The core equation—FOV = 2 × arctan(sensor / (2 × focal length))—links these parameters, and the crop factor lets you convert between sensor formats so that your virtual camera matches a specific physical camera system.

You explored how wide-angle lenses (short focal lengths) exaggerate depth and distort perspective, normal lenses approximate human vision, and telephoto lenses compress depth and isolate subjects. You practiced positioning a camera for a cinematic portrait using a Super 35 sensor preset, and you learned how perspective vs. orthographic projection serves different artistic goals. These settings form the essential foundation for more advanced techniques like depth of field, camera tracking, and the dolly zoom effect.