I Turned My Office Into a Camera Obscura
What I learned about optics, aperture size, diffraction, and exposure from turning my office into a giant pinhole camera
2016-12-09
Loading content...
What I learned about optics, aperture size, diffraction, and exposure from turning my office into a giant pinhole camera
I keep coming back to the camera obscura. It is one of the oldest ideas in optics, and it still explains every camera we use today: light through a small hole paints an inverted picture on a dark surface. I wanted to see that at room scale, so I blacked out my office and tried it.
Camera obscura (Latin for "dark chamber") is what you get when light passes through a small hole into a dark room and paints an inverted image on the far wall. Same principle behind a pinhole box and a modern DSLR.
The setup is simple: seal the room to light, cut a small aperture in one wall (I used a covered window), and the outdoor scene appears upside down on the opposite wall.
Camera obscura works because light rays travel in straight lines.
Light travels in straight lines (in homogeneous media). When light from an object passes through a small aperture, each point on the object sends rays through the hole. Those rays keep going until they hit a surface and form a point of light.
The projection math comes from similar triangles:
h'/h = d'/d
Where:
h = height of objecth' = height of projected imaged = distance from object to apertured' = distance from aperture to projection surface
The inversion happens because light rays cross at the aperture. Light from the top of an object passes through the hole and hits the bottom of the projection surface, while light from the bottom hits the top. The result is a vertically and horizontally flipped image.
Aperture size controls image quality. If the hole is too large, the image gets blurry because light cones overlap. If the hole is too small, diffraction becomes the limit.
For a pinhole camera, you can estimate optimal aperture diameter with:
d_optimal = √(2λD)
Where:
d = aperture diameterλ = wavelength of light (approximately 550nm for green light)D = distance from aperture to projection surfaceFor my office setup with a projection distance around 4 meters:
d_optimal = √(2 × 550 × 10⁻⁹ × 4) = √(4.4 × 10⁻⁶) ≈ 2.1mm
This experiment is a clean way to explain f-stops versus t-stops.
F-stop is a geometric ratio of focal length to aperture diameter:
f-stop = f/D
For a pinhole camera, the "focal length" is the distance from aperture to projection surface. So with my 4-meter setup and 2mm aperture:
f-stop = 4000mm/2mm = f/2000
This looks absurdly high compared to camera lenses (usually f/1.4 to f/22), and it explains the long exposures in pinhole work.
T-stops account for actual light transmission through an optical system. F-stops are purely geometric, while t-stops measure how much light reaches the sensor or film after accounting for:
For a simple pinhole with no glass elements, the f-stop and t-stop are essentially identical. In that sense, camera obscura is a "perfect" optical system.
As aperture size decreases, we eventually hit the diffraction limit. At that point the wave nature of light becomes more significant than particle behavior.
The angular resolution limit due to diffraction is given by:
θ = 1.22λ/D
Where θ is the angular resolution in radians. For visible light (λ ≈ 550nm) and our 2mm aperture:
θ = 1.22 × 550 × 10⁻⁹ / 0.002 = 3.4 × 10⁻⁴ radians
At a 4-meter projection distance, this translates to a resolution limit of about 1.3mm. Fine details smaller than this get blurred by diffraction regardless of geometric optics.
One remarkable characteristic of pinhole optics is infinite depth of field. Lens-based systems keep one focus distance sharp, while camera obscura renders everything from a few meters to infinity with roughly equal sharpness (limited by diffraction).
This occurs because the aperture is so small that the circle of confusion for objects at different distances remains negligible compared to the diffraction-limited resolution.
The extreme f-stop of a camera obscura (f/2000 in my setup) makes the projected image very dim. Exposure time scales with the square of the f-stop, so compared to a camera lens at f/2:
Exposure Ratio = (f₂/f₁)² = (2000/2)² = 1,000,000
This means the camera obscura requires one million times longer exposure than a lens at f/2. That is why early pinhole photographs needed exposures measured in hours.
One surprising part of the projection was color. Without glass elements introducing chromatic aberration or cast, colors looked very true to life, maybe more accurate than what I usually see through complex modern lens designs.
Contrast was excellent too, with deep shadows and bright highlights faithfully reproduced. It shows how simple optical systems can preserve color accuracy, even when they give up light-gathering ability.
Creating a room-sized camera obscura required careful attention to light leaks.
I tested multiple aperture sizes by carefully punching holes in heavy-duty aluminum foil using needles of different diameters. The precision matters more than I expected. A 0.5mm change in aperture diameter noticeably affects image quality.
One of my favorite parts was watching the projected image change in real time. Cars moving past, people walking by, and clouds drifting across the sky all appeared live on my office wall, upside down and surreal.
This real-time behavior highlights something we often forget about photography. Cameras do not just capture moments. They continuously project the world onto the sensor, and the shutter only controls when we record that projection.
The camera obscura principle has been known since ancient times:
This experiment made me feel connected to that long lineage of optical investigation, from ancient philosophers to Renaissance artists to modern photographers.
I love when math predicts what I see in the room. The calculated aperture size produced the image quality the equations suggested. The geometric relationships held up. The physics matched the outcome.
It reinforces photography's foundation in science and mathematics. Every image we capture comes from photons behaving according to physical laws that have held steady for as long as we can observe.
This experiment reinforced several key optical principles:
Understanding camera obscura principles remains relevant for modern photography:
Even with all the physics and math in mind, watching my office wall display a live inverted view of the outside world never stopped feeling magical. That may be the most important lesson here. Technical knowledge can increase wonder.
Every photograph we take is a technical miracle. Photons travel at light speed, carry information about distant objects, pass through precision optics, and end up as digital data on a sensor. Camera obscura strips that process down to the fundamentals and makes the underlying magic easier to see.
You can view the complete documentation of this camera obscura experiment on my Flickr album. The images show various stages of setup, different aperture sizes, and the evolution of the projected image throughout the day.
This experiment reminded me why I fell in love with photography in the first place. It sits at the intersection of art and science, and technical understanding deepens my appreciation for beauty and for light itself.