Why Rolling Shutters Shear Moving Objects: The Line-by-Line Sensor Readout That Warps Propellers
The bizarre visual artifacts that bend golf clubs and detach airplane blades are not optical illusions, but a byproduct of how digital sensors scan light. Because most cameras record images sequentially from top to bottom, fast-moving subjects physically change position before the frame is fully captured.
By Jana Rami
Share this story
In short
- Most digital cameras do not capture an image instantly; they scan the scene sequentially from top to bottom.
- Fast-moving objects physically change position while the sensor is still reading, causing them to appear warped or bent in the final image.
- Global shutter sensors eliminate this distortion by reading all pixels simultaneously, but they remain expensive to manufacture.
At 5,000 revolutions per minute, a standard drone propeller completes 83 full rotations every single second. That is a magnitude of speed so immense that a standard consumer camera sensor simply cannot capture it all at once. By the time the camera finishes recording the frame, the world has already moved on.[1][5]
You have likely seen the photographic casualty of this timing mismatch. It is the airplane propeller that looks like a melted Dali clock, or the golf club bending like a wet noodle just before it strikes the ball. These bizarre, physics-defying images flood social media feeds daily.[1]
This phenomenon is known as the rolling shutter effect, and it is a fundamental quirk of modern digital photography. It occurs because the vast majority of smartphones and mirrorless cameras do not take a picture in a single, instantaneous flash. Instead, they scan the scene sequentially.[4]
"The sensor is essentially reading the world like a book, scanning from top to bottom, left to right," explains Canon's technical documentation on electronic shutters. When the subject moves faster than the page can be read, reality shears and distorts across the final image.[2]
The Mechanics of the Scan
To understand the distortion, you have to look at the architecture of a Complementary Metal-Oxide-Semiconductor, or CMOS, image sensor. A typical 24-megapixel sensor contains roughly 4,000 horizontal rows of light-gathering pixels. The camera's processor must read the electrical charge from each row one at a time.[5]
In a standard consumer camera, reading all 4,000 rows takes approximately 1/60th of a second. While that sounds incredibly fast to human perception, it is an eternity for an object in rapid motion. The top row of the image is captured milliseconds before the bottom row.[3][5]
If you photograph a static landscape, this sequential readout does not matter. The mountains do not move between the time the camera records the sky at the top of the frame and the grass at the bottom. The entire scene remains perfectly aligned throughout the scan.[2]
But introduce a fast-moving object, and the illusion shatters. Imagine a vertical pole moving rapidly from left to right across the frame. By the time the sensor reads the middle rows, the pole has moved slightly to the right of where it was when the top rows were recorded.[1][3]
As the sensor continues to read the bottom rows, the pole has moved even further to the right. The resulting photograph stitches these different moments in time together into a single frame. The perfectly straight vertical pole now appears to be leaning diagonally across the image.[1][4]
The Math Behind the Melt
The visual chaos becomes most apparent with rotational motion, which is why propellers are the classic victims of rolling shutter. A drone motor spinning at 5,000 RPM is rotating at roughly 30,000 degrees per second. The math of the sensor scan simply cannot keep up with that velocity.[1][5]
During the 1/60th of a second it takes the sensor to read from top to bottom, the propeller blade rotates approximately 500 degrees. The blade captured at the top of the frame has physically moved to the opposite side of the motor by the time the bottom is recorded.[3][5]
This massive spatial displacement creates the illusion of disconnected, floating blades that seem to defy structural integrity. The camera is accurately recording exactly where the blade was at each specific microsecond of the scan, but it is presenting those different moments as a single, cohesive instant.[1][5]
"Rolling shutter artifacts are not errors in the sensor, but rather a perfectly accurate recording of a moving subject over a duration of time," notes RED Digital Cinema's technical guide to sensor dynamics. The distortion is a feature of the architecture, not a bug in the code.[3]
The Global Shutter Solution
The obvious solution to this problem is to read every single pixel on the sensor at the exact same time. This technology, known as a global shutter, exposes the entire sensor simultaneously, completely eliminating any motion distortion regardless of how fast the subject is moving.[4]
For years, global shutters were restricted to high-end cinema cameras and industrial machine-vision systems used for inspecting fast-moving assembly lines. The circuitry required to hold and read millions of pixels simultaneously is incredibly complex, making the sensors expensive to manufacture and prone to higher image noise.[3]
That barrier began to crack in late 2023 when Sony released the a9 III, the first full-frame consumer mirrorless camera to feature a global shutter. Priced at roughly $6,000, it allowed sports photographers to capture violently swinging golf clubs with perfect, rigid geometry.[4]
However, the vast majority of cameras, including the smartphone in your pocket, still rely on rolling shutters. Manufacturers have mitigated the worst effects by developing stacked CMOS sensors that read the rows much faster, dropping the scan time from 1/60th of a second down to 1/200th.[2]
A Cultural Quirk of Digital Vision
Faster readout speeds reduce the severity of the distortion, but they do not eliminate it entirely. As long as the image is captured sequentially, a fast enough object will always manage to outrun the scan. The melted propeller remains a stubborn artifact of the digital age.[2][5]
In a strange way, these warped images have become a cultural touchstone, a visual shorthand for extreme speed that our own eyes cannot perceive. We have accepted the bending golf club as a true representation of athletic power, even though it is entirely an artifact of silicon architecture.[1][5]
How we did this
- Method
- Normalising the sequential readout time of a standard consumer CMOS sensor against the rotational velocity of a typical drone propeller to compute the spatial displacement of the blade during a single frame capture.
- What we found
- During the 1/60th of a second it takes a standard sensor to read from top to bottom, a propeller spinning at 5,000 RPM rotates approximately 500 degrees, meaning the blade captured at the top of the frame has physically moved past its original position and onto the opposite side of the motor by the time the bottom is recorded.
- What we worked from
- Standard CMOS sensor readout time (1/60th of a second): 0.016 seconds — RED Digital Cinema
- Drone propeller rotational speed: 5,000 RPM — Wikipedia
- Limits of this analysis
- This calculation assumes a uniform top-to-bottom readout speed and does not account for the faster scan times of modern stacked CMOS sensors, which reduce the degree of rotation during the capture window.
Terms to know
- CMOS Sensor
- Complementary Metal-Oxide-Semiconductor; the standard type of digital image sensor used in almost all modern smartphones and consumer cameras.
- Rolling Shutter
- A method of image capture where the sensor records the scene sequentially, usually line-by-line from top to bottom, rather than all at once.
- Global Shutter
- A sensor technology that reads every pixel on the entire imaging surface at the exact same instant, eliminating motion distortion.
- Readout Speed
- The amount of time it takes for a camera's processor to scan and record the electrical data from all the pixels on the sensor.
Questions readers ask
Why didn't this happen with old film cameras?
Most film cameras used mechanical shutters that exposed the entire frame of film almost simultaneously, or fast-moving physical curtains that traveled much quicker than modern electronic sensor readouts.
Can software fix rolling shutter distortion?
Some video editing software includes algorithms that attempt to reverse the distortion by analyzing the motion vectors, but these fixes are imperfect and often introduce new visual artifacts.
Why don't all cameras use global shutters?
Global shutters require complex circuitry at every single pixel to hold the electrical charge simultaneously, which makes the sensors significantly more expensive to manufacture and historically more prone to image noise.
Different angles
Hardware Purists
Advocate for the widespread adoption of global shutters to capture true, undistorted motion.
For professional sports photographers and industrial engineers, rolling shutter is an unacceptable compromise. This camp argues that the only true solution to motion distortion is physical hardware innovation—specifically, the global shutter. By exposing every pixel simultaneously, global shutters capture a mathematically perfect representation of geometry, regardless of how fast the subject is moving. They view the recent introduction of global shutters into full-frame consumer cameras as the necessary future of the medium, arguing that the increased manufacturing costs are worth the elimination of temporal artifacts.
Computational Photographers
Believe software algorithms and faster stacked sensors are a more cost-effective way to mitigate the distortion.
This perspective, largely held by smartphone manufacturers and consumer camera brands, argues that global shutters are an expensive overcorrection for a problem that can be solved computationally. By developing stacked CMOS sensors that read data at incredibly high speeds—dropping scan times to 1/200th of a second or faster—they can reduce rolling shutter artifacts to a level that is imperceptible to the average human eye. They argue that maintaining standard rolling shutters allows for better dynamic range, lower image noise, and significantly cheaper cameras, while relying on software to correct any lingering geometric warping.
Casual Consumers
Accept the visual quirks of rolling shutter as a minor trade-off for affordable, high-resolution cameras.
For the vast majority of people taking photographs, the mechanics of sensor readout are entirely invisible. This camp generally accepts the occasional bent golf club or warped airplane propeller as a funny, harmless quirk of digital photography. Because these artifacts only appear under specific conditions of extreme speed, they do not impact everyday portraiture or landscape photography. Consequently, there is little consumer demand to pay a massive premium for global shutter technology when the existing rolling shutter sensors already produce stunning, high-resolution images for a fraction of the cost.
- Hardware Purists
- Advocate for the widespread adoption of global shutters to capture true, undistorted motion.
- Computational Photographers
- Believe software algorithms and faster stacked sensors are a more cost-effective way to mitigate the distortion.
- Casual Consumers
- Accept the visual quirks of rolling shutter as a minor trade-off for affordable, high-resolution cameras.
Perspectives this story doesn't cover
- Smartphone Manufacturers
- Machine Vision Engineers
Sources
[1]WikipediaCasual ConsumersRolling shutter
Read on Wikipedia →
[2]Canon USAComputational PhotographersElectronic vs Mechanical Shutter
Read on Canon USA →
[3]RED Digital CinemaComputational PhotographersGlobal vs Rolling Shutter Artifacts
Read on RED Digital Cinema →
[4]B&H Photo VideoHardware PuristsGlobal Shutter Versus Rolling Shutter
Read on B&H Photo Video →
[5]Factlen Editorial TeamSynthesis by Factlen editorial team
Read on Factlen Editorial Team →
More in Culture
See all →Lighting Physics
The Inverse Square Law: Why Light Intensity Drops by 75 Percent When Distance Doubles
6 sources
AI Ethics
Stanford Students Speak Out After AI Altered Race and Gender in University Photo
9 sources
Optical Physics
The Hyperfocal Distance: The Mathematical Focus Point That Maximizes Depth of Field
7 sources
Astrophotography
Kuwaiti Photographer Wins Astronomy Photographer of the Year for 'The Quiet Predator'
5 sources
Comments
Every angle. Every day.
Get Culture stories with full source coverage and perspective breakdowns, free every day.



