DOF/MOIRÉ

Chaos AV

Work out before the shoot whether the LED wall will moiré on camera — and how much defocus it takes to kill the pattern. Drag the focus plane, the subject and the wall in the side view.

Examples

1 Camera

Sensor / camera

Width · height (mm) · photosites

The photosite count drives the moiré maths: a 2/3″ HD head and a 2/3″ UHD head behave completely differently at the same position.

Low-pass filter · recording format

Circle of confusion

2 Lens

Focal length 50 mm

—

Aperture f/2.8

3 Scene

Subject distance —

Focus distance —

Lock to subject

Unlock to focus ahead of the subject — the wall goes softer while the subject stays inside the depth of field.

LED wall distance —

4 LED wall

Pixel pitch

Wall width m

Framing only. Moiré depends on pitch, distance and focal length — not on how wide the wall is.

Side view

Drag focus, subject and wall

Depth of field Focus plane Subject LED wall

Near limit: –
Far limit: –
Depth of field: –
Hyperfocal: –
Subject: –
Blur on the wall: –
Frame on the wall: –
Wall coverage: –

Low risk –

Sensor pixels per LED pixel: –
Pattern in frame: –
Remaining grid contrast: –
LED pixel on the sensor: –

–

Simulated camera image

1:1 crop

Computed, not drawn: LED grid → defocus → low-pass filter → sampling by the photosites. One image dot is one sensor pixel. Debayering is not modelled, so real moiré usually carries more colour.

Depth of field: hyperfocal H = f² / (N · c) + f, near limit s(H−f) / (H+s−2f), far limit s(H−f) / (H−s), blur disc of an object at d: B = f² · |d−s| / (N · d · (s−f)), with s the focus distance.

Moiré: the LED grid images at p · f / (d−f) on the sensor and is compared against the photosite pitch. Above the Nyquist limit it folds back into a beat pattern. Its contrast is cut by defocus (disc MTF |2·J₁(πx)/πx|), by the optical low-pass filter and by downscaling to the recording format. Planning aid — it does not replace a camera test on site.