DMZ

DMZ is a 2022 HBO Max miniseries following Alma Ortega, a New York City medic searching for her son in a demilitarized Manhattan. Set during the Second American Civil War, the series sees Alma become a symbol of hope on the island she once fled — her son having wandered off during their evacuation from Manhattan.

SCANable was responsible for 3D laser scanning and LiDAR capture of various sets for the miniseries, providing the visual effects team with accurate environment data used to extend and enhance the show’s war-torn New York in CG.

Damage is a modification of something that exists

A ruined city is not designed from nothing. It is a real place with things taken away from it, and the audience judges the result against a version of that place they already carry in their heads.

That makes destruction work a subtractive problem rather than an inventive one. A collapsed façade has to expose the structure that was genuinely behind it. A blown-out window has to sit in a wall of the correct thickness. Debris has to fall where the geometry says it would fall.

All of that depends on having the intact version measured first. Terrestrial laser scanning supplies the before — the building as built, at true dimension — and every stage of damage is then derived from it rather than invented alongside it.

The seam is a parallax problem

The join between a practical set and its digital continuation almost never fails on texture or colour. Those are correctable in the composite, and a competent team corrects them.

It fails on parallax. As the camera moves, the practical foreground and the CG extension shift relative to each other at rates determined by their distance from the lens — and if the extension was built at even a slightly wrong depth or scale, those rates disagree.

The eye is exceptionally good at detecting that disagreement and terrible at naming it. Viewers do not say the extension is at the wrong depth; they say the shot looks fake.

Building the extension outward from measured geometry means it occupies the depth it claims to occupy. The parallax resolves because it was never wrong.

What a scanner records that a photograph cannot

Photographic reference is essential and always insufficient. A still tells you how a surface looked from one position under one lighting condition. It does not tell you how far away anything was.

A scan records position first and appearance second. It captures the recesses, the returns, and the occluded faces that a camera never saw, and it does so in units that mean something outside the frame.

That is why environment capture is treated as infrastructure rather than as reference-gathering. Almost every downstream decision — placement, tracking, lighting, layout, and the render-ready geometry built on top — resolves back to a measurement somebody took on the day.

See more episodic television work, or review what we offer.