Guides / cases / source records
Search by the problem, the tool, or the department. Practical guides explain what to do; archive records establish where a technique or claim came from.
132 pieces
Epic and Bullitt’s 2021 production test is a useful lesson in stage preparation—not a promise that every shoot can finish in four days.
Mold3D’s animated short makes a useful distinction: real-time iteration and final delivery are related, but they are not the same job.
Start with the production language, then follow the workflows. Keep the 2019 and 2021 books beside—not in place of—current documentation.
A practical introduction to nDisplay, the camera frustum, and the point where a screen becomes part of a calibrated system.
A tracker tells you where a camera is. It does not, by itself, describe the image formed by its lens.
Matching numbers are not proof that the whole image pipeline is synchronized. Test timing and delay as separate questions.
Use the virtual location to settle camera, blocking, and set questions before the stage becomes the meeting room.
Understand the controller/listener relationship, then build a small readiness routine around the actual production configuration.
A convincing wall image is only part of the result. Document the whole color path before judging a calibration.
Green screen can give the crew a shared picture without requiring every compositing decision to be finished on set.
Before buying a larger system, make a small test that another person can repeat and judge.
Move from an attractive workstation scene to a tested, recoverable package with a clear owner and known limits.
A per-shot decision method that makes the image, the schedule, and the post consequence visible before a stage is booked.
A quote comparison worksheet for finding incompatible assumptions before the low number becomes the expensive one.
A booking shape that reserves time for the work the shooting day cannot safely absorb.
A production-facing environment brief that separates creative intent, camera requirements, and stage assumptions.
An operational inventory for identifying what an environment contains, where it came from, and which release questions remain open.
A test plan for discovering when two cameras want incompatible frustums, lighting, or wall real estate.
A camera-test matrix that treats moiré as an image behavior across camera settings, distance, focus, and wall content.
A practical way to prevent an LED wall from being asked to provide every kind of light in a shot.
A shot-first checklist for driving work that uses photographed plates on a wall without pretending the volume replaces road-unit planning.
A concrete deliverable package for making camera media, stage decisions, and color/calibration context usable after the shoot.
Recording a performance is not the same as knowing which take, camera state, and technical conditions produced a usable result.
Use a shared Unreal session for immediate collaboration, but make durable project state recoverable through a revision workflow.
Build a render package from an agreed image contract, then prove a small section before committing the full queue.
Frame rate is the reciprocal of time: turn the stage target into milliseconds, then profile the actual heaviest node configuration.
Send a small, declared scene package with layers, dependencies, variants, and acceptance views instead of treating a .usd extension as a guarantee.
A source can be present while the wrong role, lens stream, timestamp behavior, or mapping is still wrong for the shot.
Map the usable volume, occlusions, and failure boundaries with the real camera configuration—not only the tracker’s central sweet spot.
A restart is successful only when the team can prove the recovered scene, control state, and stage image match the named baseline.
The relevant question is not whether a background looks detailed; it is how its geometry behaves when camera position, focus, foreground, and occlusion change.
A color register makes each technical conversion and creative look traceable without turning a display calibration or a grade into an unnamed mystery setting.
ASC's own accounts of Strangers on a Train and Vertigo show rear-screen process photography as a decades-long practice, not one studio's invention.
ASC's 1968 account and the Academy's own award record show what front projection did for 2001: A Space Odyssey, and what it did not.
ASC's 1977 account and ILM's own history describe what the Dykstraflex did for Star Wars, and why repeatability mattered more than speed.
Blackmagic's product page and a trade obituary for Petro Vlahos date Ultimatte's Oscar recognition and what its process added to older bluescreen keying.
A 1991 Los Angeles Times account and the underlying front-projection patent describe what Introvision offered producers, and why blue screen displaced it.
ASC's 1982 account and the Academy's 1983 nominee list show which Tron sequences were CGI, and that the film itself was not nominated for effects.
ASC's 1993 account and the Academy's Visual Effects award record show how ILM integrated digital dinosaurs, and which department built the models first.
ASC's 1999 account and the Academy's 2001 award citation describe how The Matrix's still-camera rig simulated an impossible camera move.
A trade profile and a crew interview describe the render-farm scale and workflow Square Pictures needed to build a fully CG feature in 2001.
Two ASC accounts, thirteen years apart, describe Simulcam's real-time composite for Avatar and credit its development to different people.
Framestore's case study and the Academy's own award record describe a rig that lit a stationary actor, not a projected background.
SMPTE's own publications pages describe a peer-reviewed journal and digital library that predate the current LED-volume cycle by decades.
ILM's own pages date preliminary StageCraft testing to 2018 and the first full-scale volume to The Mandalorian's November 2019 debut.
ROE Visual's own product pages describe pixel pitch, refresh rate, and finish choices made for what a camera records, not a viewer.
A September 2023 Sony press release positions Crystal LED VERONA as purpose-built for ICVFX, distinct from Crystal LED's general display line.
Brompton's own product and application pages describe processing features built for what a lens sees, not what an audience sees.
disguise's own documentation describes a bidirectional protocol moving camera-tracking and render data between its media servers and named engines.
Megapixel VR's own Helios product page states calibration and color-management features for camera-facing LED walls, labeled here as vendor claims.
Samsung's own display page and a dated engineer interview show what The Wall's pixel-pitch and refresh-rate claims cover, and what they do not.
Absen's own PR-series page sets pixel pitch, frame rate and refresh-rate figures for virtual production, distinct from its rental and touring lines.
Unreal Engine's own ICVFX documentation defers to the LED manufacturer for panel specs while naming specific Live Link tracking partners.
Lux Machina's own case pages describe systems integration for LED volumes, distinct from the panel and engine makers whose gear it assembles.
ROE Visual's own product pages set Ruby's finer pixel pitch and refresh rate against Black Pearl's broader installed base, without a numeric comparison.
A 6 January 2021 Sony press release, not 2022, introduced the B-series for virtual sets and studio backdrops, developed with Sony Pictures.
Brompton's own site keeps the M2 in its legacy lineup at up to 1080p60, a smaller scale than the current 4K SX40 flagship.
PRG's own service pages describe design, sourced panels and on-set crew for LED backdrops, distinct from a panel maker's product page.
ARwall's own store publishes footprint, pitch and price for stage bundles from about ten feet wide up to enterprise scale.
Epic's own documentation shows nDisplay solves frame synchronization across networked render nodes, not virtual production as a whole.
Epic's quick start documents a twelve-step ICVFX setup, but calibration and coordinate work it references live in separate documents.
The 2019 field guide's own table of contents shows it organized virtual production by department rather than by tool.
Epic's 2021 announcement and the guide itself show Volume 2 added remote collaboration, expanded visualization and in-camera VFX chapters.
Unreal's licensing page and EULA together show a seat-based fee, not a royalty, applies once a company's annual revenue passes one million dollars.
Unity's own pricing page confirms a 5 percent increase for Pro and Enterprise seats effective 12 January 2026, and nothing beyond those tiers.
Unity's 2021 announcements document an agreement to acquire Weta Digital's tools and engineers, not a confirmed shipped integration.
disguise's own materials describe Designer as the authoring and control layer that sequences a show, distinct from the RenderStream protocol.
Pixotope's own pages describe a platform built first for live broadcast, with a film-making claim its product page does not itself detail.
Aximmetry's own published tiers show a port-count-based pricing ladder from a free edition to a broadcast-and-film tier.
Notch's own documentation frames it as motion-graphics and VFX software, not a camera-tracking or ICVFX rendering layer.
Zero Density's own materials frame Reality 5 as a broadcast graphics platform whose XR toolset only partly overlaps film ICVFX.
Epic's 2019 announcement states it acquired Quixel and made the Megascans library free for use in Unreal Engine.
Epic's May 2021 early-access announcement states what Nanite and Lumen do and that the build was not production-ready.
Brainstorm's own documentation frames InfinitySet as broadcast-first, with vendor-stated crossover into film and drama sets.
Ross Video's own documentation states Voyager renders broadcast AR graphics on Unreal Engine, with a named LED-workflow link.
SMPTE's RIS-OSVP documentation defines what OpenTrackIO transmits and why one open format serves a mixed-vendor volume.
Mo-Sys's own product page states how StarTracker Max tracks a camera optically and what performance figures it discloses.
Stype's own documentation states how RedSpy's infrared tracking works and where the company positions it across broadcast and film.
Vicon's own materials frame camera tracking for virtual production as an application of its motion-capture business, not a separate line.
OptiTrack's virtual-production page states a Prime-camera volume plus a wired CinePuck accessory, not a separate tracking product line.
HTC's own VIVE Mars pages state base-station coverage and tracker counts, but not an independent accuracy figure.
ARRI's own pages state LDS-2 drops calibration and speeds data, with a named, limited list of compatible cameras.
Cooke's /i Technology captures per-frame focal length, squeeze ratio and pupil data, but does not track a camera.
ZEISS states its lens metadata feeds Nuke, After Effects and ARRI hand units, not a named render engine.
Netflix's OpenVPCal repository, created in November 2023, calibrates LED walls using the production camera itself.
MRMC states its Bolt Cinebot rigs sync with Unreal Engine for repeatable moves, distinct from a camera-tracking puck.
BBC's 1997 paper set the camera-tracking accuracy targets that became the licensed free-d system.
ILM's own account of its Virtual Art Department shows environment design shifting into a real-time pipeline before a StageCraft volume shoot.
Epic Games' own program page describes a 30-day virtual-production track for film and VFX professionals, distinct from its separate games fellowship.
The National Film and Television School's own course pages describe a full-time MA and a part-time certificate covering LED-volume and real-time workflows.
SAG-AFTRA's own 2023 contract materials set separate consent and pay rules for employment-based, independently created and background-actor digital replicas.
IATSE's own release naming a 'Sr. Virtual Production Tech' role at a 2024 election shows classification arriving through first contracts, not a settled schedule.
Epic's Virtual Production Field Guide defines the brain bar as the nerve center of a volume shoot and lists who typically staffs it.
An American Cinematographer test documents that LED-wall color calibration against a chosen camera must run through the whole shoot, not only in prep.
The Visual Effects Society's own site describes free VP education and webinars built with its committees, offered as guidance rather than a binding standard.
The ASC's Motion Imaging Technology Council names a chaired Virtual Production Committee and working groups, with published reports but no binding standard.
ETC's own project and publications pages describe its virtual-production research scope, not an industry-wide survey.
Epic's own navigation routes prospective schools to a bronze-to-gold Academic and Training Partner directory, not a virtual-production-specific list.
Trade coverage and Trilith's own site describe the same 80-by-90-by-30-foot volume under two different names, four years apart.
Dimension's own pages describe its London volume's wild walls and retractable ceiling without publishing an opening date or measurements.
Vū's own pages list 29 cities and frame financing and software, not a single measured facility, as the company's product.
Trade reporting corroborated by an SPE spokesperson describes Pixomondo's closure, bracketed by its 2022 acquisition footprint.
Epic's own FAQ sets a $5,000-to-$150,000 grant range for film, TV and tooling projects, with no published running total.
The Commission's 2025 stage report tracks carbon and power demand, including virtual production's rising loads, not overall UK stage capacity.
A 2021 law, not Program 3.0 or 4.0, is the California Film Commission's own construction incentive for certified stages.
HMRC's own guidance dates AVEC from 1 January 2024 and applies it to all qualifying production, not virtual production specifically.
Cinespace Toronto's own vendor list names two on-lot virtual-production specialists rather than an in-house LED volume.
SMPTE's own standards library lists RDD 18 and RDD 55 as camera-metadata documents distinct from its newer OpenTrackIO tracking work.
The EBU's own publications catalog vendor and member seminar talks on virtual production, not a binding broadcast standard.
ASC's own history dates its Color Decision List's Emmy Award to 24 October 2012 and frames it as a primary-grading standard.
The Academy's own ACES repository dates ACES 2.0's first release to 5 September 2024, with new rendering transforms as the headline change.
AOUSD's v26.03 release adds a first-class Gaussian Splat schema, framed by its own notes as testable, not production-certified.
The Academy's 2024 Scientific and Technical Awards cited Atlas and USD for pipeline consistency from virtual production through lighting.
Epic Games' December 2021 announcement frames The Matrix Awakens as a UE5 rendering demonstration, not a virtual-production shoot.
OCIO's own documentation names specific version thresholds for native color-pipeline support in Disguise Designer and Unreal Engine.
The ASC's StEM3-VP evaluation suite, hosted by the Academy Software Foundation, lets DPs check a volume's calibration before filming.
VES's April 2026 On-Set VFX Data Guide standardizes data-sharing terms; its VP hub shows a member survey underway, not a published report.
BSI's own scope text limits BS 7909 to temporary low-voltage power for entertainment events, distinct from its construction-site counterpart.
SMPTE's Motion Imaging Journal lists ongoing peer-reviewed work on 3D rendering, distinct from its separate On-Set Virtual Production program.
A tele-immersive research paper reports 81 ms end-to-end latency to a 3D LED wall, checked against ITU-T G.114, not a film ICVFX benchmark.
Weta FX and Lightstorm crew describe swapping infrared motion-capture light for near-ultraviolet to track performers below the surface.
ROE Visual's own product page positions Diamond as a lightweight, cost-effective panel for events and XR stages, not Black Pearl's camera-first tier.
Samsung's Onyx page and its dated 2018 launch announcement describe a DCI-certified theatre screen, separate from Samsung's VP display line.
FARO's own pages state its scanners digitize sets for VFX previs, with accuracy figures published per model, not per use case.
Lightcraft's archived documentation states Previzion tracks, reads lens metadata and renders with its own built-in engine.
ADG's Production Design Initiative teaches digital-pipeline skills inside traditional art-department mentorship, not a dedicated VAD curriculum.
ZEISS's own CinCraft pages describe Scenario's tracking hardware and lens support without a comparative test against named rivals.
Christie's own solution page frames laser-projection virtual production as lower-cost than LED volumes, without a named case study.
Antilatency's own pages state tracker weight, precision and integrations, naming Vu Technologies as a client without a case study.
ISO's own standard page frames 45001 as a generic OH&S management system, not a rule written for LED-volume construction.
NFPA's own development page scopes 70E to electrical-hazard work practices, not entertainment rigging, even as stage power loads rise.
SIGGRAPH 2026's Real-Time Live! lists filmmaking technology as an eligible category, though 2026's named winners were not VP-focused.
Try fewer words, another topic, or clear the filters.
