Video is the largest and most multi-layered discipline in corporate live event production. LML staffs every video position, from the Engineer in Charge who oversees the entire video village, to specialist operators for graphics, playback, switching, projection, and cameras.
Accountable for the integrity of the entire video output: signal flow, team direction, and primary client liaison.
View role →The EIC's primary operational partner. Owns a dedicated seat and holds technical execution when the EIC is managing the room.
View role →Fills a defined seat (graphics station, playback machine, record position, or confidence monitoring) under EIC/V1 direction.
View role →Manages the pre-recorded content stack (packages, b-roll, animations) executing starts and stops on cue with the show caller.
View role →Client-facing presentation fluency, advancing decks, executing last-minute swaps, and troubleshooting without breaking composure.
View role →Runs the production switcher at 4K/UHD: clean transitions across main screen, confidence, IMAG, and broadcast simultaneously.
View role →Selects and cuts between live camera sources at the video director's call, a dedicated seat on broadcast and IMAG-heavy shows.
View role →Setup, alignment, and operation of projection systems: single units, stacked for brightness, or multi-projector blended arrays.
View role →Captures the show to file (ISO program output, individual camera feeds, or both) and confirms file integrity at each act break.
View role →Owns framing, focus, and exposure at their position, delivering a consistent, on-spec picture to the switcher throughout the run.
View role →Controls PTZ heads in real time, one op managing multiple angles from truss-mounted or inaccessible positions.
View role →Manages camera shading from a remote control panel, matching iris, exposure, white balance, and color across all positions.
View role →Operates a Steadicam rig for smooth, stabilized movement, following talent and executing tracking shots no locked camera can achieve.
View role →The roaming camera position: handheld, untethered, capturing b-roll, audience reactions, and cutaways during the live show.
View role →Nobody in a ballroom knows what a frame sync is. They know when the presenter's mouth and voice drift apart, and they know when the person on the screen waves a beat after the person on stage. Both come from the same place: the time it takes a picture to travel from the camera lens to the display, measured in frames, with every device in the path adding its own. A short, genlocked, format-matched chain keeps that number small. Audio delayed to match keeps sync honest for whatever remains. Build a path below and see what the room sees.
Milliseconds added by each device at the selected frame rate. Hover a bar for both values.
| Stage | This design | Conventional |
|---|
A camera needs most of a frame to read its sensor and build a picture. A production switcher holds a frame to cut cleanly. A scaler holds one or two to resize. A presentation switcher, an LED processor, a projector: each takes its share. None of them is slow on its own. Together they are the delay the audience sees between the person on stage and the person on the wall.
Frames are the unit, but milliseconds are what people perceive. At 59.94 a frame is 16.7 ms; at 29.97 it is 33.4. The same chain of boxes is twice as late in a 29.97 room, which is why the house format is the first decision the Engineer in Charge makes, not the last.
Audio through a console arrives in a few milliseconds. Video through the chain above arrives a hundred or more later, so without correction the voice leads the lips. Viewers detect audio leading by about 45 ms and object past 90. That is three to five frames at 59.94: an ordinary IMAG path, uncorrected, is already past the line.
The fix is in the audio department. The A1 measures the video path and delays program audio to match, for the house feed and separately for the webcast, whose encoder adds more. What the delay cannot fix is the picture itself: a late screen still lags a live presenter's gestures, which is why the path has to be short first and synced second.
Compact design in video means fewer conversions. One house format, every source conformed to it at the source, and every device locked to house reference. A locked frame sync passes video straight through instead of buffering a frame to re-time it. A scaler that never changes format can run in its low-latency mode. The presentation switcher cuts without waiting for sources to line up.
In practice: keep the IMAG path to camera, switcher, processor, wall. Scale laptops before the switcher, not after it. Put the projector in its low-latency mode and confirm it with a clapper at tech. Platforms like Ross Ultrix and Analog Way Aquilon exist to do routing, sync, and scaling in one locked frame rather than a stack of boxes.
About four frames at 59.94. The wall feels live.
Visible if you look for it on big gestures. Audio must be delayed.
The screen clearly trails the stage. Common with unlocked chains.
Applause lands twice. Presenters start watching the wall.
Six or more frames at 29.97. Rebuild the path.
Model notes: stage latencies are typical figures in frames. Camera and camera control unit 1.0; production switcher 1.0; each conversion pass 1.5 unlocked or 0.5 genlocked; presentation switcher 1.5 unlocked or 1.0 genlocked; LED processor and panels 1.5; projector 2.5. Milliseconds are frames times the frame period at the selected rate. Lip-sync thresholds follow ITU-R BT.1359: audio leading video is detectable past 45 ms and objectionable past 90 ms. Webcast encoding and distribution add seconds on top of all of this, equally to audio and video, and are not shown. Actual figures vary by model and firmware; LML measures the real path at tech.
Tell us your show date, positions, and venue. We'll staff it.
Submit a Booking Inquiry