LED walls for live events: pitch, processing and what to ask for

Pixel pitch is the distance in millimetres between the centres of adjacent LEDs, and it is the number that decides how close someone can stand before the image stops being an image and becomes a grid of dots. A 2.6mm wall has its LEDs 2.6mm apart. Smaller pitch means more pixels in the same physical area, a closer comfortable viewing distance, and a considerably higher price.

The trade rule of thumb is that the minimum comfortable viewing distance in metres is roughly the pitch in millimetres — a 3mm wall looks solid from about three metres back. It is approximate and it is conservative, but it is good enough to catch the expensive mistake, which is specifying a wall people will stand right next to at a pitch chosen for an audience thirty metres away.

Indoor conference and stage walls commonly sit in the low single millimetres. Outdoor and long-throw walls are coarser, because nobody gets close and brightness matters more than density.

LED walls are built from panels, so the size available is not continuous — it is whatever multiple of the panel dimension the supplier stocks. A common cabinet is 500mm square, which means widths and heights come in 500mm steps and a wall specified at 7.3m wide will actually be 7.5m or 7.0m.

The consequence people miss is that the pixel resolution of the finished wall is decided by the panel count and the pitch, and it will not be a standard broadcast resolution. A wall might end up 2816 by 1408, which is not 1080p and not 4K. Content built to a standard frame size will be scaled to fit, and scaled content on a fine-pitch wall is where softness and shimmer come from.

Ask for the wall's native pixel dimensions in writing, at the point you agree the size. Then build content to that number.

The processor takes a video signal and maps it onto the panels — scaling, colour, brightness, frame rate, and the mapping of which part of the picture goes to which cabinet. It is also where redundancy and signal switching live. It is the part of an LED package least visible to the client and most likely to determine whether the day goes well.

The questions worth asking a supplier are specific. What is the native resolution of the finished wall? What processor is being used and what are its inputs? Is there a redundant path if a cable or a sending card fails? What refresh rate does it run at, and has anyone confirmed that against the camera shutter if it is being filmed?

Build to the wall's native pixel dimensions, not to a standard video format, and hand over the native files rather than something already scaled. Every additional scale is a loss, and scaling to a non-integer factor on a fine-pitch wall is visible as shimmer on motion and softness on type.

Keep type larger than instinct suggests and avoid single-pixel lines and one-pixel strokes entirely — on an LED wall a one-pixel line either disappears or crawls. High contrast reads well; very dark content shows the panel grid and any brightness variation between cabinets.

If the wall is a scenic backdrop rather than a content surface, remember that it lights the stage. A white slide behind a presenter is a large soft light source pointed at the back of their head, and lighting will have to work around it.

Everything above is screen specification and applies whatever you plan with. EventBlok does not drive LED walls, does not process video and does not play back content.

Where it is relevant is the schedule that decides what should be on a screen at a given moment. Session titles, speaker names, times and room names are programme data, and when they are typed separately into a playback system they become another copy that drifts. EventBlok's signage views read the same record as the agenda and the showflow, so a renamed session or a moved slot is not a second job.

For the wider question of which systems should share a record and which should merely exchange data, see what event production software should connect.