Digital Signage Series – Video Walls

A Video Wall is an array of 2 or more monitors behaving as a single display. The lowest level of video wall is 1×2 matrix (1 row by 2 columns: just like a twin monitor setup with an extended desktop); however, 2×2 is more typical as this would result in a more typical aspect ratio of 16:9 or 4:3

At the other extreme, there are behemoths, such as the 70-foot installation for the Financial Times at New York’s Grand Central Terminal

or the 100-screen, 627 square foot, $570,000 behemoth at McCarren international airport in Las Vegas

While most video walls are box rectangular, there are products available that allow irregular shapes and even flexible options to wrap around columns, for example.

Christie’s MicroTiles are a common choice for irregular shapes:

The diagram below shows the diverse range of shapes that can be constructed with MicroTiles:

NanoLumens offers non-rectangular shapes as well as flexbile solutions using the NanoFlex wrap:

           

Any kind of display technology can be used for video wall applications. Typically. some form of Plasma or LCD panel is used, but projection can also be used for the ultimate in edgeless designs. The latter would requires considerable more setup, calibration and edge blending for a seamless display.

Bezel thickness is one of the most significant factors in achieving a seamless display. Although the human brain is capable of filtering out the bezels in a matrix of displays, skinny is still better. Today’s commercial displays offer bezels as this as 2.4mm

The Samsung UT-B Series features a bezel measuring just 2.4mm (0.1”) on the bottom and right, and 4.3mm (0.17”) on the top and left (for a total bezel width of 6.7mm (0.26”)).

The close-up below shows the difference between traditional displays and a modern display:

Calibration is a critical aspect of a video wall setup. This is one of the major factors that requires professional installation of even a 2-screen video wall. A poorly calibrated video wall does much more harm than good to product branding. The diagram below shows the difference between an uncalibrated vs. calibrated display matrix:

Physical installation is typically carried out with purpose built, as the individual displays have to be perfectly aligned in all three dimensions. Spending thousands of dollars extra on an ultra-thin is pointless if there will be gaps between the displays. Purpose built mounts offer very fine adjustments for perfect alignment of displays.

In addition to alignment, In addition to alignment control, maintenance requirements are more complex than single panel installations. Imagine having to replace one of the center panels on the FT wall (shown above). The diagram below shows a Premier Mounts solution that allows individual displays to be tilted out for maintenance.

This same model also features custom spacers that provide the correct offset for adjoining displays. This system effectively permits a matrix of any size and shape to be constructed.

The image below shows the mount for the McCarran airport video wall

Christie also provides their own solution for ease of maintenance with MicroTile products:

One of the key calculations required for video walls is resolution, or pixel pitch. At a distance of 1 foot from the display, a person with 20/20 vision can discern pixels when the pitch is great than 0.125mm (0.005”). So if the display will be 10-feet from the viewer, the pixel pitch should be less than 1.25mm (0.05”) in order for pixels to be indiscernible.

This is where the real complexity can come into play with video walls. There are several approaches to controlling a video wall:

1. Treat the display as a single monitor with the combined resolution equal to the sum of all the displays’ resolution
2. Scale the image to the combined resolution
3. Feed a single signal to the controller/displays and have it split the image between the individual displays
4. Feed different content to each controller and synchronize their display

Naturally, each method has its pros and cons.

This is probably the simplest, but scales poorly and typically requires a very expensive server. For example a 2×2 matrix comprising HD displays will have a combined resolution of 3840×2160. Rendering images at this resolution requires significant processing power. Clearly, this becomes less practical as the size of the video wall increases.

Many commercial monitors offer this capability. Displays are linked to each other in a video loop and the manufacturer’s software is used to identify, and configure the location of, each display. A single source is then fed to one of the displays, where it is scaled by a factor of 2 in each direction and spread across the display matrix. This is a very simple and effective method, where the displays are far enough away from the viewer that the pixels in the scaled image (regardless of the resolution of the hardware) do not become apparent.

Method 3 is also very processor intensive, a single very large image is fed to either a main controller or each sub controller where the image is either split or partially rendered based on the coordinates of that particular display. Clearly handling images of 10,000’s of pixels in each direction requires a great deal of processing power (whether it be centralized in a single controller or distributed across controllers located at each display—or group of displays). Hiperwall is a good example of this method.

This is an innovative approach employed by BrightSign in their BrightAuthor software and Synchronized Presentations. The designer is responsible for splitting up the content into sections that correspond to each display:

Each display has its own player (costing only a few hundred dollars). One is designated the master, and the top left image (for example) will be set to play on the master in the playlist. In the same playlist the top right, bottom left and bottom right images are set to play on their respective monitors and to synchronize with the master over the network.

When the master unit encounters a synchronize event, it sends commands to the other players and all players will start their playback in sync.

While it does require some extra work during content creation, it offers great flexibility. Not only can such a configuration be used as a video wall, but also as a multi-zone display network with synchronized content in each zone.