Imagine this scenario. You are hired as the video director for a tour, and they want to use HD. You want to display different images on seven LED screens, and you are using a media server that, at best, supports two layers of 1080p resolution media. What would you do to make this work? There is a way, and it involves creatively mapping out the raster. The desired result is for each screen to display different content. How in the world can we do that? Start with the raster. The raster could be described as the usable area of a projection or image display. It is simply that — an easily divisible area of pixels that can be broken up into as many areas as necessary. Once the areas are defined and carefully plotted out, video clips can be dropped into each area and scaled according to the available pixels in that specific area. When it is rendered, all of the movie clips will play simultaneously as long as they are at the same frame rate, and the result will be the equivalent of layers of multiple video images.
An HD 1080p raster refers to an area with 1920 x 1080 pixels of resolution and progressive scan (non-interlaced). Few media servers can play more than one or two layers of 1080p-rendered content currently, so the need for creatively breaking up the space can become an issue on large tours using multiple screens. But what goes into that usable area does not have to be all one image. It is possible to size your content accurately enough so that when the video signal is output to the display system, sections of the overall image are displayed on each screen instead of the entire image on every screen. This is a very creative use of the pixels being sent to the output.
In a standard pixel mapping system, a single image would be projected on the display(s) with specific areas of the image aligned to fit precisely on the display. By using pixel mapping software, grids are drawn and the grid represents the usable areas of the image that are to be projected using the pixels. The example above differs from the typical pixel mapping application in that it doesn’t just map certain areas of one image to any number of displays. Instead, multiple images are mapped across multiple displays. But the multiple images are rendered into one image, so it actually appears that there are multiple images on multiple layers being output by the media server when the reality is that it is still only specific areas of one “image” being mapped to multiple displays.
The secret to using pixel mapping in this fashion is in the rendering of the content. All of the content has to be carefully scaled so that once it is compiled into a single image, it doesn’t appear distorted or stretched. This is crucial because once the individual clip has been embedded into the master compilation, you won’t be able to edit its scale without affecting the entire image. The only option if you needed to adjust the scale of a specific piece of content would be to have each of the original individual pieces of content available and use one of them on another scaled layer on top of the original image. But that involves playing more than one layer of 1080p, which could cause the performance of the playback across the entire output to suffer, not to mention the very difficult task of keeping two layers completely in sync with each other while playing the same movie.
A huge advantage of using a media server is that most servers allow for live video to be input and mixed with the display. Since the media server was designed to put video at the controls of a DMX512 lighting console, that means that the LD has the ability to output a live video input at any moment.
You might be thinking, “Yes, I know this.” But have you thought about using live video images in your pixel mapping? It is entirely possible. Using the pixel map, you simply need to create a layer on top of the raster image for each of the usable “screen” areas in the raster and scale it to fit each of the screen areas in the pixel map. By saving these scaled layers as palettes or presets, this provides the ability to trigger live video on any screen in the multiple screen system at the touch of a button. (Okay, two buttons.)
These are just a few of the advantages of rendering a single piece of content from multiple clips if the intention is to play it back at HD resolution. As most professional media servers begin to adapt to HD content, the need for multiple layers of images requires the programmer to be creative in how those images are physically managed. Any time a single piece of content can be used for display, it makes the workload easier on the media server, not to mention the programmer. Also, by using a single piece of content instead of many layers of images playing simultaneously means that the playback performance will be optimum and as the client expects. And everyone likes a happy client.
