What is involved with creating a seamless Edge Blend?
Most people think that if you have two projectors and send the left half of an image to one projector and the right half of an image to the other projector, that you can just butt the two projected images next to each other to create a seamless image. In practice, however, this doesn't work very well because it is nearly impossible to line the projectors up side by side to form a seamless image. The seams are visible, and create a distraction for the audience who can see the lines between the two projectors.
So because of this obstacle a process called "edge blending" was created. It actually occurs in two steps. The first is called "data doubling" and consists of duplicating content on the projectors used. Using two projectors as an example, the right edge of the left projector shares the same information with the left edge of the right projector, like so:
Obviously, the area in which the two projected images overlap is going to be brighter than the areas projected uniquely be each projector, which will create a brighter "hot spot" at the overlap point. To compensate for this, we introduce the second step of the Edge Blending process, which is called "feathering the edge". In this manner, the data-doubled portion of the left projected image is manipulated to gradually fade to black, while the data-doubled portion of the right projected image is gradually faded in from black. If you were to graph the light output in this manner, the left and right projected edges would look like so:
Such that where they overlap, it would appear like this:
Thus reducing any "hot spotting". Here's a diagram of what we're doing with a 3 screen edge blend - the first image shows an example of 3 full projectors at 800x600 resolution each and the second shows how they are overlapped:
The results of this are better visualized with live imagery. Below you see three projected images with the data-doubled and blended areas outlined for each projector. Next, the image is broken apart to see how they were joined to create the complete seamless image.
What is required to create an edge blend?
In the past, the process of creating this kind of edge blend required the use of special equipment, such as video scalers or multi-channel video processors. At the low end, this requires an additional investment of around $1,500 per projector. At the high end, you could spend $50,000 or more depending on the quality of the image you are trying to create.
The ProPresenter Edge Blending Module makes use of the Matrox DualHead2Go or TripleHead2Go to output 2 or 3 separate images that have the data-doubling and edge-feathering already done.
The module itself has an easy-to-use interface and takes away a lot of the complex math and calculations usually necessary to accomplish a good edge blend across multiple projectors.
What are the limitations?
Because the Edge Blending Module makes use of a Matrox DualHead2Go or TripleHead2Go, you are limited to up to 3 projectors at this time. Projector placement is limited to horizontal at the moment (you cannot create a top, middle, bottom edge blend with the Edge Blending Module).
Also, like any edge blending solution, you cannot create a projected resolution equal to the total output width of the DualHead or TripleHead2Go. Because of the data doubling, usually 100-250 pixels, your finished image horizontal resolution will be 10%-20% less than the output of the computer itself.
Only two things in ProPresenter cannot be projected using the Edge Blending Module - DVD content played directly from the DVD itself or content using the Web view. Neither go through the GPU (where the software does its rendering work) and therefore cannot be manipulated and mapped to this custom resolution. The DVD issue is due to a limitation placed on us by Apple with regard to Digital Rights Management (DRM) that is not legal to circumvent. The Web View is displayed using the industry standard WebKit and therefore drawn in a custom window that we're not able to manipulate.