PRG Mbox Remote Manual 3.9 - Manual (Page 66)

60 MBOX ® REMOTE USER MANUAL
Contexts
A "context" is a rectangular area taken from the source area and onto which fixtures are placed. Each context can be
some or all of the source area, and has its size specified by width and height settings and its position upon the source
specified by X and Y offsets from the top-left corner of the source. The context is laid directly over the source image,
and is then divided into rectangular subdivisions. The context is subdivided by assigning width and height values for
sub-sampling. These subdivisions represent sub-sampled areas of the screen image, with each one being a possible
location for one pixel on a patched fixture.
In the concept illustration below, a context with 50x50 subdivisions has placed on an area of the source with a size of
800
px x 600 px . Each rectangular subdivision of the context will be (800 px /50) x (600 px /50), or 16 px x 12 px . Mbox will
average (sample) the color/intensity values for each 16 px x 12 px area of the screen image. Each sample can be used
to create one "pixel" in the map. This pixel is then used to generate pixel-mapping output levels.
The size and sub-sampling of the pixel mapping context plays a large role in how the pixels in the source are
converted into output data. Assuming you create a context that is the same size as the source, you could either
configure it to have one sub-rectangle (sub-sampling width and height of 1 x 1) or a million sub-rectangles (width and
height of 1000 x 1000). In the first case, the entire source would be reduced to one sub-sampling rectangle in the
context and therefore only one pixel in the map; the results would be fairly pointless. In the second case, you could
have a very high resolution map, but it would require a great number of fixtures to make full use of the context. In
general, try to avoid extremely large context sizes, as doing so requires extra processing. It is possible to keep
context sizes smaller and add more contexts to cover the necessary onscreen pixels for your output.
If you need a very large amount of pixel-for-pixel mapping of your video to LED fixtures, you may be better served by
using an existing video pixel mapping product rather than the Mbox pixel mapping feature. You should strive to create
a pixel mapping context that is as small as possible while still allowing you to map your content to fixtures in the
manner that you desire. Initially some trial and error may be required. If you are unsure where to begin, start with the
context at its default size of 16 x 9 and then experiment once you have patched your fixtures.
A pixel mapping configuration in Mbox may have more than one context, thereby providing for organization, and
allowing multiple sampling factors and multiple output protocols. Trying to determine how large to make a context
may be difficult before placing fixtures, however it is possible to modify the size of a context at any time.
800px x 600px Context 50 x 50 Context Sub-Sampling One "Pixel"
12px
16px