I recently, as one of the side benefits of a current project at Gensler, got to visit the visualization labs at Calit2 at UCSD, in particular those dealing with large scale screen arrays, both in 2d and 3d (with the aid of very stylish polarizing glasses). The technology we demoed ran a rather tight gamut from hemispherical projection rooms with infra-red head tracking, to gigantic arrays of networked LCD panels. That is to say, groups of display devices with sophisticated stitching software that knows how they are laid out and where the viewer might be located. I'm not trying to downplay how cool this all was (and it was cool), but the topic of conversation that kept coming up with each subsequent display was how the actual hardware was anything but cutting edge. Consumer electronics are now so powerful that these gigantic, room size displays were made of components that could be bought at your neighborhood big box store (and, in fact, some of them had been. In true engineer fashion, they hadn't removed the stickers yet.) The stitching software itself was running on three-year old gaming PCs. At one point, it was pointed out that currently it is cheaper to use arrays of LCD screens as an interior finish than it is to put up back-painted glass.
Some other things were also apparent. One is that the screen has finally caught up and surpassed the projector as a medium for environmental media. I don't know why I didn't notice this earlier. Another is that the power of this kind of environment is very dependent on being interoperable and omnivorous - it has to communicate with everything and take whatever format you can throw at it (it did the latter very well, the former being the missing piece of the puzzle).
This kind of technology is mature, inexpensive, and eye-catching, which is why I fully expect for it to be near-ubiquitous well before we get our hands on a Google self-driving car (although likely after we see someone wearing Google glasses run into a light pole).
Finally, a note on museums: it seems to me that, after decades of disappointing edutainment at science and technology museums, virtual displays might actually be close to beating out reality in certain situations. Not sure how I feel about that one. I'd probably prefer for my kids to look at real fish.
While I wasn't looking, sawapan have released a suite of Grasshopper components they are calling "Millipede." This is a diverse suite including finite element analysis, topology optimization, fourier transforms, eigenvalue partitioning, and much much more! I got to play with the alpha or beta versions of many of these components in Panagiotis' classes at the GSD and they are powerful stuff.