Maddy

Saw Meet The Robinsons 3D at a “state of the art” theater, i.e. an almost brand spankin’ new theater with the RealD deal.
Firstly, let me say that it is not possible for me to make a precise evaluation of any 3D movie without having it available on a computer (and yes, I can assimilate “most of” the theater viewing conditions with a computer, at least enough for a good evaluation).
Took my handy-dandy stereo FOV (field of view) print-out with me, that gets me within a degree or two of accuracy of gauging the FOV of any particular seat. Before the show, I was surprised to find out that the front row seats have nearly a 90 to 100 deg FOV… not good for most 3D, BTW, unless it is a “special show” shot with a “typical” stereo camera with a 16-17mm focal length lens… (never seen one of those “special shows”, yet…
BTW, the back row seats have close to a 30 deg FOV.
My “optimum FOV” for viewing “standard” stereo 3D images (based on 35mm “standards”, and not medium/large format, 70mm film, etc) is 33.3 deg, and the “total acceptable range” is about 30 to 40 deg (not an extremely huge difference from 30 to 40 deg).
In this theater, that range was from about in the middle of the theater to the last row… anything closer than about half way would give you more than a 40 deg FOV, all the way up to close to a 90 to 100 deg FOV (front row).
I sat in a seat with close to a 35 deg FOV for most of the movie, but for a short time (starting somewhere near the middle of the movie), I sat in a seat with close to a 50 to 55 deg FOV (a bit “too close”, and found the experience a bit unpleasant). It was not possible for me to use my FOV gauge at that time, so I am only guessing about the FOV. IMO, special theaters should be set up for 3D movies, with _only_ a 30 to 40 deg FOV possible (again, for “typical” camera settings). If you let people sit closer than what results in a 40 deg FOV, you stand the big risk of giving people a bad viewing experience!
Now… about the stereoscopic aspects of the movie…
Firstly, one word about the RealD setup… I do indeed detect a motion artifact, i.e. a slight “fluttering” effect from the shuttering, especially noticed in fast-moving scenes… probably nothing too extreme, and there is probably not a good alternative, short of using dual projectors with passive glasses. Most people probably don’t even see this motion “flutter”, but it could indeed be “bad enough” to cause a headache after long periods of viewing (e.g. a full length movie???).
In the seat with close to a 35 deg FOV, I have few complaints about the stereoscopic aspects of the movie. I really feel that I probably would have come up with something quite similar, given the same circumstance, i.e. a movie that was rendered _first_ as a 2D movie, and not strictly rendered with stereoscopic aspects as a primary goal, i.e. the stereoscopic rendering involved adding a second camera “after the fact”, basically making the “best 3D possible” from the original compositions/scenes. A movie made “strictly for stereoscopic purposes” could potentially give you a much better stereoscopic “platform”.
As Phil mentioned, most of the movie had something close to 5 inches of positive parallax… not enough to exceed the “1.5 deg of eye divergence maximum rule”, as long as you sit in the 30 to 40 deg FOV range! (as I did, most of the time). I think all the movie had at least _some_ positive parallax… and although that doesn’t really hurt anything, I don’t think it is necessary… especially if it exceeds 2.5 inches, which can be very taxing on peoples’ eyes sitting “too close” (i.e. closer than a 40 deg FOV).
The floating windows were effective, rarely exceeding 1 or 2% of the image width (just guessing) of negative parallax. (I commonly use a floating window with up to 3.3% of the screen width of negative parallax, with no “problems”.) I never saw an animated floating window… just a changing window (per scene). An animated window might be another option for “added interest” or function (eliminating edge violations “only where needed”, and not necessarily for an entire scene). Also, I never saw any “out-of-rectangle” or rotated floating windows, i.e. rotated in 3D space on any axis. (Another idea worth trying, IMO.) Also, there was no feathering of the floating window edges, that I could see… a good trick, BTW, if you want to bring them out farther and not introduce the potential for ghosting at those edges.
Most of the movie seemed to have a bit “less than optimum” amount of total deviation, i.e. I would guess closer to 2% than 3.3%… with the exception of some scenes (mostly “fast-action” type scenes), where I would guess there was more than 3 to 4%, but not extreme amounts (probably rarely more than 6.6%, with the exception of some “rain” effects or something similar to that). You can correct me if I’m wrong about that… it’s hard to accurately estimate in the theater.
That’s about all I have to say… really, if the stereoscopic aspects of every 3D movie were “this good”, I don’t think anyone would have too much to complain about.
Fun stuff!
BTW… what was the movie about?… I didn’t get a chance to actually “watch” it… ;-)