Here are geometric illustrations of what Clay described above:
http://www.arielnet.com/Main/images/fig1f.gif
Technically, this is actually an illustration of barrel distortion in a lens, which is a true distortion like pincushion, but lack of pincushion produces a very similar perceptual effect in optics.
Holger Merlitz uses moving images in his technical report to illustrate the the "globe effect" (his name for "rolling ball") that illustrate closer to what you actually see (if you see it!).
http://www.holgermerlitz.de/globe/distortion.html
The upside to "rolling ball" is that the images in the center field look even larger than normal. Given that roof prisms give a larger image scale than comparable magnification porros to begin with due to their close set barrels, the image scale through the 10x42 LX/LX L was huge in the center.
Unfortunately, the image scale became smaller at the edges (or to use the technical jargon - "squished").
It's this imbalance between the inside and outside image scale that gives the impression of the image scrolling over a ball, and when you move the bin, you get a "rolling ball".
Even the night sky looked like curved like Ptolemy's Crystalline Spheres through the LX L.
Some manufacturers can give you too much of a good thing by adding too much pincushion.
Then the views look like this:
http://upload.wikimedia.org/wikipedia/en/9/97/Pincushion-distortion.jpg
When I looked at my computer screen with the ZR 7x36 ED2, it had a saddle shape appearance.
Fortunately, while panning at longer distances, the pincushion wasn't distracting like "rolling ball," but I have used bins with too much pincushion that produced an inverse "rolling ball" while panning, for example, the 10x42 Swift Ultralite and the Swift 8x44 ED Ultralite.
The good news is that some people who initially see "rolling ball" eventually adapt to it.
So if you are not sure how you react to "rolling ball," order your SV EL from a store with a liberal return period so you have time to see if you can adapt to it.