I posted this as a comment vimeo but it might be more useful to post here, so here it is again:
I did a very brief test on a preproduction 7d a few weeks ago and unfortunately it seemed like the rolling shutter skew at 24p was worse than at 30p, so i guess the framerate may be achieved by lower clock rate rather than extra vertical blanking. Of course this was just what it seemed to me waving around the camera at approximately the same speed across a scene with some vertical lines. I could be wrong though, id love to see a test like this with two 7d's side by side in 24p and 30p. Of all the shots, I dont think you really need anything besides a closeup shooting as straight on to the side of the train as possible, so the skew is maximized and can be measured/compared more easily, like the shot at 0:25 but make sure both cameras have the same field of view.