When I mentioned a specially shaped ramp in the previous post, I was referring to the b-curve, not just any curve. Read as, "I fail to see why one would go to the effort of building a ramp with a b-curve when a more ideal choice would be a smoothed out L shape, or one more like the picture in the wiki(which is not technically a b-curve)".
I wrote all this up, then reread your post and think I've found the miscommunication, but I'll leave this here for future reference
Regardless of ramp shape, the idea of conservation of energy will be held, so assuming all vehicles start from the same height, they will reach the transition point of the ramp at the same speed(also assuming the same friction, rotational inertia, etc). The point behind the discussion is that the vehicle can hit a 'bump' while transitioning from the ramp to the floor, this is where the ramp shape effects velocity. As Mrsteven said, as long as the ramp shape is as close to horizontal as it can be(which minimizes this 'bump') the velocity of the car leaving the ramp will be as high as possible, regardless of the shape above the transition. Now, that being said, the fact that you want as horizontal a transition as possible does mean that the ramp shape is a smooth curve. However, if you were to use a b-curve, the curve would need to extend from the start of the ramp all the way to the target point in order to be most effective, which obviously is impractical. The curve also does not provide the highest possible velocity at the end of the curve, due to the slight slope up at the end, so it isn't an ideal ramp choice in that regard either.