Adrian Newby wrote:Once again, even with the best case robot-driver scenario, in the slowest hairpin, everything goes right - you are still accelerating directly at the wall you are trying to avoid.
At some instant in time, yes your instantaneous inline acceleration vector is pointing outside the bounds of the track. So what? At all points in a corner you are invariably pointing outside the track. But since you're cornering and rotating and not driving in a straight line, you can continue to rotate and follow your intended path even while accelerating.
Now then, my position hasn't changed. I'll grant you that what I think of as the apex of a path may differ from yours and that may have been a source of consfusion. As I said, to me it's the point of tightest radius rather than where you come to clip a kerb. IMO it's an easier working definition as from car data you don't get to see an overhead of where the car was on the track - you just have the sensors. Additionally, outside the scope of road racing you can take a corner on a big track (particularly with variable banking) or on an asphalt pad and never really come close to any inside boundary.
Irrelevant of where that max curvature point is, I maintain that trail braking is the fast way around your generic corner. That's not just my opinion or experience as an engineer. Car data from Michael Schumacher to Lewis Hamilton shows that that is the preferred method. Performance driving schools teach it. Literature covers it. Assuming that they've got all that pretty well figured out, it is just not possible to trail-brake a corner while on a constant radius. It would ask either too much of the tires on entry and you'd wreck, or you'd be diabolically slow mid corner. Ergo we can conclude (and again, car data will show this - either with "real world" or sim racing drivers) that the ideal line through a corner is variable radius. Furthermore, the general wisdom is that "slow in, fast out" tends to be the fastest way around the track. To do so, taking what some would call the "late apex" (or really late clipping) line geometrically requires that the tightest part of your turn be before the 50% mark. It's unavoidable. The whole premise is to maximize path radius on exit so you can get back to throttle quickly, at the expense of some corner entry performance. There are times however where that may not be possible - particularly while attacking, defending, or in traffic.
In my mind, that's all been pretty well proven out in the world of racing. It's not just this one guy's opinion. So to that point, I'm pretty content being left to my own devices and I'll admit that my views aren't going to change. I'm just trying to give some explanations and reasoning behind how that all pieces together. But indeed, I will grant I have not grasped your points. I'm certainly willing to hear them out if presented with some data / car telemetry / personal trackside experience which back them up.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.
