Yeah, and that 15% is the problem. "15%" (no matter how unscientific a number, just an illustration here) is simply too often.
I've driven adaptive cruise control since 2006 (premium Germans), I've been through three, four hardware generations (some a couple of times) of them. They were not perfect, but the lack of perfect meant maybe one ghost braking during the ownership of a car (perhaps when the car in front turned 90 degrees away) as well as the obvious misses within limitations of the system before cameras came along (watch out for stopped cars etc.). The adaptive cruises quickly become second nature and one could anticipate very well what the car would do. I would use them in all kinds of traffic and roads with great success. As I said, second nature.
Enter my AP2 Model X. It ghost brakes pretty much every single time I use the TACC and already in the space of months it has done this a couple of times seriously (as in brakes down in the middle of the motorway). Most braking events are luckily milder than that, but even they are a constant reminder of how risky it seems going under an overpass. The only second nature that it has been incuding in me is tensing up every time an overpass comes to sight.
I don't know if it makes any sense, but to me the Tesla TACC seems somehow thin, strained, less robust than those e.g. Audi has. I don't get the same sense that the AP2 one knows what it is doing than I got from the Audi. Sure, AP2 is trying to do more and eventually is expected to do a lot more, than my old cars, but in the process - as it currently stands - it is missing that singular purpose, that precision tuning for a specific mission.
AP2's TACC currently seems like a scatterminded teenager on a Friday night compared to a singleminded professional on a mission from the competition. AP2 seems nervous, where as the German's I've had were confident.
Some say perfect is the enemy of good. I say bad is the enemy of good as well.
