In my opinion, though, they were more imaginative with designs back then (as an example I am thinking of the Mustangs that came out in the 60's compared to later versions). They are considered classics for a good reason.
Campbell Ritchie wrote:But can you remember the bad cars from the 1960s?
Definitely! Those cars just screamed bad @ss!!! Muscle cars built for speed that sounded like a tank going down the road!
Referring back to aforementioned '67 Impala, I remember us kids piling into the back of the car, no seatbelts, no car seats - heck, sometimes my little sister would be sitting on mom's lap in the front, while dad took us wherever (not that he was a bad driver). Today he would be pulled over and likely hauled off to jail (or at the very least, been ticketed to Timbuktu and back).
I can remember old Morris vans with catches on the windscreen so you could open the windscreen. And wipers operated by the vacuum from the carburettor, which wiped slower if you put your foot down. And I survived sitting between the driver and the driver's door in a van. That was before the 1960s, though.
Slip‑roaded, surely, not side‑tracked. Or access‑roaded
Randy Maddocks wrote:. . . I have side-tracked your original topic, . . .
That happens all the time; both face to face and on these fora discussions gradually veer off their original course. I did once drive a car with automatic lane following, which was reluctant to veer off its original topic.
salvin francis wrote:Just read this today, thought about bumping this thread :
Yeah. I think the real question is whether a human would have been able to stop in time. It appears to be someone jumping out into a dark road. I don't drive at this time so I don't have an opinion.
If a human driver would have had the same problem, we run into what is acceptable. If autonomous cars are twice as safe as people, is that ok? Do they have to be perfect? (I don't think they will ever be perfect)
Indeed, if the "driver" had been concentrating then she would have spotted her.
Wonder why the LIDAR didn't pick her up, as she was well across the road.
The video makes it look like she appears out of nowhere, but that's more to do with the camera trying to prevent the bright headlights washing out the image, causing the darkness at the edges to be really black.
But the video shows the car taking no action about Ms Herzberg.
Jeanne Boyarsky wrote:I think the real question is whether a human would have been able to stop in time.
I have a slightly different perspective here...
As a human, we are capable of doing things that a machine can't do. A driver would behave differently if he was alone vs if his family were in the car with him. When a collision is inevitable, we would at least do our best to avoid as much as possible including hitting the breaks, turning the steering to another direction, in some cases even accelerate to lead ourselves out of harm's way. If a pedestrian was far enough, the driver would at least honk.
A kind human would stop his car and even get out to inspect if he "thought" he hit something. We are alert to every kind of sounds our car makes and any sound that's peculiar like some kind of "tick" or a "bang" would get our mind's "curiosity engine" up and running. Our collision logic is based on sight, sounds, feeling (bump), and a sixth sense. There is also a "known danger" logic that we internally possess : ie. driving cautiously when there is a forest on the both sides of the road, driving slower than the road limit in times of rain/snow, etc...
We can predict the future by observing that a pothole is present far in the distance and that we need to slow down and either avoid it or if it is small enough, ensure that its encountered at all. Visually, a pothole would look like some "blob" in the road to an image processing algorithm.
Jeanne Boyarsky wrote:we run into what is acceptable. If autonomous cars are twice as safe as people, is that ok? Do they have to be perfect?
If all the cars of a given company follow the same algorithm, then there is a possibility that a flaw present in one of them is present in all. This would not be the case with humans who learn from their own mistakes.
What I am scared of is the fact that the pedestrian is merely a bunch of "pixels" or "dots" to an algorithm.
While it is lovely that many apps can detect a "face" and even tag us with a 99.999999% accuracy, I would be very scared if that same app was assigned to a robot to cut my hair
However, these simple engines needed constant tuning, and they leaked.
Modern cars are made of sophisticated parts that it would challenge an aerospace machine shop to reproduce and a limbo expert to replace. They're much more complicated and they carry a ton of systems that were neither available or affordable on '60s cars like backup cameras, and a lot of things like power windows and mirrors used to be exceptional where they're now the rule. To say nothing of the intricate computer buses running between things.
I'd think that automated cars would likely do better in NYC than in, say, Orlando. The traffic might be (slightly) denser, but you can see some really creative insanity on Florida roads. One major intersection in particular I know of had 2 left-turn lanes and 3 straight ones, and somedays I swore that the sole purpose of the left-turn lanes was to allow people to pull out from them in front of the people in the straight lanes so as to avoid the long wait in line for the straight line (Orlando traffic lights were also a good place to grow old - except for the one where a neighbor T-boned a Ferrari with her Ford Fiesta. That one was practically red before the green faded.)
Actually things have improved since I lived down there - they finally put in a decent beltway and the infamous turn intersection was recently replaced with a flyover. But people do crazy things down here.