Test Drove a Tesla Today

NoCoPilot wrote:

So, does the loss of three people to Tesla auto-drive justify scrapping the whole program??

If you are one of those three people, the program loses a lot of its appeal.

My only point is that the system is not ready.

What criterium will you use to determine when it is ready?

I assume you mean criteria or criterion (criterium is a bicycle race).

I will consider it ready when it works properly.

Define “works properly.”  Include relevant statistics.

And I assume bicycle races are not part of it?

NoCoPilot wrote:

Define “works properly.”

That seems obvious.

No it doesn't. Three people have been killed under very unusual circumstances: one was watching a movie, one stepped in front of a tesla on a darkened road, and one hit a defective jersey barrier.

Meanwhile, hundreds of millions of miles have been auto-driven accident-free.

I ask again -- is the only acceptable rate of accidents 0%?

NoCoPilot wrote:

one was watching a movie,

That was not possible to ascertain. But even if he was, the car should not have driven into the side of a huge truck.

NoCoPilot wrote:

one stepped in front of a tesla on a darkened road,

Actually, I think it was a Volvo.

NoCoPilot wrote:

and one hit a defective jersey barrier.

Note that the car HIT the barrier. Would it have made a difference if the barrier had not been previously damaged? It's still a big piece of fucking concrete - something to be avoided regardless of condition.

NoCoPilot wrote:

I ask again -- is the only acceptable rate of accidents 0%?

You are the only one who has suggested that. And it's a bit silly.

NoCo wrote:

There have now been THREE fatalities from Tesla's self-driving feature*.  The first one occurred after 130 million miles driven.  Not sure (can't find) how many road miles accumulated before #2 and #3.

The fully self driving (FSD) mode on Teslas has not been activated except in testing (the hardware is there and you pay for it; you just can't use it). Only the first stage of FSD, Tesla's Autopilot, has been active. As of May, 2016, Teslas had logged about 100 million miles with Autopilot activated. That number may be up to 300 million by now. The only FSD cars on the road are Alphabet's Waymo and some of Uber's cars.

In the US, there are 1.18 fatalities for every 100 million miles driven. Uber's cars have logged about two million miles in FSD, with at least one fatality. Tesla's have at least two fatalities in autopilot (not even FSD!). Those numbers do not impress me.

So zero it is then. Nothing else would be sufficient.

There you go again. I've not made that statement.
But just for grins, you pick a number that you like better than zero.

And while you're pondering that number, consider this: what is the benefit to society for self-driving cars?

The number I pick is 926.  

Remember, statistically 94% of all accidents are human error. Therefore, the sooner and faster we can get all human-driven cars off the roads the better.

Realistically, it won’t happen overnight. Realistically there are steps we could take which would have the greatest effect for the least investment. First I would embed wire in the freeways and design cars to travel at 80 mph with a foot in between them.

Rural roads and roads with pedestrians and cross-traffic  — as we have seen — present an exponentially more difficult challenge. Overcoming these challenges will be exponentially more expensive. The benefit is still there — more accidents occur under these more difficult driving conditions — but new technology will be required, beyond mere following a wire or having proximity sensors.

But listen. We can’t even get the EPA to keep emissions standards. What hope do we have for automotive safety standards?

NoCoPilot wrote:

First I would embed wire in the freeways and design cars to travel at 80 mph with a foot in between them.

I want to stand at the on- and off-ramps and watch the fun.

NoCoPilot wrote:

... 80 mph with a foot in between them.

In California, that's just called traffic.

In New Jersey, they only do that during the rush hour, but then it's probably 70 (in a 55mph zone).

NoCoPilot wrote:

First I would embed wire in the freeways and design cars to travel at 80 mph with a foot in between them.

You realize that something like this would require that only self-driving cars would be allowed on the road and that the occupants of the cars would have no control: with only eight thousandths of a second separating the cars, that would be a necessity. And, of course, if one car had a mechanical problem, everything comes to a standstill for hours while the debris from the massive collision is cleared up.

But you are on the right track: the roadways must be designed specifically for autonomous vehicles. But the cost would be in the trillions of dollars for redesigning and rebuilding the roads. And that would not include local roadways. I don't think there is any way to make rural and small urban streets suitable for self-drivers. And most fatalities occur on rural roads.

You can't completely run down a lithium ion battery. They will not take a charge, except with some special chargers which, unfortunately, can cause explosions if things don't go just right.

Tesla has an option of a standard automotive battery that prevents complete discharge (don't know why it is an option), but that system won't keep the battery from going dead while you're driving, only while the car is out of service for a long time.

This guy's a believer. I think I am too.

Elon Musk: Tony Stark Gonna Tony Stark

Did I mention I was hoping to rent a Tesla for a road trip on our anniversary next month? We were initially thinking of driving down to Crater Lake, but all of the hotels down there are already booked. So I got a simpler/cheaper room at the Snoqualmie Falls Lodge instead.

Thought it might be fun to rent a Tesla for the trip.

But, $300/day plus 15¢/mile (anything over 150 miles). Wowsa. Won't be doing that!

Funny. And great proof that paying a real mechanic to fix your car is often a brilliant move.

wtf?

Tesla in turmoil after Elon Musk smokes marijuana on live show and top executives resign

Elon smoking weed is the least of their problems. The stock has fallen twenty percent in the last month, and is thirty percent below its all-time daily high. They have sold (taken deposits) on thousands of their new lower-priced car and have only built a few hundred. We may be seeing the demise of Tesla coming up. Anyone remember the DeLorean? And that was built by a guy who actually knew something about cars and their production.