With a lot of the decisions Tesla's been making lately, I'm not sure I'm going to buy another either. I'm mainly speaking of:
- The butterfly steering wheel.
- The removal of radar for Autopilot (I'd like to have multiple systems for this, not just cameras/vision).
- The removal of steering column stalks (and using the sensors to figure out which direction you want to go).
- The door open mechanism changing from handles to whatever this button situation is.
I'm sure there will be other stuff, but yeah, may have to cancel my reservation for the Cybertruck.
The talk by Karpathy sheds some light on while they removed the radar. TLDW: Radar is inaccurate and buggy for a few seconds, then typically returns to normal. And the sensor fusion is difficult too. FWIW, I still think they're dangerous. I don't want my car to be "production." Engineers were SSHing into tons of cars to apply updates according to an insider thread a while back.
> The Tesla Model 3 has regained its status as a Consumer Reports Top Pick after independent tests proved the effectiveness of its new camera-based automatic emergency braking (AEB) and forward collision warning (FCW) systems.
> Earlier today, the Insurance Institute for Highway Safety released the first evaluations of new Tesla vehicles that use a camera-based system for AEB and FCW. Because of their performance, the Model 3 will once again get a Top Safety Pick+ designation, which is the IIHS’ highest safety award.
> “Given the IIHS’ recent evaluations of Tesla’s new camera-based system on its Model 3 and consistent with CR’s integration of IIHS ratings into our recommendations, CR is restoring the car’s Top Pick status,” says Jake Fisher, senior director of CR’s Auto Test Center.
Radar based cruise control, like what my Toyota has, is pretty inaccurate. The shape of a vehicle affects how close the system estimates it to be. Its size too.
Maybe that's just one badly implemented system. But my eyes don't judge a vehicle's distance differently because of its size or shape. Usually :-)
If you have two different systems for judging distance, you still need to decide which one is correct. Maybe if there were three, it would be more reliable.
This is a human factors oversight. The emergency handle and button to open the door should be combined.
When you put pressure on the button the door opens electronically. Putting a lot of pressure -- as when panicking -- should result in the mechanical lever unlatching the door.
I recently pulled over to help a Model S driver with a flat.
Every doorhandle on the car had a thin zip-tie looped under it because the "present" mechanisms no longer functioned. One had to tug the zip tie to manually force the handle out of its flush position before being able to grasp it.
Quality.
The poor guy was stuck in the summer desert heat, waiting for a tow with a flat in a ~5000lb luxury sedan too heavy to limp 10 miles to the tire shop without destroying its alloy wheel, that apparently can't manage to carry a spare.
I'm pro-EV, but it felt like Teslas are a disaster as I left him in my tiny 90s miata that manages to fit a spare in the trunk, has previously limped that exact route on a flat without destroying its own alloy wheel (~2400lbs), and has perfectly functioning doorhandles still after 320k miles.
Not defending Tesla here - but most manufacturers no longer include full size spares and many don't even include donuts anymore. The former is reportedly because of weight savings which contributes to fuel savings for their federal requirements. The latter is because donuts were never maintained by owners so they never worked when needed (ie, already deflated or aged beyond safe).
Holy shit - it has _no spare at all_ and no run-flat tyres? That is totally insane to me - living in the country, I hate the idea of the donut spare, but at least you can drive a reasonable distance on it.
When friends are in my car they almost always use the emergency mechanical handle instead of the electronic one and I always have to remember to show them how to open the door the non-emergency way.
When you press the button, the window rolls down slightly before the door latch is opened. This prevents you from putting stress on the window when you push the door to open it.
When you use the handle, the door latch opens and the window rolls down slightly and there’s a larger period of time where you can push the door and put stress on the window. So the danger with using the handle is that you _might_ damage the windows if you aggressively open the door, but the button will not.
The 2020 Dodge Challenger had the same problem due to a similar door design. They solved it by causing the mechanical action of the door open lever to also roll down the window slightly, and electronically rolled up the window when the door was closed.
The easy solution that literally every other car uses is to have a piece of frame on top of the window as a part of the door. But that doesn't have the same aesthetic!
Originally they didn't want you to use the manual handle regularly but they added a software feature that rolled the window glass down a bit when you pull the handle and stopped warning not to use it unless it was an emergency.
Whatever you do - if you don't already have one, buy a combined window breaker/seatbelt cutter for all of your cars and Velcro it to the side of your chair/bottom of side of center console/wherever you think would be best to have it.
Good advice. I have a model S and a model X. The seatbelts in the X are especially bad - I’ve had a service call where the belt was stuck closed and there was nothing I could do to open it and twice I’ve had small children become trapped for several minutes in their car seat because there was no way to release tension on the belt and get them out.
I was within seconds of cutting them out with my pocket knife the second time it happened.
Most people don't know that the headrest (head restraint) can also be used to break a window in several ways. I'd known about this for a while, typically there's just a simple mechanical button you have to depress to remove the headrest to be able to use it in this fashion.
Then, for ergonomic reasons, I went to change the headrest orientation one day (I know this is stupid) - and, it turned out, in addition to a mechanical push button to remove, I also had to shove a tiny ass pin punch deep into a hole in a really weird place to fully remove the headrest. This was in a Lincoln, and I'd never seen something as stupid as it in any other car. So much for the idea in the back of my head Id've been able to rely on it if I'd ever needed to break a window.
Even more of a reason to just buy the purpose made breaker/cutter, never know what unconventional BS some brands/models may be doing.
edit: sorry, speaking in general - Tesla's dont/sports cars tend to not have detachable headrests. Or maybe most newer cars don't in general, haven't looked into it much.
Every time something like this happens, people point out that gas powered cars catch fire much more often than electric ones.
Quote: "That’s five fires for every billion miles traveled, compared to 55 fires per billion miles traveled in gasoline cars."
So why don't I want to own an electric car? Because my assumption is that the typical electric car fire happens while it is being charged, whereas the typical gas-powered car fire happens while it is being driven.
There's a huge difference between a fire that potentially destroys my home, quite possibly starts while I'm asleep, and a fire that just totals my car. And that holds even more for people who live in multi-family buildings.
This particular incident didn't happen at night while the car was being charged. But is that representative?
I knew a guy growing up who's gas car caught fire in his garage while it was parked and burned it down. I think your assumptions about when and how different kinds of cars catch fire are wrong.
Ok, but why do you think they are wrong? Your first sentence isn't enough to tell me.
An anecdote is a data point, but it's only one, and I've got my own.
I have seen a car on fire by the side of the road, and I've also owned a car with a history of recalls for issues with the fuel injection that caused fires for some owners. I've also heard of (read about) people who had fires that started under the hood while driving.
Another point is that while sometimes old cars have messed up wiring either because of design issues or amateurs messing with it, on the whole, I'm more confident it's still working as designed than my home, especially when I'm driving something in the 0-10 year range.
My home will be used indefinitely, and I don't know for sure what people may have done to the wiring over several decades before I lived in it. But I've seen some stuff.
I think most electric car fires are the result of serious damage to a vehicle from a serious crash. I don't believe there are any documented cases of a Tesla catching fire in a private garage and burning the owner's house down. There are some instances of them catching fire in public garages but it isn't clear those incidents aren't arson.
Gas cars have electrical systems that cause them to catch fire. There are loads of news stories about gas cars catching fire in private garages. It isn't clear they are less risky in this regard than an EV.
The stat of fires per miles driven isn't saying that fires only happen while driving it is just a way of comparing the relative prevalence adjusted for the fleet size and usage.
> a Tesla catching fire in a private garage and burning the owner's house down
The thing that sticks in my mind is that Ford had some issues with their charging cable and had a big recall to substitute one with a (presumable 50 cent) thermistor.
I guess there is a very general sense in which I associate electric cars with electronics where the tiniest software or hardware error can be catastrophic, whereas I expect that the scope of problems is easier to limit in a conventional vehicle. Like, gas powered cars rely heavily on electronics, but if a specific sensor or something goes out, they tend to have a "limp home" mode, and mechanical failures are planned for in the worst case.
I observe that shipping a lithium ion battery is a big deal, considered hazardous. They have brought down aircraft in the past. So I would tend to discount people saying that they are only an issue in a collision, just because without particular numbers, I see people worrying about spontaneous combustion. Due to manufacturing defects, I guess.
I think you're much more likely to have a counterfeit/faulty USB C cable that somehow makes its way to you lead to a fire with any of several objects containing Li-Ion that people tend to have in their houses anymore.
EV charging systems are pretty robust and fail-safed.
> For whatever reason, speculation is pretty rampant whenever it comes to Tesla accidents.
That, and we seem to hear about every single Tesla that gets into an accident in detail, whereas with other cars there's just a brief blurb during the news about how another 10 car pile-up blocked the freeway for an hour.
So I wonder at this point which PR firm is making sure that every single Tesla accident makes the news?
It's not a conspiracy. Tesla is among the most watched and followed companies in the world. A large number of people have huge positions on its stock, both long and short. Its CEO is one of the richest and possibly the most divisive businessman/celebrity in the world. All news related to it – both positive and negative – is greatly amplified.
Shorts must have the stock go down at a certain time. Longs don't. So shorts are far more incentivized to promote fake news every now and then (or they get margin-called).
I also don't know very many companies with a group of short-sellers so determined and coordinated that they've named themselves ("TSLAQ"), so makes sense they are also much more keen on ensuring every shred of negative press gets signal-boosted as much as possible.
>tried to unlock and open the doors but had to force his way out of the vehicle, as the locks seemed to malfunction. After he left the car, it began to move on its own
If this is true it's not great news.
I wonder, does the parking pawl engage at power-off?
Many years ago I had a vehicle catch fire while I was driving it. The first indication was that the digital speedometer started showing random numbers for this very reason.
Tested. Seatbelt off, door open, Model S shifts to park within 1-2 seconds. Even when shifting to another gear from park, it shifts back to park rapidly if door is open and driver seatbelt is unbuckled. Seatbelt on, door open, vehicle does not shift to park.
3. Why aren't regulators reviewing and approving firmware updates? For example, if Tesla management decides to brick all cars or an engineer is bribed to hack a particular Tesla to crash.
4. Why are moving vehicles being "live patched" over SSH? This seems insane like "I don't always test my code, but when I do so, I do it in production."
Maybe he had trouble getting out because he was occupied calling his attorney.
I’m pretty sure this is the first article on the first report of a Tesla issue where an attorney is giving statements that I recall seeing.
Not to diminish the seriousness of the issue but certainly seems likely they’re going to go for all they can against Tesla. I just wonder what the other side of the story will be.
Note, not a Tesla owner, mixed on fandom, but I do have Tesla Solar on my roof.
Sure, there might be a mechanical backup to the electronic door handle. But will I remember to use it in an emergency?