5 years...? I bought mine with FSD in 2018, and that was years after it was "right around the corner." Worst Kickstarter of all time... Though I do like the car itself.
Yeah, FSD was promised with the release of the first model S, 2012 - to be ready by "next year"
As such, it's been roughly 12 years just around the corner for Tesla and Musk enthusiasts.
To be more specific: Musk explicitly said in that marketing event that buying anything other the Tesla wouldn't make economic sense, as they'll earn their own price back as self driving Taxis within the following two years.
Still blows my mind that people believed him anything as these kinds of unrealistic promises were at the heart of every event since the start.
Believing someone with promises like that is - from my perspective - begging to get scammed.
From that point of view, people should be glad the delivered cars were decent. Most purchases with outlandish promises end with merchandise that is borderline unusable.
This is at the other extreme end though. They could do nothing and call the agreement to explore satisfied. Would rather they wait till they've removed at least three of the hedging words.
The ODD for drive pilot is so limited, I don’t think it’s really comparable. I have very little faith that their approach will scale to anything more than a traffic jam pilot gimmick.
It’s fair to argue that FSD is limited as well but I believe their approach is much more scalable.
It's not just speed, its things like requiring lead vehicles, only highways speeds in the far right lane (likely for localization), weather, etc. The approach drive pilot is taking, with their technical investment and sensor suite, will not scale to urban/suburban driving.
FSD, in the states, works on approximately every street, parking lot/garage, etc.
All I’m saying is that starting FSD from park in my driveway and having it drive to my destination with my hands on my legs and then having it park itself when it gets there seems reasonable to call “full self driving” to me. I pay for the subscription and I would continue paying if it never got any better. I do live in a rural state, so maybe that’s why it works so well.
Having used cars that had that "supervised" driving feature.... Gosh, I hope you were paying 100% attention the whole time of that driving experience you described. Even the smart cruise control features I've used allowed my mind to drift, and I was glad for the beeping from the steering wheel telling me to pay attention. I don't use those features anymore.
If it's full self driving, then I assume that Tesla is paying for your insurance and taking all responsibility for any crashes it causes in your car?
Let's see it do that in the snow, heavy rain, anything that doesn't replicate ideal conditions in SoCal. You're riding on the sweet spot of a Gaussian and at some point you're going to experience an outlier when the machine makes a wrong interpretation of its inputs.
Do you even drive? Or have you tried using any of the features discussed above. I think I’m going insane seeing people comparing cruise control (lol) to FSD. One is a line follower, the other is a teenage driver with a fresh license. They’re not the same.
Oh please, just because it can do (bad, sometimes horribly dangerous) turns doesn't mean that 99.9% of the time it's not just a glorified cruise control with lane centering.
I owned a Model 3 with FSD. I own a Mach-E with BlueCruise. They're equivalent for the majority of drives I've done.
They edit their videos to remove the mistakes. It's all a lie if it only works 90% of the time and you don't know when it's going to fail after being lulled into inattention.
Waymo made a very good point about this, fundamentally that's the problem with a progressive rollout - drivers will stop paying attention when the software is "good enough", and the result will probably be crashes. So for them, it was all or nothing - either it fully drives itself, or it's not worth deploying.
Mercedes system is dumber in every way, they merely set the criteria narrow enough to get dibs on L3 for gullible people to repeat on the internet. It’s OBVIOUSLY more dangerous than FSD under the same circumstances.
No, they really are no different. A legal guarantee doesn't actually mean the car is safe, it means they will pay for it when the safety features fail. Those fees paid out can just be considered a marketing expense to make the car appear safer.
This is misinterpreting what I'm saying. I'm not arguing for the safety of Tesla's system. I'm saying that judging the safety based off of corporate marketing decisions is a mistake and putting a guarantee on a product is a marketing decision.
Even if this originated as a marketing thought bubble, there's no way that such a decision could've been made without direct approval from the executive (including the CEO), and only after taking advice from their general counsel and consulting with the board. The potential reputational damage is too immense for such a decision to be made by "marketing" alone. What you're describing has happened before and the courts awarded massive punitive damages against the motor company.
Yes, they are different. The degree depends on the company and if they have a history of trying to weasel out, but a legal agreement makes it harder to dodge liability in court.
Tesla would love to offer the reassurance to buyers but there’s a reason they haven’t done so: they’d lose money on it.
You are failing to understand what I'm saying. They don't have to weasel out of legal liability in court. They just bake the legal settlements they know they will have to pay into their marketing budget.
Has no one watched Fight Club and heard the anecdote about how a company will only recall a car if the cost of the recall is lower than the cost of settling all the lawsuits? All this guarantee tells us is that Mercedes did a similar calculation. Taking legal liability is not proof the car is safe. It is proof that they think the value of customers thinking the car is safe is more valuable than the cost of paying out settlements. Tesla not making the guarantee does not prove their cars are unsafe. It is evidence that if they did the same calculation, that got a different result. Maybe that is because the car is more dangerous, but it could also just be a different marketing philosophy and Tesla notably does not approach marketing like most other car companies.
The conclusion that you reached in which the Mercedes is safer than the Tesla is valuable to Mercedes and that opinion was indirectly purchased by Mercedes paying out legal settlements.
Yes, we know that companies exist to make money. My point is simply that when they are willing to make a stronger legal commitment in a country famous for litigation it suggests that they have a higher confidence level in their system.
Think about it like this: company A says “our government product is military-grade. We have a 1 year warranty.” while company B also says their product is tough but offers a 5 year warranty. Which one do you think has better data supporting the durability of their product?
That scene is a reference to Grimshaw vs Ford Motor Co.
The precedents set in that case mean that the liabilities arising out of legal action based on 'strict liability' are likely to be extremely punitive (these days, well upwards of the $147M awarded against Ford in 1980, and into the billions). Any company that did not factor such a payment in their calculation in addition to the indirect costs of reputational damage, deserves everything they get. I doubt this is the case with Mercedes.
One question I do have that perhaps someone here will know - is the Mercedes guarantee limited to certain locales? e.g. Germany only as the roads there are in good condition and well marked? (I'm assuming here).
Yes, and them putting their money where their mouth is means they are reasonably certain the accidents wont eclipse the profits, which is a much better signal than whatever Tesla marketing is putting out, seeing as they're not able/willing to do the same.
I'm not familiar with the facts of the matter but if it is indeed the case that Mercedes is indemnifying drivers for accidents caused by FSD, then that's far more than marketing, and your comment (without presenting any facts to the contrary) is unwarranted.
> The bigger difference is that Mercedes’ system only works on highways, under 40 mph, and you need a car in front of you that it essentially follows.
And geofenced to specific highways, only during the day and during good weather.
It's still cool (to me at least). But it's bizarre seeing people dismiss FSD as being the same as adaptive cruise control while touting Mercede's Drive Pilot. Drive Pilot is a lot closer to adaptive cruise control than FSD.
It's unfortunate that there's so much misinformation that gets thrown around whenever this topic comes up.
People are interested because Drive Pilot is L3 while FSD is L2. People are naturally more interested in the more advanced systems, but that would include FSD if Tesla can improve it to perform closer to the way it’s been marketed. Exaggerating the capabilities for over a decade juiced their share price but it also gave them a reputation for failing to deliver which is going to need hard data to shake: putting their money on the line would be one way to do that.
Progress would be get certified for self driving. For comparison, Mercedes, BMW, Honda etc have L3 cars on the market. Mercedes just got approved full highway speeds in EU and working on L4 certification.
I just checked out Mercedes, and it appears to be geofenced with a lot of restrictions[1]:
> DRIVE PILOT can be activated in heavy traffic jams at a speed of 40 MPH or less on a pre-defined freeway network approved by Mercedes-Benz. DRIVE PILOT operates in daytime lighting conditions when inclement weather is not present and in areas where there is not a construction zone. Please refer to the Operator’s Manual for a full list of conditions required for DRIVE PILOT.
Only on select freeways and only under 40 mph (and only during daytime with good weather conditions) sounds like it wouldn't be particularly useful.
Still, the tech is cool, and moving in the right direction. It's just always hard to really tell the state of things without doing some digging, because there's a ton of misinformation that gets thrown about whenever this topic comes up.
That’s what the government allows them to do. They got approved for freeway speeds in EU and are trying to get approved in CA. You need to prove that the car is safer than a human driver.
There's now a thing called "FSD", yes. But it's not FSD as in Full Self-Driving, as in L4. It's still an L3, the driver still needs to be at the wheel and paying attention. "Full Self Driving" implies L4. What Waymo has, with no one at the wheel, is L4.
FSD is Level 2. The driver must, at all times, be supervising the car. L3 implies that the car, in exceptionally ideal and limited circumstances, takes over full control unsupervised.
Even if you were to take Mobileye's definitions, it would still be equivalent to other companies at best (hands off, eyes on).
Compared to "FSD this year", every year for the past five years, I honestly find the approach pretty refreshing.