-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
- Status
While I cannot say a lot about the US Situation, I am afraid that TSLA ignores the European Situation and the established thinking about safety. We here in Europe have very clear standards regarding safety (EN 13849, EN 62061, btw according to what I know there is no such equivalent in the US). If you build a machine which consists of a control system and a mechanical system (TSLA is clearly doing both) and this machine can produce a risk for any human, you have to consider all these regulations. For example you will have to prove that the MTTF (mean time to failure) is lower than let's say once per 100 years. The timeframe now depends of course on the size of the risk. The problem is, that you have to consider and combine the MTTF values of each sub-system (e.g. the mechanical brakes, the analogue output signal, the control system, incl. processor and also the programing, all the cameras). Typically you can never reach the required combined MTTF values for such a complex system, would be even very difficult with full redundancy. The redundancy I do not even see currently, so I consider it as impossible to reach high combined MTTF values at all.
The only positive thing is that the German car industry tries to avoid that these regulations will be used for autonomous driving. However, you can be sure that there are standards to come within the next years. These standard will consider for sure the complete hard- and software (anything else doesn't make sense). Thus I am very sure that the now foreseen TSLA system will never allow autonomous driving in the EU (which I find very disappointing and also misleading by EM).
The only positive thing is that the German car industry tries to avoid that these regulations will be used for autonomous driving. However, you can be sure that there are standards to come within the next years. These standard will consider for sure the complete hard- and software (anything else doesn't make sense). Thus I am very sure that the now foreseen TSLA system will never allow autonomous driving in the EU (which I find very disappointing and also misleading by EM).
jesselivenomore
Member
The headliner is a lot of sizzle that reinforces Tesla's lead in autonomous driving, which is nice. The bears will point to this being another example of carrot on a stick since the roll out date is uncertain(potential delays, regulatory hurdles etc).
That's fair, but ignores the steak. $5k+$3K today for self driving package = higher margins and ASPs for Q4. I was wondering how they planned to achieve "approaching 30%" GM by year end when they were at 21% in Q2 and expecting only slight improvement in Q3. This should help a lot. Also increased demand after pulling some forward with the discounts in the Q3 push. You guys have mentioned the fence sitters waiting for AP 2(waves). I don't think this is actually that big of a pool of people since I doubt most people even knew about AP2 coming outside of online fan communities and diehards. Instead of fence sitters, I believe it will be new demand after all the headlines today, and the fact that a Tesla Model S or X is the only car you can buy today that can future proof you for self driving. While bears can point to that promise as a carrot on a stick all they want, each new purchase generated today due to this "carrot" is unequivocally steak for Q4 and beyond.
The only "negative" to the news is a rollback on AP features until validation is complete, which they said should be in December. Is that what everyone is fretting about, a 2 month wait? Isn't that exactly what they did when they first shipped AP1?
That's fair, but ignores the steak. $5k+$3K today for self driving package = higher margins and ASPs for Q4. I was wondering how they planned to achieve "approaching 30%" GM by year end when they were at 21% in Q2 and expecting only slight improvement in Q3. This should help a lot. Also increased demand after pulling some forward with the discounts in the Q3 push. You guys have mentioned the fence sitters waiting for AP 2(waves). I don't think this is actually that big of a pool of people since I doubt most people even knew about AP2 coming outside of online fan communities and diehards. Instead of fence sitters, I believe it will be new demand after all the headlines today, and the fact that a Tesla Model S or X is the only car you can buy today that can future proof you for self driving. While bears can point to that promise as a carrot on a stick all they want, each new purchase generated today due to this "carrot" is unequivocally steak for Q4 and beyond.
The only "negative" to the news is a rollback on AP features until validation is complete, which they said should be in December. Is that what everyone is fretting about, a 2 month wait? Isn't that exactly what they did when they first shipped AP1?
Last edited:
racer26
Active Member
Such regulations will be swiftly changed when confronted with the real hard data that demonstrably proves autonomous vehicles are safer than humans driving.
Tesla is building the tool they need to generate that data.
Mario Kadastik
Active Member
I don't quite get the 2 month wait even. The people who order today no matter where will not get the car before December where the software will be enabled already. Those who get the car before were buying it without the AP2 hardware which they I guess get for free now (or based on their prior price) so waiting 2 months for the usage should be trivial considering the benefit they got on the pricing.
sounds good... but the data doesn't currently exist... and until then the suggestion that "autonomous vehicles are safer than humans driving" is only theoretical... and it isn't even proven that the "tool" that Tesla is building is even the right tool to collect such data.
i find the lack of scientific reasoning on this subject crazy... EVERYTHING Elon has ever said about fully autonomous vehicles is only theoretical... and that includes yesterday's "product release".
Mario Kadastik
Active Member
Tesla has already demonstrated that AP causes less deaths. The factor was 2:1 in favor of AP driven cars. With full autonomy I'd not be surprised to see this number get even better quickly. All they need is cars on the road and that's exactly what they're getting starting from today.
they have proven this to YOU... dude... they have 150k cars on the planet... how exactly did they "prove" that Tesla Autopilot is safer than the other billion cars?
Johan
Ex got M3 in the divorce, waiting for EU Model Y!
Do you have a better idea? Really, if so what is it?
Where is the lack of scientific reasoning? Elon is putting forward a theory: autonomous automated driving can be made far safter than human driving. The validity of this theory will be empirically easy to test in the real world. We will have an answer. We must of course not jump to conclusions here, and say that there is overwhelming data to show empirically that the hypothesis is valid, but there is good preliminary data to suggest that it's a good and plausible theory. The next step toward making sure that theory does hold up to scrutiny is to get the massive amounts of real world data required, and it will of course have to be subject to regulatory approval which I assume will be gradually widened as the data pools up. Is this diffucult for you to wrap your mind around? Is your claim here on of incredulity - that you believe Elon's theory will likely turn out false? If so, considering your call to scientific reasoning, what data and what analysis is your case resting on?
I consider the price drop a gift. Added some premarket. Plan to add more during the day and next few days.
"Do you have a better idea? Really, if so what is it?"
YES... make an affordable M3 and deliver it on time... what are they doing spending cycles on this?... Elon should be working on "the best factory ever"... and here we are talking about fully autonomous fleets of cars and about to talk about SCTY... all this does is further guarantee that the M3 will be late and next year at this time we will be talking about all this as reasons for the miss.
On the other hand, I know plenty of people - mainly younger than myself - who aspire to owning a Tesla, but certainly can't afford one, even a M3 is a tall order. However, in theory they could buy one, pimp it out on the Tesla Network, and maybe end up getting the car for free, or making a clear profit. Even those not previously looking at a M3 could now become interested as a pure business case.
But I do agree there will be some, maybe most, that would never do it, espcially the older generations.
"Where is the lack of scientific reasoning?"
it's not from Elon... it's from you guys making suggestions that Fully Autonomous is here!... and it's already proven to save lives!
Mario Kadastik
Active Member
Model 3 will be able to generate income to cover the car payment. Maybe not right away, but it's going to happen. Once people realize they essentially get the car for free, the waiting list will grow really fast. The new vehicle market is > 80 million a year.
ev-enthusiast
Active Member
I like very much that AJ was right about Tesla and Uber. He knew Tesla's plans about the ride sharing platform in advance and already a lot of month ago. Thus his precise and repeated questions about ride sharing during the recent earnings calls. Thus AJ's high price target for TSLA.
Last edited:
mutle
Member
There are probably many other important stats to be learned from the data Tesla collects, aside from deaths.
For sure TSLA has not demonstrated this. I fully agree with myusername that there is a lack of scientific results. The current assistance system doesn't save lifes in average and is clearly below the average driver regarding the capabilities. The comparison EM always likes to make is incorrect: just because TSLA has collected XXX miles on autopilot and only suffered YYY known deaths per mile in average doesn't mean that it does drive safer than a human. How many times did a human interact and avoided an accident which the AP would have caused? Nobody knows, but if I see such videos, I am sure there are many of them!
Tesla Model S: Autopilot birgt hohe Unfallgefahr (at 38 seconds)
Even if there would be no interaction necessary, this comparison of EM is incorrect: within the death statistic, the guy who e.g. drives drunk and dies in an accident is included at the human death rate. As I drive always fully awake and clear in mind, you will have a complete different and lower figure. You can be sure that EM will wish some day he would not have underestimated this topic this much.
I still remember the people telling that speech recognition could reach near 100 % accuracy some day (much easier than autonomous driving). What have they reached? For a car that drives completely autonomous you would need 99,99999 % even under worse conditions (bad light / rain, snow, ...).
It's a mistake to focus on the end goal of full autonomy.
Tesla AP 2.0 hardware is here now, *in production*, no more speculation when 1.0 will become obsolete, in few months it'll exceeded 1.0 capabilities and continue getting better while more precise GPS will make fleet learning even more effective.
Parallel parking will get much better, changing lanes will be much safer, so will AEB... Better awareness/less false positives due to better sensor coverage and new features such as red light detection, pedestrian and obstacle detection will keep Tesla as one of the safest cars on the road.
Last edited:
- Status
Similar threads
- Locked
