Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Elon: "Feature complete for full self driving this year"

This site may earn commission on affiliate links.
Just more 3 card Monte The same feature set (full FSD capability) most of us paid $8,000 for would still be $8,000 (Full FSD capability) as you described Tesla monkeys around with the base car price so much that my 2017 $122,000 car is less than $100k today so a $2,000 car increase or decrease is just a temporary thing one way or the other

05CB0B07-59AC-4E16-AAA3-7D8DAC491409.gif
 
  • Helpful
Reactions: TaoJones
Just to add to your excellent list of regressions in the 2019.8 series, there is now a spot on my commute where it believes the speed limit has dropped to 25mph (on the highway) and it slams on the brakes. This is an area where I'm on an overpass and there's a fairly complicated interchange happening in 3-dimensional space -- which is really common on urban highways, not like some weird edge case. I can't tell if it has decided that I actually got off the exit or if it thinks I'm on the road under the overpass, but either way it didn't used to happen, and it's dangerous, and I hope they fix it soon.


Thanks for mentioning this. I've seen this, too at half a dozen different spots on my normal route, where suddenly the car tries to go to 45 (or even 25) in a 65 zone. Every one of those behaviors, if a human didn't intervene, would likely be a citable violation of the California vehicular code.

I think they used to do some sort of blacklisting of the map data near intersections, and they don't now, in preparation for allowing NoAP on city streets. So now we're seeing all the bugs revealed by the nav system not understanding the difference between a slower speed on the road above you and a slower speed on the road you're on. At least that's my best guess about what's happening.

Either way, 2018.8.x is the worst software regression I've ever seen in all my years of tech, and I'm including the Microsoft Office ribbon atrocity in that. :/


On a practical note, does anybody know of any way to actually get the attention of somebody inside Tesla to look at an issue like this? Bug report is a black hole. (Also, in 2019.8, it usually takes several minutes to get bug report to work...)


Email customer service, but I'd imagine they've already heard the screams at this point. But the fact that they shipped this fetid piece of dingo turd to more than half of all Telsa drivers makes me wonder whether Elon is the only one at Tesla who is smoking something. I'm just saying.


The best thing Tesla could do for their customers, IMO, would be to roll back every car that got a 2019.8.x release to whatever they were running before, assuming that's possible. The problems with this release are simply too severe, and given that they've been trying to fix this nightmarish weekly for... well, I guess six weeks now... I have no confidence remaining that they'll be able to fix them before the body count starts piling up. This release just needs to go away. Chalk it up to a learning experience, roll it back, and try again when you've gotten more of the kinks worked out.
 
Elon's mindset certainly is not.

Before you naysayers jump in rejoice that is actually a good thing.

Jeff Bezos: People who are 'right a lot' make decisions differently than everyone else—here's how

"Consistency of thought is not a positive trait." Even dumbest people can be consistently wrong. Recognize a few of them on this forum? ;)

Elon is very consistent. I don’t recall him often changing his mind about anything that matters. This is one reason why we are in the current mess in the Tesla world: holding onto no advertising and no deals so resorting to evermore weird demand levers, doubling down on the 2016 FSD fiasco still so you get what we have here now etc...

How fresh it would be if he’d just at times say ”fullstop, was wrong, we’ll change the company”. The closest thing to that was his failure with the robot factory but that is rare. He really had no choice there so there wasn’t really a choice to stick with the robots...
 
Elon is very consistent. I don’t recall him often changing his mind about anything that matters. This is one reason why we are in the current mess in the Tesla world: holding onto no advertising and no deals so resorting to evermore weird demand levers, doubling down on the 2016 FSD fiasco still so you get what we have here now etc...

How fresh it would be if he’d just at times say ”fullstop, was wrong, we’ll change the company”. The closest thing to that was his failure with the robot factory but that is rare. He really had no choice there so there wasn’t really a choice to stick with the robots...

I think you mistook principle to practice.
 
Read the article if you don't understand. People are complaining Tesla changine option and pricing structures, stores situations etc.. Basic priciples should never change.

I understand, I just disagree what here constitues as a basic principle. Not advertising and not doing discounts is not in my view a basic principle in this mission, let alone preselling FSD the way it was sold (especially if you are unable to really stick to delivering those either). A more flexible leadership would in my view have seen sticking to these moves detrimental to the real core values and changed tactics.

It is hard to see Tesla’s mission or any core principle related to it would have been hurt if FSD would have shipped exactly as it had but without the pre-sold promises that misled customers. Or if many of the questionable demand levers — that clearly have also eroded customer trust — would simply have been replaced with advertising and traditional price action.

Now instead we have this mess where a lot of us just can’t really take Tesla seriously anymore while still paradoxically appreciating the basic product. That seems like a completely unnecessary predicament for a company which has such a solid underlying BEV mission and product — and a predicament more flexible leadership would in my view have avoided.
 
I understand, I just disagree what here constitues as a basic principle. Not advertising and not doing discounts is not in my view a basic principle in this mission, let alone preselling FSD the way it was sold (especially if you are unable to really stick to delivering those either). A more flexible leadership would in my view have seen sticking to these moves detrimental to the real core values and changed tactics.

It is hard to see Tesla’s mission or any core principle related to it would have been hurt if FSD would have shipped exactly as it had but without the pre-sold promises that misled customers. Or if many of the questionable demand levers — that clearly have also eroded customer trust — would simply have been replaced with advertising and traditional price action.

Now instead we have this mess where a lot of us just can’t really take Tesla seriously anymore while still paradoxically appreciating the basic product. That seems like a completely unnecessary predicament for a company which has such a solid underlying BEV mission and product — and a predicament more flexible leadership would in my view have avoided.

Sorry but it shows just how little you know about Elon and his company. Elon's priciple has always been to put all resources into engineering instead of advertising. In other words to make the best product instead of appearance of it.
 
Sorry but it shows just how little you know about Elon and his company. Elon's priciple has always been to put all resources into engineering instead of advertising. In other words to make the best product instead of appearance of it.

I don’t think my knowledge is the problem here. I simply think that this is an area where — we actually agree — Elon is consistent. Where we disagree is: I don’t think this consistency is a ’positive trait’ in this case.

Clearly if Elon felt Tesla succeeded without questionable demand levers this would be a moot point. But since he uses questionable demand levers so often he clearly feels Tesla needs those questionable demand levers — and this is the time where his old method is not working for him optimally at Tesla in my view.

A change in thought and action would have in my view been warranted. Replace questionable demand levers (like over-promising in product specs and chaotic price and product action, both of which have been increasing since late 2014 really) with more traditional tools that can drive demand and sales, and Tesla would be left with less trust-eroding antics yet still the same great product and product progress...

I think it would have made a overall better result for Tesla — and much better for customers. Just my opinion.
 
  • Funny
Reactions: CarlK
I think they used to do some sort of blacklisting of the map data near intersections, and they don't now, in preparation for allowing NoAP on city streets. So now we're seeing all the bugs revealed by the nav system not understanding the difference between a slower speed on the road above you and a slower speed on the road you're on. At least that's my best guess about what's happening.

I think that's not it, based on my drive this morning. I was going quite slowly through the problem this morning because of traffic, and the speed limit never blipped. I actually think this isn't about the map but about the logic that decides when you are on an exit ramp and slows you down. Going slowly, the car did not get confused, perhaps because the IMU data plus GPS at that speed was good enough to localize the car accurately to the highway all the way. At higher speed there is probably more uncertainty about where the car is at any time. Or perhaps it is not IMU+GPS but camera+NN; when you're going slowly the camera+NN doesn't get as confused? Anyway, it's a problem, and it's not my job to figure it out for them, but if it were only maps then the car would get confused at any speed.
 
(Another interesting idea, since AP is now standard, they can branch off NOA and smart summon into something like an upgraded autopilot for 5k and totally self driving into another 3k package. Totally self driving will enable the car to totally drive itself, with no person in the car, or needing to supervise it!)

That would certainly complete the cycle of redefining FSD as EAP and then offering "real" FSD separately. But they will still need to be cautious with wording. I suggest they call it Partially Totally Self Driving, or PTSD.
 
Stupid teenager should have known that Teslas only have full self driving capability on the highway!

“Well, we already have Full Self-Driving capability on highways. So from highway on-ramp to highway exiting, including passing cars and going from one highway interchange to another, Full Self-Driving capability is there."

Elon Musk - Jan 30th 2019
 
  • Informative
Reactions: rnortman
Stupid teenager should have known that Teslas only have full self driving capability on the highway!

“Well, we already have Full Self-Driving capability on highways. So from highway on-ramp to highway exiting, including passing cars and going from one highway interchange to another, Full Self-Driving capability is there."

Elon Musk - Jan 30th 2019

Yah, like Panasonic has 35GWh of cell generation capability at GF1, but they aren't making 35GWh of cells (yet)...
Teslas have had the (sensor) HW capability for FSD for years, but they do not currently have FSD.
Don't they still have the startup confirmation box that says stay alert and keep hands on wheel?
 
  • Like
  • Disagree
Reactions: CarlK and rnortman
Yah, like Panasonic has 35GWh of cell generation capability at GF1, but they aren't making 35GWh of cells (yet)...
Teslas have had the (sensor) HW capability for FSD for years, but they do not currently have FSD.
Don't they still have the startup confirmation box that says stay alert and keep hands on wheel?
Should be "the possibility of Full Self-Driving capability" :p
Just saying that Elon Musk doesn't help make it more clear to people that the system is nowhere near FSD.
 
  • Like
Reactions: rnortman
I think that's not it, based on my drive this morning. I was going quite slowly through the problem this morning because of traffic, and the speed limit never blipped. I actually think this isn't about the map but about the logic that decides when you are on an exit ramp and slows you down. Going slowly, the car did not get confused, perhaps because the IMU data plus GPS at that speed was good enough to localize the car accurately to the highway all the way.

I'm almost certain that it is caused by it thinking you were on the ramp. But what I'm saying is that I think they used to throw away the first fifty or a hundred feet of speed limit data on the ramp to prevent that from causing problems, and they no longer do. Just speculation, though. :)