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

Autopilot lane keeping still not available over 6 months after delivery

This site may earn commission on affiliate links.
Musk, in answering one of the questions about autopilot lane-keping, said something along the lines of lane-keeping working best when the Model S had a target car to follow, and not quite as well when the Model S was driving on its own, without a target car. (I'm restating this from memory, but that was the gist of it.)

I found that interesting. We've never heard that as a potential even temporary limitation of the lane-keeping autopilot. Thinking about it, and knowing what Musk said a week or two ago about the faint lines being an issue, it makes sense, but even so, I thought this was noteworthy.
I did not listen to the conference call, but if AP needs another car in front to help it do lane-keeping that will not be something to write home about.
 
Yep, I heard that, too. (And bravo to you for what is likely to be a futile attempt at refocusing this thread. :) ) I thought at the time that it was a really nice piece of transparency & spoke to some of the challenges they've been facing. Which also fit with his assertion that the software would get better with time, after release. (though from reading this thread, I suspect there are some that won't accept that, either...)

The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.

This is not, however, a field within my expertise, and I can see that it's an inherently complex problem, so I'm just thankful that others are responsible for solving it so that I don't have to. :)

Especially if you could just license a system from a sub-contractor like the Drive PX system. Tesla's differentiator is the EV thing and being a tech-driven platform. I'd rather they were a fast-integrator with other best-in-breed offerings and differentiated not by doing all of their own software, but rather having an elegant software system that can integrate well with others. Or at least, integrate for the really hard problems so they can focus their limited resources on differentiation.
 
The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.

I'm generally on the side of "Tesla misled us" in this thread. To be fair, and to address your post, I'd like to hope that one of the reasons this is hard is that Tesla isn't trying to replicate what any other company has done, but is trying to do it a heck of a lot better. For that I certainly give them credit. I can understand why it's taking longer than expected. And I give them the benefit of the doubt that they did not intend to mislead. I would like to see them do something--anything (well, almost anything)--to acknowledge the patience of those of us who had no choice but to wait.
 
Perhaps I should. Do you know if there is a link to the recording of the call?

Tesla always posts calls (if you didn't dial in to listen) on their Investor Relations site. Today's call is now posted, also. Tesla Motors, Inc. Second Quarter 2015 Financial Results QA Conference Call | Tesla Motors

- - - Updated - - -

The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.

Especially if you could just license a system from a sub-contractor like the Drive PX system. Tesla's differentiator is the EV thing and being a tech-driven platform. I'd rather they were a fast-integrator with other best-in-breed offerings and differentiated not by doing all of their own software, but rather having an elegant software system that can integrate well with others. Or at least, integrate for the really hard problems so they can focus their limited resources on differentiation.

I'm generally on the side of "Tesla misled us" in this thread. To be fair, and to address your post, I'd like to hope that one of the reasons this is hard is that Tesla isn't trying to replicate what any other company has done, but is trying to do it a heck of a lot better. For that I certainly give them credit. I can understand why it's taking longer than expected. And I give them the benefit of the doubt that they did not intend to mislead. I would like to see them do something--anything (well, almost anything)--to acknowledge the patience of those of us who had no choice but to wait.

The problem with just licensing someone else's system is that you are then limited to that system. I suspect, based on actions to date, that they have a much bigger picture in mind and are putting the building blocks in place towards that bigger picture.

Yes, it would be much faster to just plug in the Drive PX system. But it probably wouldn't be the fastest path to even more functionality. That's the only logical explanation.

Those of you here who have designed systems, knowing there are next generation versions yet to come, are familiar with leaving hooks in the software for the future - even if you don't need that today. It's the smart thing to do. I'm sure that's what is going on here. They have to do much of this from scratch if they want to implement the future versions the way they are envisioning.
 
This is the key. Some of you seem to think that Tesla is some sort of advanced research lab, kind of an MIT with an assembly line attached, seeking truth, scientific advancement, and clean transportation for all.

That's not what they are at all!

They did the demo and announcement when they did because they knew that the competition was already getting ahead of them. It was a classic move that Microsoft used to do to perfection in the 1980s and 1990s-- see what your competitors are working on, hack up a demo, and announce that you have the product done and are about to ship. Take all the fawning press coverage, etc., and then spend several years working to make it a reality while your competitors are already shipping and hardly anyone notices.

The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.
 
The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.
Much like, sending a man to the moon can't be THAT difficult, you point a rocket that-a-way, put a LOT of gunpowder in it's ass-end, some scuba tanks for the astranawts a Vooosh! Off ye go.

(Not making fun of you, Gizmoboy, people in general)

I am stunned at the lack of understanding in this thread. People on the "Get over it" side have stated, again and again, that we understand the "I've been mislead" group and empathize. But that doesn't make Tesla's work any easier.

Fitting one system into another system is not necessarily easy work. Especially when you've got custom code and hardware on one end and a demanding and intricate API on the other. And to Bonnie's point, there are certainly inherent limitations or structural demands that someone else's system adds to the mix. Although painful, sometimes the best route is to Develop It Here. And that takes a lot of time, debugging, effort, discipline and courage.

Was Tesla right to shoot the moon so early when they didn't know what it would take? No.
Is that unusual? No.
Does this satisfy WK's problem? No.
Will WK continuing to snipe and grouse about it make things go any quicker? No.
Will this wind up permanently damaging Tesla stock? No.
Does it even realistically damage the order pipeline? No.

This is tough stuff and the importance of accuracy in the code and system cannot be understated. If Tesla screws up on this one, the damage will be FAR greater than them being late on delivering it. I get disappointment. But I'm partial to them getting it right.
 
It's still amazing to me that all of this is still in the present tense on the website. Tesla has shown a willingness to change the website quickly and radically but something so fundamentally misleading has remained for months. I'm, frankly, more upset as a Tesla shareholder than as a P85D owner and driver.

Totally agree with this.
This is what really confuses me about the investors like jeffro/bonnie/etc, and others defending Tesla in this thread. Its not only a question if Tesla did mislead a bunch of people 10months ago or not. Or even how much misleading took place.

-The big question you should be asking yourselves is why are they still doing it?
-And even worse why are you OK with your beloved company conducting business in this manner?
-Why arent you pursuing Tesla in order to get them to change their webpage etc to protect themselves from anti-EV movements, shorts etc ready to jump all over this?

If you really are long on TSLA this should really scare you. Honestly I am surprised we havent seen more negative press about the 691hp claim and the missing autopilot yet.. I know lots of ICE-guys would love to laugh about the car not being close to the HP it was sold with and that the tech company not delivering on the promised tech. When any of this breaks into mainstream media the stock certainly would take a hit.

We have even seen CPO-listed cars without the asterix about autopilot now.... So its possibly getting worse even.
 
Direct quote from CC (Elon)

"That said, in terms of what the capability of the system is, I think it's capability in steering and control of acceleration and braking is excellent. When it has a tracking vehicle in front and you can basically have high confidence in steering, braking, and acceleration when basically when you're in some kind of traffic situation where there's a car on the road in front of you. I think it's pretty good in the absence of that, so if it's just lanes, it's pretty good, and it will get better over time as we refine the software. So I would certainly not take the initial version of Autopilot as the final version. It will just get better and better over time."
 
I've already said that if Tesla doesn't remedy the situation to my liking, which honestly is probably way less trouble and infinitely less expense for them than even a legal battle that I ultimately lose (which I'm far from convinced would be the case), then yes I will likely sue Tesla and yes it will likely end up being a class action case that rolls in all of the issues at hand as is well within my rights to do so.

Again, I am not taking any legal action today, nor do I foresee the need to do so personally in the future as it is much more likely Tesla will remedy this before it comes to that.

At this point, what would you consider sufficient remedies? Warranty extension? A partial refund? I agree that you definitely have a basis to sue, but if you did, what specifically would you seek to obtain? I don't think your complaints are unjustified, just curious what would you like to see happen.
 
You are going to be sorely mistaken and I hope you have enough money for all the legal fees your going to have to pay Tesla in response. Please pursue legal action, if anything it'll force you to stop whining here all of the time about how you didn't get what you feel you're entitled to. I'm not going to pick out each of your issues and explain to you why you'll loose, badly, but just beware you're looking at fighting a battle you simply can't win. Tesla will likely settle out of court to keep it's costs down though, so hey there's that. Although I can promise you that you'll never be allowed to buy one again...

Jeff

I'm sure there are class action lawsuit lawyears willing to pursue this at their own cost.
 
I'm sure there are class action lawsuit lawyears willing to pursue this at their own cost.

/RANT ON

And this is when things all go down the toilet.

Elon Musk and team are not doing this for money. Elon could have retired to an island a LONG time ago and anybody who has the capacity to make it as a key member of his team could doubtless be earning more elsewhere. These people are doing what they're doing because they have a passion and a mission - to advance sustainable transport and to advance automotive tech in a whole host of ways. I'm sure that ALL of these people are working absurd hours, are faced with severe financial constraints and are doing their damnedest to deliver high quality products & systems.

I think they are succeeding admirably. Yes there are problems and yes, they've probably made a few errors or over-promises. And yes some things are late. That comes with the personality type. People who have the guts and brains to reach beyond what exists inevitably fail to see the challenges sometime. That's the way that it works.

This ongoing whining, moaning and complaining that is starting to pervade some corners of this website HAS to be incredibly disheartening to the people who are driving like hell to get this done. And when people start to talk about lawsuits and such I'm sure some of them are ready to throw in the towel.

THIS is how companies like Tesla turn into companies like GM. Eventually people tire of the pissing and moaning and they lose the drive to push or innovate and they start to do the safe things. The enthusiastic engineers get pushed to the back room. The lawyers, accountants and PR people take over and they only commit to deliver what they can deliver without taking much risk or pushing too hard.

NOTHING can build enthusiasm like a group of equally enthusiastic supporters. And NOTHING can kill it like a group of whiners who see only downside and look for the failures.

Personally, I am tired about hearing about why my P85D doesn't deliver exactly 691HP. So bloody what? My P85D in amazing car, goes like stink and never fails to impress. And I'm tired about hearing that they're X months late on the autopilot. So bloody what?! I'll keep my hand on the steering wheel for a while longer while they work out the bugs and look forward to when it comes out. Being able to take my hands off the wheel for a bit is not going to change my life in any significant way.

Let's cut these guys some slack. Shooting off about lawyers, compensation and so-on isn't going to accomplish anything other than, perhaps, to damage the company. Elon & gang are putting things on the market that we shouldn't have seen for another decade.

Let's all have a beer and enjoy the ride.

/RANT OFF
 
Last edited:
The part that I find difficult to wrap my head around is that this feature already exists from other manufacturers without those limitations, and I just don't know what makes it so hard for Tesla to replicate.

This is not, however, a field within my expertise, and I can see that it's an inherently complex problem, so I'm just thankful that others are responsible for solving it so that I don't have to. :)

Especially if you could just license a system from a sub-contractor like the Drive PX system. Tesla's differentiator is the EV thing and being a tech-driven platform. I'd rather they were a fast-integrator with other best-in-breed offerings and differentiated not by doing all of their own software, but rather having an elegant software system that can integrate well with others. Or at least, integrate for the really hard problems so they can focus their limited resources on differentiation.
I fully agree with your statements. Tesla should focus on differentiating the company as the pioneer and best in class on BEV, and license other well developed features from other integrator or OEM to provide the bells and whistles.
 
I continually get the feeling (yep, just a feeling... no proof) that those most vocal about getting what they were promised are those that have produced the least amount of meaningful original value add. To echo what was said above - "cut these guys some slack".




What I found most interesting and telling about the conference call was the question regarding selling cars to Uber or simply providing services to the end user. I can just imagine the horned one on Elon's shoulder going "yea, that uber billion dollar valuation for an IPhone app; we will see how long that lasts :) "

Going after the autonomous taxi service business would explain the push for autopilot. Elon has said on many an occasion that it takes three fundamental revisions of any technology before it starts to show its true potential. The "soon" to be released version one starts that development clock and having an end goal/use for the technology means there will be continued focus and attention on the functionality. That bodes well for MS owners even if later revisions are paired with a change in the sensor suite.

As for other vendors carrying some of the load, the comments about high resolution GPS data were telling. Elon did not think there was an acceptable source so Tesla was rolling its own presumably using MS to collect that data. The continued free cellular connection makes more sense if Tesla is using the link to collect data for the company.
 
Last edited:
Just to be clear - is it your understanding that the cars with current gen autopilot hardware (not fully autonomous next-gen hardware) will be able to self-park and retrieve on private property?
I don't think the current AP H/W can do the above, simply because many other cars require more H/W to do less than the above. Parallel park is not the same as self park by backing into my garage. I am just an armchair speculator:wink:
 
Last edited:
I'm an owner in the not-so-upset camp. I'm disappointed, sure, but not yet as stressed as many of the rest of you. I'd really like to have the feature, I'm quite shocked a few months has turned into 9-10 months at best, probably a year given Elon Standard Time.

Just to be clear - is it your understanding that the cars with current gen autopilot hardware (not fully autonomous next-gen hardware) will be able to self-park and retrieve on private property?

This "feature" is a particularly interesting one to me. I told everyone I know when I bought the car that I didn't believe for one minute that they would actually deliver this function on my car. I'm not 100% sure why I decided that, other than that during the press conference it was pretty clear that Elon was pretty much making it up as he went along.

At this point, what would you consider sufficient remedies? Warranty extension? A partial refund? I agree that you definitely have a basis to sue, but if you did, what specifically would you seek to obtain? I don't think your complaints are unjustified, just curious what would you like to see happen.

This is kind of the line of thought that crystallized my thinking. I'm not sure what legal remedy I think an owner should be entitled to. The strongest would be to be made whole -- give the car back and get your money back. Not only do I think that solution is unlikely, I wouldn't take it even if they offered it to me. I don't want to give up my car, I just want it to be better. The more likely remedy is some kind of return of the difference between what you paid for and what you got. Presumably this would be limited to the extra cost of the autopilot suite. Except, I'd probably keep TACC for the same money or at least half of the money. Plus, maybe you should just be entitled to the interest you would have earned on that money in the year you didn't get the feature you wanted.

I do think it would be smart of Tesla to offer some kind of conciliatory gesture to owners for waiting a year, but I don't have any particularly helpful suggestions.

I'm sure there are class action lawsuit lawyears willing to pursue this at their own cost.

This is the nightmare scenario for Tesla. I can tell you how that one would end. We'd all get $500 credits toward future repairs and the lawyers would get $180 million in fees.
 
As for other vendors carrying some of the load, the comments about high resolution GPS data were telling. Elon did not think there was an acceptable source so Tesla was rolling its own presumably using MS to collect that data. The continued free cellular connection makes more sense if Tesla is using the link to collect data for the company.
Not sure about your reference to GPS data collection, but if you watch the Mobileye technical presentation, you might agree that Tesla and Google are trying to achieve autonomous driving from two opposite directions.