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

FIRMWARE UPDATE! AP2 Local road driving...and holy crap

This site may earn commission on affiliate links.
Sorry those statements are not supported by the data from the NHTSA and IIHS.

The NHTSA reported a 40% decrease in serious accidents after AP1 was enabled. The IIHS's extensive analysis of AEB and FCW covering five manufacturers' systems (Mercedes, Volvo, Acura, Honda and Subaru) over a total of four years concluded that the overall decrease in accidents from both AEB and Forward Collision Warning Systems was only 6%, which did not even reach the level of statistical significance. http://orfe.princeton.edu/~alaink/S...s/IIHS-CicchinoEffectivenessOfCWS-Jan2016.pdf at page 15.

I think the confusion arises because the IIHS report also concluded that AEB plus FCW reduced the number of rear-end collisions by 39%. But rear-end collisions are only about 25% of total collisions, and the IIHS only looked at 1/2 of rear-end collisions -- those in which the car with AEB/FCW rear-ended another car.

Bottom line: the IIHS did an extensive analysis and found that other manufacturers' FCW/AEB systems reduced overall accidents by only 6% -- so small that it was statistically insignificant. So the data just don't support your conclusion that the 40% reduction in serious accidents achieved by AP1 is something that has been obtained by other manufacturers' collision avoidance systems. The IIHS looked deeply into this very issue and came to a different conclusion. In fact, the 40% reduction in serious accidents the NHTSA reported for AP1 is many times higher than the accident avoidance the IIHS found was achieved by other manufacturers' FCW/AEB systems.

How is it that you don't understand?
that stats is meanless because it doesn't validate ap driving behavior.

IF everyone with ap2 were to activate autopilot on local roads for more than 10 seconds they would be dead.
that's a guarantee. how is that safe? why can't you look past your fandom and see this for what it is?
you come off as one of those people who will cover up things to further your bottom line.

you could careless about safety aslong as you can continue your tesla fandom.

Do you need more videos?

how about a guy who nearly died 5 times
after watching that, how clueless can u be?
 
One of the biggest mysteries to me is why clear, ongoing communication to owners about current AP2 development status, exactly what to expect and when, and so on is nonexistent.

It is clear that there are knowledgeable people on the top and this just makes it even more confounding.

A lot of bad experiences, sour taste, and customer angst can be relieved with simple, honest, ongoing communication. And you don't even need an MBA to know that.

Given all that, the lack of communication augurs more concern because knowing all that why are they afraid to speak?

[/tinfoil time]

But really! Can anyone suggest a cogent answer without belittling the question and attacking the desire for basic decent communication ?
 
One of the biggest mysteries to me is why clear, ongoing communication to owners about current AP2 development status, exactly what to expect and when, and so on is nonexistent.

IMG_0051.JPG


Transparency and hype are mutually exclusive in the Tesla realm.
 
One of the biggest mysteries to me is why clear, ongoing communication to owners about current AP2 development status, exactly what to expect and when, and so on is nonexistent.
Does Tesla ever communicate ANYTHING properly? Look at the complete mess of the audio player.

But really! Can anyone suggest a cogent answer without belittling the question and attacking the desire for basic decent communication ?
Tesla is bad at communications?
 
Tesla is bad at communications?

Clearly so, but I'm surprised since they seem to be at the highest level in other areas (...cough, cough...not mentioning fit and finish, some customer service issues, and did I mention communication?)

Elon is sharp and with it and so are probably those who work in the upper offices making decisions about these kinds of issues, hence the surprise when I hear this aloof way of interfacing with customers has been going on since 2012.

It's just a shame to see so much unnecessary displeasure in areas that could be relatively easy to fix.

Of course the experience overall score is a 10 out of 10 in other areas, but is it so hard to step up with the communication to make it a 10 out of 10 all around?
 
  • Like
Reactions: neroden
Of course the experience overall score is a 10 out of 10 in other areas, but is it so hard to step up with the communication to make it a 10 out of 10 all around?
Apparently it is hard for them. I think the problem is something like too many engineers, not enough English majors... if you see what I mean. I'm hopeful hiring Andrea James will help.
 
That is, of course, the problem. When the orders are off the hook like this they don't get much pressure to fix the communications problem. It'll bite them in the ass in about 10 years, when all cars will be electric and they'll have to compete with other electric cars. Until then, it just won't matter to the bottom line...
 
That is, of course, the problem. When the orders are off the hook like this they don't get much pressure to fix the communications problem. It'll bite them in the ass in about 10 years, when all cars will be electric and they'll have to compete with other electric cars. Until then, it just won't matter to the bottom line...

It's just surprising, for such a large company it takes very little effort, relatively speaking, to communicate.

How hard could it be already?

And as Tesla is Elon's own baby, doesn't he personally care about what matters a lot to his customers?
 
One of the biggest mysteries to me is why clear, ongoing communication to owners about current AP2 development status, exactly what to expect and when, and so on is nonexistent.

It is clear that there are knowledgeable people on the top and this just makes it even more confounding.

A lot of bad experiences, sour taste, and customer angst can be relieved with simple, honest, ongoing communication. And you don't even need an MBA to know that.

Given all that, the lack of communication augurs more concern because knowing all that why are they afraid to speak?

[/tinfoil time]

But really! Can anyone suggest a cogent answer without belittling the question and attacking the desire for basic decent communication ?
Caveat: what I am saying below only applies to software development and not necessarily other aspects that may require communication (like hardware development, design, and marketing, which may follow a different development cycle).

Giving an honest public predictive timetable for software updates with specific predicted features is rarely a good idea (a regular release cycle without specific predicted features is another story). If you ask a software engineer to give their honest estimate of how long it will take to develop something, it will often take 2-3x as long. Nowadays, some engineers learn to pad their estimates, but sometimes even that is not enough.

The nature of software development is that there is almost always unforeseen bugs and issues (even without feature creep adding to the burden). And because they are unforeseen, time estimates easily become invalid. Some features also run into problems that can not be fixed in a reasonable amount of time and need to be dropped.

Most likely with such a timetable, you set multiple false expectations and when those expectations are not met (either a deadline or feature), then that leads to massive disappointment for the consumer (multiple broken promises rather than just a general one).

Another thing to understand is that Tesla is operating under the model that the car continually improves, so there is really no specified "end" point in which the software is deemed its final release. That means releases of features deemed "acceptable" in quality (which is subjective) can be released just to get features into consumer hands, and further refined in future updates.

As the company matures and if the uncertainty becomes an issue in terms of sales, I can see them move to a regular release cycle (similar to car model years or mobile OS releases), for example annual major release, quarterly sub-release, etc. This lets people know for certain when a major update will be released, but not what features would be included.

Elon is mostly the one giving out time estimates, but the nature of how he releases them (buried on twitter and earnings calls) makes them softer estimates. Elon's current estimate for AP2 development is high speed autosteer in 2 weeks to a month:
Software Update Brings Autopark To Tesla's Equipped With Enhanced Autopilot
 
@stopcrazypp
I agree with all the above except, when you release something that does not work as expected, you need let your client know what the expectations are, and it is better to release later, with less features that at least work as expected.
This looks more and more like a marketing decision and I will not blame any SW developer for it. I can only blame the people up top that push these kind of things just to have a release. It feels to me like Tesla is becoming more like GM, a brand I would never buy because of their ethics.

I am more pissed about them telling us something was done(poorly) than telling us they are still working on it, on something we will soon enjoy and be proud of. I would have preferred seeing an auto rain sensor, a simple CC, line departure functionality done right. Like most, I bought the car for what it will be and because I think electric cars can make a difference.
 
Last edited:
From an historical perspective I remember being displeased with how AP1 performed on local roads. Anytime it lost a line it would go hunting that way, and it couldn't track correctly in mild corners. I was mostly happy with freeway performance except for the whole diving for an exit thing.

Now I didn't write a bunch of posts here being upset with it, nor did I create a bunch of youtube videos. I didn't because my expectations were set pretty low. The people that did post various fail videos I simply wrote off as being nuts, and expecting too much from a single camera system.

AP1 did get better over time, but I didn't actually use it that much. All it takes is a few scary incidents to dissuade a person from using a feature. As time went on the number of uses case I even considered using it for rapidly dropped. I also had concerns along the lines of how AP would impact my own driving. Where I felt like TACC made my driving safer, and AP made my driving worse. This was over thousands of miles of analyzing how it changed my freeway driving. With any technology people need to be conscious of how it changes them or the story of you will end up on Black Mirror.

For me AP was more about the adventure/experience of it than anything else. It was a fun adventure and I'm glad I got to take apart in it.

Fast forward today and the reality is completely different for people with AP2. The expectations are set WAY above what my expectations were set, and the real tragedy of it is that it had ZERO chance of matching up to the expectations. It's hard for most of us to grasp the engineering challenge of taking a brand new system, and matching the performance of the old one let along exceeding it.

This wasn't what Tesla really wanted. They wanted to put MobileEye+HW2 into a car, and then slowly enable features on the HW2 as they became available. So we know from this that Tesla knew how big of a task this was for them to pull off. I have no idea why MobileEye didn't allow for this. MobileEye wanted to be rid of Tesla anyways, and unlike what someone else said there was no attempt to have the MobileEye teach the Tesla Vision anything. That's not how these things work. If what Tesla is reporting is true then MobileEye acted completely irresponsible. How is any other manufacture going to trust them not to pull the same damn thing? I would avoid them like the plague because it takes a lot of work to build a platform around someone's product. No wonder NVidia has so many manufactures signing up.

In any case the situation we find ourselves is the result of a lot of things coming together. It was a perfect storm so to speak.

Now you can pick who you want to blame for this. But, nothing is going to change the fact that AP2 owners are left with a choice. The choice being to actively use/report issues with AP or stay on the sidelines for the next 3--6 months and wait for all this to settle down.

Long term I do expect the situation to improve a lot. It's probably going to be some mix of tempered expectations plus it getting better. Hopefully within 6+ months it will actually be able to excite again.
 
Last edited:
I completely agree. I think that some here are either blindly allegiant, or have stock holdings that bias them toward overlooking the unpleasant truth: Tesla's autopilot, on its own without Mobileye, is so far a disaster.

Fully agree :) Four months back, here is what I, the clairvoyant @mmd, said, and got bunch of dislikes :( Now, people agree it is actually AP 0.2, not even AP 1/2. An expensive and sophisticated piece of hardware without working software is no more than a big piece of rock.
But we can see why Tesla did this mislabeling. Without AP 2.0, Tesla would have missed Q4 sales goals by a much wider margin.

Short-Term TSLA Price Movements - 2016
You are asking way too many relevant questions. How about using the Occam's razor, that the sudden end of MobileEye-Tesla relationship has forced Tesla to redo AP w/o MobilEye in a hurry? To make it sound good, call the fresh start version 2.0 instead of version 0.5. We also don't know for sure, who has the right to the data collected using MobilEye systems.
 
Last edited:
  • Love
Reactions: u00mem9
Oh man, in some of those videos, I wouldn't even try with my AP1. I mean, one of them looks like pretty much a 90 degree left turn...and he's suprised it did not navigate that?! And the windy-road at night business... yeah good luck getting my AP1 to work on that either. I only use AP1 on normal straight-ish roads, with traffic going one way. While I'm sure it can do single-lane no-divider opposing traffic type driving... do you have a death wish? And it is a definite no that my AP1 car could do that drive with the centre island things. For certain.

All this being said, AP2 was advertised as some sort of super-improvement over AP1. So far, it looks like the same or maybe a little worse because it isn't enabled for highway yet.
 
It's just surprising, for such a large company it takes very little effort, relatively speaking, to communicate.

How hard could it be already?

And as Tesla is Elon's own baby, doesn't he personally care about what matters a lot to his customers?

You must be new to Tesla :) The missed communications only happen when the news is negative :(
It could be very hard, if it hurts the sales or bottom line.
 
  • Like
Reactions: u00mem9
Really? This is me on a local road yesterday with the latest release 17.7.2.


drshan, my Tesla with AP 1.0 would not have handled those quick turns at 30 mph. You are asking too much of the autopilot. When fully autonomous driving comes, there will need to be speed controls that limit the speed to an appropriate speed for those roads. As long as the driver is allowed to choose the speed, there is no guarantee of successful outcome. Get to know the current capabilities of your autopilot and choose a speed which is appropriate if using the lane-keeping feature.
 
  • Like
Reactions: wesley888
One of the biggest mysteries to me is why clear, ongoing communication to owners about current AP2 development status, exactly what to expect and when, and so on is nonexistent.

Tesla's communication strategy:

Angry Mob: "Where's AP1?"
Elon: "Hey look over there. SUMMON!"
Angry Mob: "Ooh. Shiny." Forgets about AP1 for a while.

Angry Mob: "Why doesn't summon work to the door like you promised?"
Elon: "Hey look over there. 0-60 in 2.8 seconds."
Angry Mob: "Ooh. Shiny." Forgets for a while.

Angry Mob: "Why's AP2 so far behind AP1?"
Elon: [knows he has nothing] ....
Angry Mob: Still Angry
 
  • Funny
Reactions: mmd