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

[_jal_]

I don't think they knew how bad it was, otherwise they would have provided a way to disable TACC and enable CC.

This is where I bring up my wife's Outback again - cue the eyerolls - because it allows you to enable just CC apart from TACC. Tesla could have easily put a switch to enable TACC right under the Autosteer switch on the settings screen. If it was disabled, then you'd just have regular CC.

I use TACC all the time and the vast majority of the time it is great - but no better than the other systems I've used. I mean there's a limit to how awesome cruise control can be. But it is the only system I've had Sign Panic with. Sometimes it just slows down a bit. Sometimes it puts the brakes on, then proceeds. But on Sunday I was driving pretty much by myself on the interstate and there was an overhead sign. There display showed the red car which was actually the car way ahead on the road, and the brakes went on. I slowed from 75 to 45 before I punched the accelerator.

I'm going to make a bracket for the car and GoPro every drive I do just to keep a record for fun to see exactly what is going on.

 

[disagree]

If you track back through some of the twisted interpretations of what most would consider Plain English contractual language, irrational advocate aka FanBoy seems just about right. Funny that juvenile and idiot don't seem to qualify as name calling in the FanBoy lexicon.

EinSV's latest move to call it irrational to take words for what they mean is irrefutable, though.

We've always been at war with Eastasia.

 

[disagree]

This is where I bring up my wife's Outback again - cue the eyerolls - because it allows you to enable just CC apart from TACC. Tesla could have easily put a switch to enable TACC right under the Autosteer switch on the settings screen. If it was disabled, then you'd just have regular CC.

I use TACC all the time and the vast majority of the time it is great - but no better than the other systems I've used. I mean there's a limit to how awesome cruise control can be. But it is the only system I've had Sign Panic with. Sometimes it just slows down a bit. Sometimes it puts the brakes on, then proceeds. But on Sunday I was driving pretty much by myself on the interstate and there was an overhead sign. There display showed the red car which was actually the car way ahead on the road, and the brakes went on. I slowed from 75 to 45 before I punched the accelerator.

I'm going to make a bracket for the car and GoPro every drive I do just to keep a record for fun to see exactly what is going on.

It is bizarre how this is happening routinely to Chicago drivers, when we know that we are covering these routes regularly and the cars are supposedly learning.

 

[_jal_]

We've always been at war with Eastasia.

Fake news. Oceania. It's Oceania we've always been at war with.

 

[_jal_]

It is bizarre how this is happening routinely to Chicago drivers, when we know that we are covering these routes regularly and the cars are supposedly learning.

It's the people on the coast trying to kill the midwest elite.

 

[NerdUno]

I'm driving back to Bowling Green just to see if driving through a war zone will fix things.

 

[_jal_]

It is bizarre how this is happening routinely to Chicago drivers, when we know that we are covering these routes regularly and the cars are supposedly learning.

Oh, BTW, @disagree, I get them EB on the Kennedy by the Harlem el stop. There is a pedway there that makes it kind of hesitate. But I've had it happen with electronic signs and with just the regular signs too. I'm not sure, but it seems like it happens when light reflects off the signs or structure just right because I also had it happen when a stop light changed to yellow.

But it's like watching for a falling star. You never know when it is going to happen. So I just make sure no one is tailing me whenever I use it unless I'm in stop and go traffic.

 

[ABC2D]

OK, I think many of you are irrationally advocating against EAP and Tesla's rollout of it. I will also say that I am skeptical that those posting on this thread were surprised by the way this is being rolled out.

Irrationally, huh. You're funny.

OK, I think many of you are irrationally advocating against EAP and Tesla's rollout of it. I will also say that I am skeptical that those posting on this thread were surprised by the way this is being rolled out.

Also, for those of you who like to throw around words like "fraud" you might consider the all-too-common phenomenon known as the Planning Fallacy, in which "predictions about how much time will be needed to complete a future task display an optimism bias and underestimate the time needed." Planning fallacy - Wikipedia

Most of us are guilty of succumbing to the Planning Fallacy. Given that Tesla is doing things that no one has tried before, accurately predicting timelines is especially difficult. I think a fair criticism of Tesla is that its predicted timelines are often too optimistic. But to say this is fraudulent is ridiculous IMO.

Also, for those of you who like to throw around words like "Planning Fallacy" please tell me since when buying a product requires customers to familiarize themselves with "Planning Fallacies" phenomenon

See, those "Irrational" arguments above usually prompt the use of "fanboy" term.

 

[EinSV]

I hate to be a spoilsport but we have seen this movie before (with AP1) so let me tell you how it ends.

(1) AP(2) is released. A great hue and cry rises up -- "unsafe" "Tesla BAD!!!"

(2) 1 or 2 years later -- data comes out showing AP(2) is significantly safer than an average human driver.

(3) Not one person who was shouting "unsafe" and "Tesla BAD" at the top of their lungs will acknowledge they were wrong, and that AP(2) avoided accidents, reduced injuries and will save lives.

Sorry to ruin the suspense.

 

[disagree]

I'm beginning to think that guy was playing a trick on us by defining fanboy.

 

[Joe F]

I hate to be a spoilsport but we have seen this movie before (with AP1) so let me tell you how it ends.

(1) AP(2) is released. A great hue and cry rises up -- "unsafe" "Tesla BAD!!!"

(2) 1 or 2 years later -- data comes out showing AP(2) is significantly safer than an average human driver.

(3) Not one person who was shouting "unsafe" and "Tesla BAD" at the top of their lungs will acknowledge they were wrong, and that AP(2) avoided accidents, reduced injuries and will save lives.

Sorry to ruin the suspense.

You don't know how it ends, unless you're clairvoyant. Not a personal attack; Just keeping it real.

(1) Right now, an AP2 car is unsafe. As Elon said "be safe". Tesla BAD, as you portray what's being argued here, are you words.

(2) Sorry, not sorry. Read tea leaves much?

(3) Not shouting here, but warning all who drive an AP2 Tesla (do you?) that, right now, you can be involved in a serious rear end collision when AP decides to react to, not sure how to put this politely, invisible objects perhaps, that causes the car to attempt a panic stop. Now I'm sure said accident will probably be blamed by some here on the driver that rear ends the Tesla, and legally that's true, but when was the last time you experienced a car in front of you slam on its brakes for no sane reason?

 

[Bladerskb]

I hate to be a spoilsport but we have seen this movie before (with AP1) so let me tell you how it ends.

(1) AP(2) is released. A great hue and cry rises up -- "unsafe" "Tesla BAD!!!"

(2) 1 or 2 years later -- data comes out showing AP(2) is significantly safer than an average human driver.

(3) Not one person who was shouting "unsafe" and "Tesla BAD" at the top of their lungs will acknowledge they were wrong, and that AP(2) avoided accidents, reduced injuries and will save lives.

Sorry to ruin the suspense.

They weren't wrong, the system based on your own analogy would have improved in 1-2 years to make it competent and safe.
Bro your logic is beyond warped.

 

[EinSV]

You don't know how it ends, unless you're clairvoyant. Not a personal attack; Just keeping it real.

(1) Right now, an AP2 car is unsafe. As Elon said "be safe". Tesla BAD, as you portray what's being argued here, are you words.

(2) Sorry, not sorry. Read tea leaves much?

(3) Not shouting here, but warning all who drive an AP2 Tesla (do you?) that, right now, you can be involved in a serious rear end collision when AP decides to react to, not sure how to put this politely, invisible objects perhaps, that causes the car to attempt a panic stop. Now I'm sure said accident will probably be blamed by some here on the driver that rear ends the Tesla, and legally that's true, but when was the last time you experienced a car in front of you slam on its brakes for no sane reason?

If actual data comes out showing AP2 is less safe than an average driver, I will happily admit I was wrong. I think the chances of that happening are extremely low based on Tesla's track record and the careful roll-out of AP2.

If it comes out the other way -- which I think it almost certainly will -- I will be interested to see whether you or the others claiming AP2 is unsafe will acknowledge that you were wrong. My bet is that is a big "NO" but would be happy to be proven wrong.

 

[ABC2D]

If actual data comes out showing AP2 is less safe than an average driver, I will happily admit I was wrong. I think the chances of that happening are extremely low based on Tesla's track record and the careful roll-out of AP2.

If it comes out the other way -- which I think it almost certainly will -- I will be interested to see whether you or the others claiming AP2 is unsafe will acknowledge that you were wrong. My bet is that is a big "NO" but would be happy to be proven wrong.

RIGHT NOW AP2 is UNSAFE - why is it wrong to say that?

When in 2 year, or some other time, AP2 becomes safe(r) - why would it make all who had called AP2 unsafe to be wrong as AP2 was indeed unsafe at that time?

Bro your logic is beyond warped.

That ↑↑↑↑↑

 

[Joe F]

If it comes out the other way -- which I think it almost certainly will -- I will be interested to see whether you or the others claiming AP2 is unsafe will acknowledge that you were wrong. My bet is that is a big "NO" but would be happy to be proven wrong.

So, by inference, you're position is that an AP2 Tesla, speed controlled by TACC, IS SAFE when, for no apparent reason, it slams on the brakes hard enough to trigger anti lock brakes, right?

You really need to read more carefully what others are writing if you expect to be taken seriously.

 

[EinSV]

RIGHT NOW AP2 is UNSAFE - why is it wrong to say that?

When in 2 year, or some other time, AP2 becomes safe(r) - why would it make all who had called AP2 unsafe to be wrong as AP2 was indeed unsafe at that time?

That ↑↑↑↑↑

The fearmongering on AP1 reached a fever pitch last summer after the Josh Brown accident -- long after AP1 was first activated. It continued up until the day the NHTSA report issued showing a 40% reduction in airbag-triggering accidents after AP1 was enabled. Since then the critics have quieted down quite a bit. Here is the data in case you haven't seen it:

It turned out that the individual, anecdotal reports of problems with AP1, some undoubtedly well intentioned and some not, were completely unreliable in terms of predicting overall system safety. In fact, they were just plain wrong.

A 40% reduction in accidents is a safety leap of historic proportions -- statistically as revolutionary a safety benefit as seat belts and air bags assuming it is confirmed by further data. And AP2 is a much more robust system and is very likely to provide even greater safety benefits than AP1.

Tesla has the data and is gradually enhancing AP2's capabilities as the data shows it is safe to do so. It would be against its own interest to do otherwise. Since anecdotal reports have in the past proven totally unreliable in predicting safety of AP1, I trust Tesla's data and track record more than I do anecdotal reports. I should also note that there are many anecdotal reports of AP2 doing very well, although there is still plenty of room for improvement, which should be expected at this early stage.

 

[stefan.bc]

AP2 does well until it doesn't. Here's a video of someone almost getting rear ended. Imagine that at 75mph not 50.

 

[ABC2D]

The fearmongering on AP1 reached a fever pitch last summer after the Josh Brown accident -- long after AP1 was first activated. It continued up until the day the NHTSA report issued showing a 40% reduction in airbag-triggering accidents after AP1 was enabled. Since then the critics have quieted down quite a bit. Here is the data in case you haven't seen it:

It turned out that the individual, anecdotal reports of problems with AP1, some undoubtedly well intentioned and some not, were completely unreliable in terms of predicting overall system safety. In fact, they were just plain wrong.

A 40% reduction in accidents is a safety leap of historic proportions -- statistically as revolutionary a safety benefit as seat belts and air bags assuming it is confirmed by further data. And AP2 is a much more robust system and is very likely to provide even greater safety benefits than AP1.

Tesla has the data and is gradually enhancing AP2's capabilities as the data shows it is safe to do so. It would be against its own interest to do otherwise. Since anecdotal reports have in the past proven totally unreliable in predicting safety of AP1, I trust Tesla's data and track record more than I do anecdotal reports. I should also note that there are many anecdotal reports of AP2 doing very well, although there is still plenty of room for improvement, which should be expected at this early stage.

Comparing AP1 with AP2 is .... ludicrous: different sensors, different hardware, different software - EVERYTHING IS DIFFERENT.

Will AP2 be better? Probably, but not right now. So, let's keep this discussion to present time

 

[Joe F]

AP2 does well until it doesn't. Here's a video of someone almost getting rear ended. Imagine that at 75mph not 50.

Interesting. Thanks, hadn't seen that. Red car showing up under the overpass. Had that happen to me too.

Unfortunate the driver left the center seat upright. Would have had a better shot of the car behind him reacting.

 

[EinSV]

Comparing AP1 with AP2 is .... ludicrous: different sensors, different hardware, different software - EVERYTHING IS DIFFERENT.

Will AP2 be better? Probably, but not right now. So, let's keep this discussion to present time

My post was intended to point out that the anecdotal data of the type people are using in this thread (and others) has essentially zero value in assessing the overall safety of AP2, now or in the future. So those on this thread who are claiming AP2 is unsafe because they think AP2 did not behave as it should are making assertions that they can't support in any meaningful way. And, in the very recent past in the closest analogous situation available (AP1) similar claims were proven to be incorrect.

Tesla has the data and has every incentive to ensure that AP2 is safe. Their careful rollout with AP2 is consistent with that. I believe that those who are claiming otherwise based on anecdotal data will be proven wrong -- just like those who made similar claims for AP1 right up to the issuance of the NHTSA report. Tesla has the data and the track record with AP1 to back them up. The track record for people claiming AP is "unsafe" based on their own experience, on the other hand, is very poor.