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

Updated autopilot sucks!

This site may earn commission on affiliate links.
I would actually say that 21.9 is more dangerous than not using AP. I actually turned it off this weekend.

My 5ft 3" wife could not put enough pressure on the steering wheel to stop the nanny alerts. It caused her to pull off the side of the road and reset the car so she could reuse the AP. So I tried - 6ft and 210lbs and my normal way of driving had the same issues. Finally the alerts were so annoying - I turned it off.

Sad - 3 years later and AP2 is a downgrade compared to AP1.
You’re clearly trying to spread misinformation and fear because of some agenda.

The amount of torque force required by autopilot to register a hold wheel is a fraction of the amount of force required to actually turn the steering wheel. If your wife and yourself can’t apply enough force then how would you ever be able to turn the wheel to begin with? You’re either doing it wrong, have an issue with your sensor, or more likely trying to spread negative misinformation.
 
  • Funny
Reactions: Shock-On-T
We've had many discussion on TMC about both hands not being recognized on the wheel, though, which would be the best and safest driving position to counter any irregularities. But it also means the steering wheel is held so steadily by two hands that the torque sensor simply can't regonize them.

I don't have the latest update yet, but I drive with one hand on AP because of this (someone suggested it to me on TMC last year and it worked pretty well) - and even then an occasional jug at the wheel is needed. I never, ever drive without hands, I always have one or two hands on the wheel and nags still happen in either case... more so with two hands.

It hasn't been a big deal, but to suggest the torque sensing works all the time would not be truthful. It does not sense normal two-hand, steering wheel holding very reliably, which would be the safest driving position.
 
You’re clearly trying to spread misinformation and fear because of some agenda.

The amount of torque force required by autopilot to register a hold wheel is a fraction of the amount of force required to actually turn the steering wheel. If your wife and yourself can’t apply enough force then how would you ever be able to turn the wheel to begin with? You’re either doing it wrong, have an issue with your sensor, or more likely trying to spread negative misinformation.

Clearly that is not the case. I have been the biggest fan of Tesla since 2012. My family owns 5 - I have talked at least 7 people into buying a Tesla and I will order a P100D model S in October.

My father bought the first P100D to replace his P85D - he hates the car so bad because the AP2 in his 100 is inferior to his AP1 - he cancelled two model 3s he had on order. I just replaced my P85D with a P100D.

I am pretty sure I love Tesla and Musk more than most people on this forum but the truth is the AP1 that I used for 60,000 miles and 3 years is way better than the AP2 on my new P100D X.

My only agenda - I want AP2 to be equal or better than my 3 year old AP1.
 
Drove 200 miles this morning with 2018.21.9. Didn't have any problem with the nag. I rested one hand on my wheel as usual when I was using autopilot. No, any problem!

To whom said eyes on the road is enough, you should know when you see something suddenly happens on the road, it takes at least a second for your brain to react. Then you have 1 second to your brain ask your hand move to the wheel, another second for hand move on the wheel, then another second to turn. Yes, that is 3 more seconds. You will have died!
 
  • Love
  • Like
Reactions: _jal_ and jeffro01
Clearly they aren't or you wouldn't get the nag. It's not really complicated or hard. I can't even remember the last time I got a nag, it's been so long.... Why? Because I keep one hand on the wheel at all times usually at the 8 o'clock position with my elbow resting on the drivers door. Zero nags, ever. Including on 2018.21.9...

Jeff
Umm.... Not necessarily.
 
  • Like
Reactions: MIT_S60
Clearly that is not the case. I have been the biggest fan of Tesla since 2012. My family owns 5 - I have talked at least 7 people into buying a Tesla and I will order a P100D model S in October.

My father bought the first P100D to replace his P85D - he hates the car so bad because the AP2 in his 100 is inferior to his AP1 - he cancelled two model 3s he had on order. I just replaced my P85D with a P100D.

I am pretty sure I love Tesla and Musk more than most people on this forum but the truth is the AP1 that I used for 60,000 miles and 3 years is way better than the AP2 on my new P100D X.

My only agenda - I want AP2 to be equal or better than my 3 year old AP1.
You also must hold the wheel with AP1. So, for you, what is so egregious about the current implementation of AP2 vs the current implementation of AP1. Please provide details of why AP1 is way better than 2 still.
 
Let me say - my new P100D is better than my old P85D in almost every way. Ride, sound, performance, space - I love it.

But in my observations - my drive to Laguna Seca Sunday in the X (a trip I have made dozens of times in my P85D) - AP2 is not as good.

The amount of nags with AP2 was a very significant annoying amount more than AP1. It was the first time AP was deactivated and my hand was always on the steering wheel! I also felt the AP1 did a better job with the curves near 156.
 
  • Like
Reactions: fablau and MIT_S60
Clearly that is not the case. I have been the biggest fan of Tesla since 2012. My family owns 5 - I have talked at least 7 people into buying a Tesla and I will order a P100D model S in October.

My father bought the first P100D to replace his P85D - he hates the car so bad because the AP2 in his 100 is inferior to his AP1 - he cancelled two model 3s he had on order. I just replaced my P85D with a P100D.

I am pretty sure I love Tesla and Musk more than most people on this forum but the truth is the AP1 that I used for 60,000 miles and 3 years is way better than the AP2 on my new P100D X.

My only agenda - I want AP2 to be equal or better than my 3 year old AP1.

Details please.

You make it seem as if AP2 doesnt even work next to AP1.

Even the MobileEye CEO says Tesla hit parity even if it was a left handed compliment.
 
Let me say - my new P100D is better than my old P85D in almost every way. Ride, sound, performance, space - I love it.

But in my observations - my drive to Laguna Seca Sunday in the X (a trip I have made dozens of times in my P85D) - AP2 is not as good.

The amount of nags with AP2 was a very significant annoying amount more than AP1. It was the first time AP was deactivated and my hand was always on the steering wheel! I also felt the AP1 did a better job with the curves near 156.
That's the thing though, this isn't an AP1 vs AP2 with being nagged to hold the wheel. It isn't a limitation of either AP1 or AP2. AP 1 also has increased nags, especially since it's original version of like 5 minutes or something between notifications. This is clearly a legal response to everyone being stupid with AP and the negative press that then puts on Tesla. Again, it's always been required that the driver be attentive and hold the wheel until such point that Tesla says it isn't required. Just hold the wheel people, we have been doing it for over a hundred years.
 
Details please.

You make it seem as if AP2 doesnt even work next to AP1.

I apologize if my post conveyed it didn't work. It works - it works well but I find it to not perform as smoothly on curved roads. With the speed set at 80 (yes I know that is above the speed limit). The AP1 car stayed in the lanes smooth and carried a consistent line through the corner. I found the AP2 car to have an inconsistent line and on some occasions drift off the preferred line through the corner.

Plus the nagging when my hand is on the wheel

Those two examples are why I feel AP2 does not perform equal to AP1.
 
Are you folks getting the nags holding with one hand or two?

Two hands on the wheel and AP won't detect them. One hand, on one side; it see the torque and you're good to go.

It encourages one-handed alertness.
Mine registers a hold either way, though, I assume if your applying exactly equal torque both ways and not hindering auto steer in the slightest then it wouldn't register
 
Just hold the wheel people, we have been doing it for over a hundred years.
What? Um... Yes you did... AP CLEARLY STATES that you are to keep your hands on the wheel while using it... Where in the world did you get the idea that you'd be excluded from this requirement...???

Jeff
I'm not convinced that you two aren't trolling, but if you're serious, you really need to get behind the wheel of a complainer's AP 2 Tesla, because false notifications are obviously a regular thing for most of us, myself included. My steering in comfort mode is also stiffer than most of the demo and loaner Teslas I have driven in comfort mode, so maybe it's just that there are a lot of variables here that the right people haven't gotten figured out yet.
 
I'm not convinced that you two aren't trolling, but if you're serious, you really need to get behind the wheel of a complainer's AP 2 Tesla, because false notifications are obviously a regular thing for most of us, myself included. My steering in comfort mode is also stiffer than most of the demo and loaner Teslas I have driven in comfort mode, so maybe it's just that there are a lot of variables here that the right people haven't gotten figured out yet.
What's trolling about saying hold the wheel? Holding the wheel is literally the point of the complaint. Even if you don't like it, AP isn't ready for us to not do so. Secondly, I never said false notifications are not a thing. I questioned the statement about someones 5' 3" wife not being able to put enough "pressure on the wheel". The amount of force required to register is a fraction of the amount of force required to turn the wheel and if that's not your experience, and instead requires more force than typical driving, you should probably take it to a service center to get it fixed because that is not typical of mine or any of the other Teslas I have driven.

So back to my comment about holding the wheel, and it's pretty simple, if AP is not usable for you because it's not registering you holding the wheel when you truly are then take it to a service center to get fixed. Then when it is fixed go back and make sure you are still holding the wheel because that is literally your responsibility as the driver of the vehicle until Tesla says you no longer need to hold the wheel and it becomes their responsibility or however that driverless vehicle liability law ends up being written.
 
What's trolling about saying hold the wheel?
The part where the false positives have already been discussed in the thread and "just hold the wheel" doesn't take care of them. However, I should have considered the possibility that you didn't read the entire thread since most people clearly don't. This, on the other hand:
if AP is not usable for you because it's not registering you holding the wheel when you truly are then take it to a service center to get fixed.
is a good point. I think most people just jiggle the wheel when they get the false positives and don't worry about it, and I suspect that's what most people that get the false positives do. For me, it's not worth the effort to deal with the service center on something infrequent enough (in my case) that there is a 95%+ chance they are going to say "could not reproduce" or "everything is within spec."
 
Which is of course, a perfectly reasonable and valid request... I was under the impression that feature/function parody is there if not slightly better than AP1 now...???

Jeff

Oh, feature and function parody is most definitely there. Function parity is still somewhat lacking. :p

(Though it's been getting closer and I've been told there are now some scenarios AP2+ handles better.)
 
  • Like
Reactions: jsmay311
I questioned the statement about someones 5' 3" wife not being able to put enough "pressure on the wheel". The amount of force required to register is a fraction of the amount of force required to turn the wheel and if that's not your experience, and instead requires more force than typical driving, you should probably take it to a service center to get it fixed because that is not typical of mine or any of the other Teslas I have driven.
I wanted to address this separately, I didn't quote the message where you were suggesting someone was spreading misinformation when I quoted you, and I wasn't responding to that. However, I am going to now. I do believe certain posts in this thread have some exaggeration to them, including the one you refer to here, but that exaggeration could come from half truths or one-off experiences. For instance, I'm 6'1' and can turn the wheel when steering assist isn't even powered (as happened to me during startup a couple times last year due to a firmware bug), but when I'm driving with EAP and trying to keep my Model X straight where it wants to veer into an on/off ramp, the counter-force I am providing with a firm grip with both hands often leads to a slight swerve away from said on/off ramp if EAP disengages. That is a normal thing in my Model X, and I have been in loaners where it disengages with next to no resistance.

As I mentioned a few posts back, my steering is stiff in general, I have had it to the service center a number of times where diagnosis was attempted on that specific issue and no issue was found (no less than three times). Today, on my first real EAP drive with 2018.21.9, my Model X beeped twice (I've never heard it beep before and there was no card on the IC when it did) and started swerving back and forth within the lane (to be clear I'm not sure if the beeps or the swerving were first), and my firm grip on the wheel trying to keep it straight did not disengage AP in spite of my fighting it. I'm not saying I couldn't have triggered a disengage with more force, but the amount of resistance I was providing most definitely should have disengaged it, however, I am hopeful that was a one-off experience that won't turn into a regular thing. So, while the premise is questionable (and I didn't challenge your questioning it in spite of my experience this morning), I'd suggest you might consider it could be an exaggeration as opposed to a fabrication.