Tesla Autopilot, worth it?

Questions on how we spend our money and our time - consumer goods and services, home and vehicle, leisure and recreational activities
squirm
Posts: 1306
Joined: Sat Mar 19, 2011 11:53 am

Tesla Autopilot, worth it?

Post by squirm » Sun May 20, 2018 1:18 pm

Wife and I have been considering a model 3. One reason I'd go for the Tesla is the supercharger network. However when considering the options, I'm not sure if the autopilot is worth it. I'm very particular about driving, in that I always want to be in full control, so I'm not sure how autopilot will work with someone like me... The other issue is judging from YouTube videos it always seems to be a hit or miss whether the car will take a curve correctly or slow when it should....

I was thinking maybe it's best to rent one for a week to find out, however I have a feeling that's not a long enough time to get the feeling for it without it being a novelty item.

stan1
Posts: 5558
Joined: Mon Oct 08, 2007 4:35 pm

Re: Tesla Autopilot, worth it?

Post by stan1 » Sun May 20, 2018 1:22 pm

If you are entertaining this question it seems like the cost of the autopilot should be well under 1% of your net worth. I'd treat it as a toy or hobby and get it only if you'd find it enjoyable.

TX_Drew
Posts: 33
Joined: Wed May 16, 2018 11:37 am

Re: Tesla Autopilot, worth it?

Post by TX_Drew » Sun May 20, 2018 2:03 pm

I’d get the autopilot. It will help on those long trips between charger stations.

But be mindful...seems some highly publicized wrecks lately, perhaps due to autopilot. So it’s still new tech.

User avatar
nisiprius
Advisory Board
Posts: 35895
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Tesla Autopilot, worth it?

Post by nisiprius » Sun May 20, 2018 2:18 pm

We have the package of ordinary new safety and driver-assist features on our ordinary 2016-model car: pre-collision sensing, late departure alert, blind spot sensing, and so forth. I think it's worthwhile. Three features we find really valuable are the blind-spot sensing, the "rear cross traffic" alert, and the "adaptive cruise control," which is wonderful--it means you can actually use cruise control when you're on normally-busy highways.

I'm not sure what features you get on a Tesla without autopilot. I'd probably get it and just consider it to be an enhanced version of the electronic-assist features you can now get in every other car.

The thing is that I think it's kind of a scam their calling it autopilot (but complaining when drivers expect it to live up to the name), and suggesting that it is capable of becoming a full autonomous car eventually through a software upgrade. "A simple matter of software" is almost a joke in software engineering circles.

I believe that self-driving cars cannot match human visual perception and will not soon. I think this is one of those problems where it is easy to get 90% of the way there and almost impossible to get the last 10%. I think self-driving cars need to make up for nonhuman perception by having superhuman senses. I think one of the reasons adaptive cruise control works so well is the car's radar is better at judging the relative speed between me and the car ahead than I am.

Almost-but-not-quite-autonomous seems like the worst of both worlds, and dangerous.

All of the serious self-driving cars currently have an ugly bump on the top for LIDAR. Tesla has a story about why they don't use LIDAR and don't think it helps, but it doesn't sound convincing to me.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

PFInterest
Posts: 1393
Joined: Sun Jan 08, 2017 12:25 pm

Re: Tesla Autopilot, worth it?

Post by PFInterest » Sun May 20, 2018 2:22 pm

yes get it.
but if you dont even have your deposit in......

cuihang
Posts: 9
Joined: Sun Apr 05, 2015 1:08 pm

Re: Tesla Autopilot, worth it?

Post by cuihang » Sun May 20, 2018 3:18 pm

I’d say get it. I’ve been using autopilot for almost 3 years on a model s (first gen hardware). It really reduced a lot of stress on the highway especially for long commute. But the key thing is you don’t totally rely on it and still pay attention to the road. It’s been amazing to witness its improvement over time by firmware update. I love driving too and having the feature doesn’t prevent me from having fun.

Longdog
Posts: 1066
Joined: Sun Feb 09, 2014 6:56 pm
Location: Philadelphia

Re: Tesla Autopilot, worth it?

Post by Longdog » Sun May 20, 2018 3:36 pm

While I think it’s a cool feature to have, I cannot reconcile the claim that it makes driving safer while at the same time you must pay full attention to the road and be able to take over immediately if it goes astray. In other words, it keeps you safe but you cannot rely on it to keep you safe. There is a report on the Internet about a woman who recently survived a terrible crash with her Tesla while using autopilot. It speaks highly to the safety of Tesla vehicles, but Tesla’s response identifies all the times the woman took her hands off the steering wheel, implying that because she didn’t have her hands on the wheel she was not being attentive and it’s not the fault of autopilot. I actually like Tesla quite a lot and am super impressed by what they’ve accomplished under Elon’s leadership. But I cannot help but feel that with regard to autopilot, they are trying to have their cake and eat it too. “It’s a great driver assist feature well worth $5,000!” but “If something goes wrong with it, it’s your fault for not paying attention and taking over.” Huh? Treat it as an experimental (beta) feature, and if that’s worth the money for you then get it.

I will not be ordering it for my Model 3. You can add it after the fact if you like, for an additional $1,000.
Steve

User avatar
Offshore
Posts: 382
Joined: Wed Apr 09, 2008 9:43 am
Location: Michigan's West Coast

Re: Tesla Autopilot, worth it?

Post by Offshore » Sun May 20, 2018 3:44 pm

I have the differing opinion of "not now", and here's why. If you are like me, you are unwilling to shell out the cost of a model S, so you wait for the model 3 that fits your wishes. The cost of EA is $5000. at time of configuration, or $6000. later. Even at $5k, EA is about a 15% option. For me, I want to stay true to my goals, buy a Tesla and control cost. Add EA later, if desired.

I've received my configuration email about a month ago, but am waiting for the $35,000. vehicle.

PhilosophyAndrew
Posts: 310
Joined: Sat Aug 13, 2016 10:06 am

Re: Tesla Autopilot, worth it?

Post by PhilosophyAndrew » Sun May 20, 2018 3:56 pm

Longdog wrote:
Sun May 20, 2018 3:36 pm
While I think it’s a cool feature to have, I cannot reconcile the claim that it makes driving safer while at the same time you must pay full attention to the road and be able to take over immediately if it goes astray. In other words, it keeps you safe but you cannot rely on it to keep you safe.
I don’t think there is any contradiction in the claim that driving assistance safety features both make driving safer but don’t allown drivers to pay no attention. Likewise, I see no contradiction in the claim that these features make driving safer and yet sometimes fail.

I think of it this way:

We know that humans often fail as drivers. If any of these new technologies help prevent or lessen the severity of an accident caused by human failure, they make driving safer. That these new technologies have limitations and don’t help in every such situation just means that they are imperfect. But why expect perfection?

We know that these technologies will sometimes fail. If a human driver is paying attention, he or she may be able to navigate through a dangerous situation. If, however, the technology fails and the humans also fail to pay attention, then the worst outcomes will occur.

As long as the technologies have a reasonably low failure rate and drivers don’t foolishly choose to abrogate their responsibilities, the technologies should make driving safer. The car I drive (a Mercedes) has the full suite of currently-available features — I do not perceive them as tempting me towards inattention and have experienced them as supporting me when I drive.

My spouse did not order the Autopilot for her Model 3, but this because she perceives it as an unwanted complication. I understand her perspective even as I love the similar technologies in my car.

Andy.

Nate79
Posts: 2574
Joined: Thu Aug 11, 2016 6:24 pm
Location: Portland, OR

Re: Tesla Autopilot, worth it?

Post by Nate79 » Sun May 20, 2018 4:08 pm

Where can you rent a Model 3?

User avatar
jabberwockOG
Posts: 1281
Joined: Thu May 28, 2015 7:23 am

Re: Tesla Autopilot, worth it?

Post by jabberwockOG » Sun May 20, 2018 4:34 pm

Folks with an engineering level of grounding in the internals of hardware and software of AD technology and the challenges involved understand that AD is still far from perfect at this point and will likely remain relatively rudimentary especially in exception situations far into the future. The system can seem like magic and appear to work perfectly when environmental and traffic conditions are just right. But exceptions will certainly occur due to a multitude of factors that can completely fool even a very sophisticated AD system for a couple of seconds and at 70 mph even tiny momentary mistakes can be 100% catastrophic if the driver is not fully aware. As long as the driver completely understands that they cannot trust an AD system even for a second they should be fine however most drivers will understandably be lulled into excessive trust and complacency because when the AD system is operating in good conditions it appears to be so very sophisticated/powerful/magic/etc.

Longdog
Posts: 1066
Joined: Sun Feb 09, 2014 6:56 pm
Location: Philadelphia

Re: Tesla Autopilot, worth it?

Post by Longdog » Sun May 20, 2018 4:56 pm

PhilosophyAndrew wrote:
Sun May 20, 2018 3:56 pm
Longdog wrote:
Sun May 20, 2018 3:36 pm
While I think it’s a cool feature to have, I cannot reconcile the claim that it makes driving safer while at the same time you must pay full attention to the road and be able to take over immediately if it goes astray. In other words, it keeps you safe but you cannot rely on it to keep you safe.
I don’t think there is any contradiction in the claim that driving assistance safety features both make driving safer but don’t allown drivers to pay no attention. Likewise, I see no contradiction in the claim that these features make driving safer and yet sometimes fail.

I think of it this way:

We know that humans often fail as drivers. If any of these new technologies help prevent or lessen the severity of an accident caused by human failure, they make driving safer. That these new technologies have limitations and don’t help in every such situation just means that they are imperfect. But why expect perfection?

We know that these technologies will sometimes fail. If a human driver is paying attention, he or she may be able to navigate through a dangerous situation. If, however, the technology fails and the humans also fail to pay attention, then the worst outcomes will occur.

As long as the technologies have a reasonably low failure rate and drivers don’t foolishly choose to abrogate their responsibilities, the technologies should make driving safer. The car I drive (a Mercedes) has the full suite of currently-available features — I do not perceive them as tempting me towards inattention and have experienced them as supporting me when I drive.

My spouse did not order the Autopilot for her Model 3, but this because she perceives it as an unwanted complication. I understand her perspective even as I love the similar technologies in my car.

Andy.
Andy, I don’t think a normal driver can sit with his hand in his laps, relaxed and enjoying the freedom that auto-steering affords, and instantly recognize a catostraphic situation and take over from autopilot. That immediate context switch, in my opinion, is not possible, especially when you’ve come to trust its behavior in so many situations. I think to be able to do that requires keeping your hands on the wheel at all times. And then.... what’s the point?
Steve

Longdog
Posts: 1066
Joined: Sun Feb 09, 2014 6:56 pm
Location: Philadelphia

Re: Tesla Autopilot, worth it?

Post by Longdog » Sun May 20, 2018 5:04 pm

jabberwockOG wrote:
Sun May 20, 2018 4:34 pm
Folks with an engineering level of grounding in the internals of hardware and software of AD technology and the challenges involved understand that AD is still far from perfect at this point and will likely remain relatively rudimentary especially in exception situations far into the future. The system can seem like magic and appear to work perfectly when environmental and traffic conditions are just right. But exceptions will certainly occur due to a multitude of factors that can completely fool even a very sophisticated AD system for a couple of seconds and at 70 mph even tiny momentary mistakes can be 100% catastrophic if the driver is not fully aware. As long as the driver completely understands that they cannot trust an AD system even for a second they should be fine however most drivers will understandably be lulled into excessive trust and complacency because when the AD system is operating in good conditions it appears to be so very sophisticated/powerful/magic/etc.
Exactly! And why would someone who fully understands you can’t trust it for even one second shell out $5,000?
Steve

seawolf21
Posts: 217
Joined: Tue Aug 05, 2014 7:33 am

Re: Tesla Autopilot, worth it?

Post by seawolf21 » Sun May 20, 2018 5:06 pm

AutoPilot is not auto pilot per se. The driver will still have to maintain constant vigilance with eyes on the road and be prepare to take over if AutoPilot gets confused due to suboptimal road conditions.

One can pretty much conclude the Tesla incidents were preventable had drIver paid attention and be aware the situation AutoPilot was were about get them into.

wrongfunds
Posts: 1687
Joined: Tue Dec 21, 2010 3:55 pm

Re: Tesla Autopilot, worth it?

Post by wrongfunds » Sun May 20, 2018 5:36 pm

Nate79 wrote:
Sun May 20, 2018 4:08 pm
Where can you rent a Model 3?
I think Turo but be aware that rent will be steep; couple of hundreds for a day but if it prevents you from making $60K "mistake" then it would be worth it :-)

Personally, if you are buying it just as an EV vehicle, then it probably is not worth paying for the autopilot. Unfortunately Tesla does NOT sell adaptive cruise control as a cheaper option. If you want that, then you take the auto pilot. I can certainly see the novelty wearing off pretty quickly. It would be great to impress your friends and colleagues when you are proudly showing off your baby. Personally, I don't even use regular cruise control on my vehicle even on a 14hr straight drive.

Having said that, I will still purchase auto pilot because my Tesla 3 purchase has very little to do with logic but much more to do with the cachet of Tesla and emotion. At least I have accepted that.

PhilosophyAndrew
Posts: 310
Joined: Sat Aug 13, 2016 10:06 am

Re: Tesla Autopilot, worth it?

Post by PhilosophyAndrew » Sun May 20, 2018 6:18 pm

Longdog wrote:
Sun May 20, 2018 4:56 pm

Andy, I don’t think a normal driver can sit with his hand in his laps, relaxed and enjoying the freedom that auto-steering affords, and instantly recognize a catostraphic situation and take over from autopilot. That immediate context switch, in my opinion, is not possible, especially when you’ve come to trust its behavior in so many situations. I think to be able to do that requires keeping your hands on the wheel at all times. And then.... what’s the point?
When auto-steering exists, the context switching you describe may be a serious problem. However, that problem doesn’t arise with the current technology, which merely supports drivers who remain responsibly for driving their own cars.

The steeeing support may help me make a turn that I’m late making, but no context switching g is required because there is never a time when that technology justifies my taking my hands off the wheel. The point of adding this assistance is just what I described before: minimizing or sometimes preventing bad consequences of human drivers’ errors.

Andy.

User avatar
jabberwockOG
Posts: 1281
Joined: Thu May 28, 2015 7:23 am

Re: Tesla Autopilot, worth it?

Post by jabberwockOG » Sun May 20, 2018 10:34 pm

seawolf21 wrote:
Sun May 20, 2018 5:06 pm
AutoPilot is not auto pilot per se. The driver will still have to maintain constant vigilance with eyes on the road and be prepare to take over if AutoPilot gets confused due to suboptimal road conditions.

One can pretty much conclude the Tesla incidents were preventable had drIver paid attention and be aware the situation AutoPilot was were about get them into.
I think the real danger is when the AD system fails when the operator thinks conditions are totally optimal. That is a real possibility with any autonomous system because humans and machines perceive environments and complex situations so differently.

The danger early on as these systems come into general use is that everybody understands that AD system is going to be less reliable at night in rain, snow, super heavy traffic etc. But there are more than a few other conditions and exception situations, some of which can very subtle and may occur in what appears to be absolutely normal or even optimal conditions, where an AD system can momentarily struggle and fail and at speed it only takes a couple of seconds of inattention for some very bad things to happen.

Almost certainly the general public will learn over time to pay better attention (maintaining situational awareness - a term familiar to aircraft operators) because these systems will occasionally fail at the worst possible time and if the driver is not fully and immediately attentive, people will get hurt.

golfCaddy
Posts: 504
Joined: Wed Jan 10, 2018 10:02 pm

Re: Tesla Autopilot, worth it?

Post by golfCaddy » Mon May 21, 2018 7:06 pm

PhilosophyAndrew wrote:
Sun May 20, 2018 3:56 pm
As long as the technologies have a reasonably low failure rate and drivers don’t foolishly choose to abrogate their responsibilities, the technologies should make driving safer.
This seems like an optimistic assumption. It's difficult enough to convince people to stop using their smart phones, when they're expected to drive normally. It seems hopeless to convince people not to text and drive, when driving means waiting for the 0.01% case where the autopilot fails and the human is expected to takeover. Even the hypothetically responsible drivers will have a tendency to lose focus on the road, when steering, speed control, and braking is automated.

User avatar
Leif
Posts: 2408
Joined: Wed Sep 19, 2007 4:15 pm

Re: Tesla Autopilot, worth it?

Post by Leif » Mon May 21, 2018 7:18 pm

I have a Tesla Model S with enhanced autopilot (EAP) and full self drive (FSD). This is sometimes called Autopilot 2.0.

I think the EAP is too expensive for what you get. I do use it, particular on trips. However, the best part is the traffic aware cruise control(TACC), as Tesla calls it. Others call it adaptive cruise control. It uses a radar to control your car's speed based on traffic. That works fine. The exception is when a car cuts in front of you it is slow to react. If I'm controlling speed I will immediately slow down to establish a safe following distance. The lane changing works most of the time. It use to work about 50% of the time. Now I would say it works 90% of the time. When it does not work I need to disable EAP, make the lane change, then reengage EAP. Not really worth much to me. Lane keeping works better then in the past. It rarely fails. However, on a recent trip it did want to take me from the highway to an off ramp. I had to pull it back. Of course this generated questions from my passenger.

Some of the features that were suppose to be part of EAP are still missing. So in summary the TACC is not worth $5,000 in my Model S. I would guess even less so for the Model 3 if you consider the relative prices. I might pay $1,000 for the TACC.

So far I've got nothing for my $3,000 for FSD. That is total vaporware. The worst part is the reason I got EAP was that it is necessary to FSD.
Last edited by Leif on Mon May 21, 2018 7:37 pm, edited 3 times in total.
Investors should diversify across many asset-classes so that whatever happens, we will not have all our investments in underperforming asset classes and thereby fail to meet our goals-Taylor Larimore

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Mon May 21, 2018 7:24 pm

Personally, I would not buy a Tesla without it. I've been using AP1 on Model S since December 2015, and have loved it. Have been using AP2 on a newer Tesla since December 2017, and I'd say it's on par with AP1 if not a bit better after an update a month or two ago. And it keeps getting better.

For my 2017, I went for the enhanced AP, but not the full self driving, since I'm skeptical that the latter will become a reality for a long time, and it'll only cost me an extra $1K to get it if it does become real.

As others have said, where AP shines is on long freeway drives--it eliminates the need to hold the steering wheel tightly, and therefore reduces fatigue, and I'm able to use it without taking control like 99% of the time. Every now and then it does something funky, and I feel like I have to take control to maintain safety, so it's true that you can't abdicate responsibility for driving.

Perhaps also already mentioned, but it also can reduce fatigue in stop and go traffic, although sometimes it stops a bit more abruptly than I would.

Others can debate the safety or usefulness of it. I love it, and am very comfortable using it.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

PhilosophyAndrew
Posts: 310
Joined: Sat Aug 13, 2016 10:06 am

Re: Tesla Autopilot, worth it?

Post by PhilosophyAndrew » Mon May 21, 2018 7:24 pm

golfCaddy wrote:
Mon May 21, 2018 7:06 pm
PhilosophyAndrew wrote:
Sun May 20, 2018 3:56 pm
As long as the technologies have a reasonably low failure rate and drivers don’t foolishly choose to abrogate their responsibilities, the technologies should make driving safer.
This seems like an optimistic assumption. It's difficult enough to convince people to stop using their smart phones, when they're expected to drive normally. It seems hopeless to convince people not to text and drive, when driving means waiting for the 0.01% case where the autopilot fails and the human is expected to takeover. Even the hypothetically responsible drivers will have a tendency to lose focus on the road, when steering, speed control, and braking is automated.
Yes, time will tell about this as these technologies advance.

My Mercedes is engineered to present the current implementation as I’ve described — an assistive backstops protecting alert drivers. Other implementations may not be engineered so well, or people may not interact with the technologies as intended. Also, the context switching problem may become acute when we start to see cars with fully autonomous driving capabilities.

About this last point: I think you are over-estimating the extent to which current technologies amount to autonomous driving. We are nowhere near the point where the driver can safely fail to pay attention 99.99% of the time. The current technologies are the exact opposite of that; they are designed to protect drivers who are alert 99.99% of the time — they are there to help with the .01% situations where an alert driver does not react to a dangerous situation; the current technologies do not automate routine driving, as you seem to assume.

Andy.

User avatar
4nursebee
Posts: 976
Joined: Sun Apr 01, 2012 7:56 am
Location: US

Re: Tesla Autopilot, worth it?

Post by 4nursebee » Mon May 21, 2018 7:36 pm

If it were me and I wanted to make an informed decision, I would go test drive the closest thing available. The internet is full of mis informed and click baiters.
4nursebee

golfCaddy
Posts: 504
Joined: Wed Jan 10, 2018 10:02 pm

Re: Tesla Autopilot, worth it?

Post by golfCaddy » Tue May 22, 2018 6:52 pm

PhilosophyAndrew wrote:
Mon May 21, 2018 7:24 pm
About this last point: I think you are over-estimating the extent to which current technologies amount to autonomous driving. We are nowhere near the point where the driver can safely fail to pay attention 99.99% of the time. The current technologies are the exact opposite of that; they are designed to protect drivers who are alert 99.99% of the time — they are there to help with the .01% situations where an alert driver does not react to a dangerous situation; the current technologies do not automate routine driving, as you seem to assume.

Andy.
Here's one description of level 2:
Consumer Guide® Nutshell: The vehicle will essentially drive itself, but the driver is required to remain aware of driving conditions and be prepared to resume control at short notice. Automation may not be available in all traffic conditions.

anoop
Posts: 688
Joined: Tue Mar 04, 2014 1:33 am

Re: Tesla Autopilot, worth it?

Post by anoop » Tue May 22, 2018 6:59 pm

This is my view on Tesla's implementation of autopilot.
http://anoopsplace.blogspot.com/2018/04 ... pilot.html

PhilosophyAndrew
Posts: 310
Joined: Sat Aug 13, 2016 10:06 am

Re: Tesla Autopilot, worth it?

Post by PhilosophyAndrew » Tue May 22, 2018 7:11 pm

golfCaddy wrote:
Tue May 22, 2018 6:52 pm
PhilosophyAndrew wrote:
Mon May 21, 2018 7:24 pm
About this last point: I think you are over-estimating the extent to which current technologies amount to autonomous driving. We are nowhere near the point where the driver can safely fail to pay attention 99.99% of the time. The current technologies are the exact opposite of that; they are designed to protect drivers who are alert 99.99% of the time — they are there to help with the .01% situations where an alert driver does not react to a dangerous situation; the current technologies do not automate routine driving, as you seem to assume.

Andy.
Here's one description of level 2:
Consumer Guide® Nutshell: The vehicle will essentially drive itself, but the driver is required to remain aware of driving conditions and be prepared to resume control at short notice. Automation may not be available in all traffic conditions.
This description reads like media hype -- at least, it doesn't reflect how Mercedes has engineered these technologies into its most recent cars. (Mercedes itself initially marketed the technology along those lines, but stopped doing so because this characterization is inaccurate.)

Andy.

Strummer
Posts: 45
Joined: Thu Dec 28, 2017 4:15 pm

Re: Tesla Autopilot, worth it?

Post by Strummer » Tue May 22, 2018 9:46 pm

I think the answer to the question depends on your driving patterns. If you have a daily freeway commute with a lot of stop/start freeway traffic, you'll really enjoy being able to turn on either Traffic Aware Cruise Control or AP and have the car handle the braking and accelerating for you. Either setting will slow when traffic slows, all the way down to a complete stop, and accelerate to your defined limit when it's moving. TACC does that; AP will keep you between the lines, too. (Please note that I said "freeway commute." If you're on surface streets, neither TACC nor AP will notice when a stoplight turns red, so you're on your own in those circumstances.)

Autopilot will also provide automated parallel parking in some conditions. That's a novelty but it's fun to press a couple buttons and watch the car park itself.

User avatar
Cycle
Posts: 470
Joined: Sun May 28, 2017 7:57 pm
Location: Minneapolis, USA

Re: Tesla Autopilot, worth it?

Post by Cycle » Tue May 22, 2018 10:22 pm

Absolutely, consider it an investment in self driving cars. We need data to show the safety of self driving cars so governments can more quickly outlaw humans from driving cars. According to the national safety council, one has a 1/98 chance of dying in a car wreck in their lifetime.

User avatar
Leif
Posts: 2408
Joined: Wed Sep 19, 2007 4:15 pm

Re: Tesla Autopilot, worth it?

Post by Leif » Wed May 23, 2018 10:39 am

Strummer wrote:
Tue May 22, 2018 9:46 pm
Autopilot will also provide automated parallel parking in some conditions. That's a novelty but it's fun to press a couple buttons and watch the car park itself.
Have you tried it? I rarely parallel park, but I would like to try it some time. Have you ever experienced a case when the "P" lights up, but parallel parking is not possible? It has happened to me sometimes. I'll be in local traffic with a couple of lanes of traffic at a stop light. Then the "P" comes on. That means the car has recognized that it can park. You need to press the break and put the car in reverse. But I wonder what my car is thinking. There is not parking opportunity, or even much space, as the cars are waiting for the light to turn green.

I guess that is called a false positive.
Investors should diversify across many asset-classes so that whatever happens, we will not have all our investments in underperforming asset classes and thereby fail to meet our goals-Taylor Larimore

hicabob
Posts: 2659
Joined: Fri May 27, 2011 5:35 pm
Location: cruz

Re: Tesla Autopilot, worth it?

Post by hicabob » Wed May 23, 2018 10:54 am

nisiprius wrote:
Sun May 20, 2018 2:18 pm


All of the serious self-driving cars currently have an ugly bump on the top for LIDAR. Tesla has a story about why they don't use LIDAR and don't think it helps, but it doesn't sound convincing to me.


I think the omission of LIDAR is perhaps for cost saving? I've seen the mapping produced by a lidar an associate of mine was using (albeit for a slower moving device) and it looked really, really good relative to what can be deduced from cameras. Cheap LIDAR should happen.

MotoTrojan
Posts: 1631
Joined: Wed Feb 01, 2017 8:39 pm

Re: Tesla Autopilot, worth it?

Post by MotoTrojan » Wed May 23, 2018 10:57 am

Add it later on. It’ll extend that new car feeling. It’s just a software change.

Get the dual motor for $5K more. More range and performance/reliability.

Enjoy!

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Wed May 23, 2018 6:43 pm

MotoTrojan wrote:
Wed May 23, 2018 10:57 am
Add it later on. It’ll extend that new car feeling. It’s just a software change.
A software change plus $1,000 more than if you get it when you buy.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

MotoTrojan
Posts: 1631
Joined: Wed Feb 01, 2017 8:39 pm

Re: Tesla Autopilot, worth it?

Post by MotoTrojan » Wed May 23, 2018 6:46 pm

Kevin M wrote:
Wed May 23, 2018 6:43 pm
MotoTrojan wrote:
Wed May 23, 2018 10:57 am
Add it later on. It’ll extend that new car feeling. It’s just a software change.
A software change plus $1,000 more than if you get it when you buy.

Kevin
Ah. Fair.

Strummer
Posts: 45
Joined: Thu Dec 28, 2017 4:15 pm

Re: Tesla Autopilot, worth it?

Post by Strummer » Fri May 25, 2018 9:20 am

Leif wrote:
Wed May 23, 2018 10:39 am
Strummer wrote:
Tue May 22, 2018 9:46 pm
Autopilot will also provide automated parallel parking in some conditions. That's a novelty but it's fun to press a couple buttons and watch the car park itself.
Have you tried it? I rarely parallel park, but I would like to try it some time. Have you ever experienced a case when the "P" lights up, but parallel parking is not possible? It has happened to me sometimes. I'll be in local traffic with a couple of lanes of traffic at a stop light. Then the "P" comes on. That means the car has recognized that it can park. You need to press the break and put the car in reverse. But I wonder what my car is thinking. There is not parking opportunity, or even much space, as the cars are waiting for the light to turn green.

I guess that is called a false positive.
I have used it a couple times and found it useful. I've also seen what you describe — the P lights up on the Tesla's control panel, indicating that it's ready to park if you want to initiate it. I can see why that happens when, for example, there's a line of cars stopped in a left-turn lane and I pull past them in the lane next to them. The car would see that as a line of stationary cars with a parking-space-sized gap between two of them, and would then offer you a chance to park there. As long as you don't initiate the parking sequence, which takes a couple steps, you'll be fine — because why would you initiate it then? The cars are smart but, with any luck, the people driving them should be smarter.

Longdog
Posts: 1066
Joined: Sun Feb 09, 2014 6:56 pm
Location: Philadelphia

Re: Tesla Autopilot, worth it?

Post by Longdog » Fri May 25, 2018 7:29 pm

Strummer wrote:
Fri May 25, 2018 9:20 am
Leif wrote:
Wed May 23, 2018 10:39 am
Strummer wrote:
Tue May 22, 2018 9:46 pm
Autopilot will also provide automated parallel parking in some conditions. That's a novelty but it's fun to press a couple buttons and watch the car park itself.
Have you tried it? I rarely parallel park, but I would like to try it some time. Have you ever experienced a case when the "P" lights up, but parallel parking is not possible? It has happened to me sometimes. I'll be in local traffic with a couple of lanes of traffic at a stop light. Then the "P" comes on. That means the car has recognized that it can park. You need to press the break and put the car in reverse. But I wonder what my car is thinking. There is not parking opportunity, or even much space, as the cars are waiting for the light to turn green.

I guess that is called a false positive.
I have used it a couple times and found it useful. I've also seen what you describe — the P lights up on the Tesla's control panel, indicating that it's ready to park if you want to initiate it. I can see why that happens when, for example, there's a line of cars stopped in a left-turn lane and I pull past them in the lane next to them. The car would see that as a line of stationary cars with a parking-space-sized gap between two of them, and would then offer you a chance to park there. As long as you don't initiate the parking sequence, which takes a couple steps, you'll be fine — because why would you initiate it then? The cars are smart but, with any luck, the people driving them should be smarter.
Wow, that would be a disaster if you attempted to "park" in that situation. It does show that there is a certain context we have when driving that goes beyond the mere placement and current movement of vehicles. There are other adjustments I've recognized I make when driving, when I observe certain behavior in a nearby car or even certain vehicle types that I know will either accelerate faster or slower than I expect to or will block my view. If it is safe, I try to position myself to my advantage, such as for better visibility, better safety (e.g., staying away from cars that I recently observed weaving in an out of traffic), better access to an upcoming turn or lane adjustment, etc.
Steve

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Fri May 25, 2018 7:47 pm

Longdog wrote:
Fri May 25, 2018 7:29 pm
Wow, that would be a disaster if you attempted to "park" in that situation.
Sure, but who would do that??? I've seen this a number of times, and I just chuckle.
It does show that there is a certain context we have when driving that goes beyond the mere placement and current movement of vehicles. There are other adjustments I've recognized I make when driving, when I observe certain behavior in a nearby car or even certain vehicle types that I know will either accelerate faster or slower than I expect to or will block my view. If it is safe, I try to position myself to my advantage, such as for better visibility, better safety (e.g., staying away from cars that I recently observed weaving in an out of traffic), better access to an upcoming turn or lane adjustment, etc.
Exactly, and that's when you take over from AP. I use it daily, but there is no way it's even close to fully autonomous driving yet, and I take over whenever it exceeds the parameters of how I would drive beyond a certain limit.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Leif
Posts: 2408
Joined: Wed Sep 19, 2007 4:15 pm

Re: Tesla Autopilot, worth it?

Post by Leif » Fri May 25, 2018 10:50 pm

Kevin M wrote:
Fri May 25, 2018 7:47 pm
Longdog wrote:
Fri May 25, 2018 7:29 pm
Wow, that would be a disaster if you attempted to "park" in that situation.
Sure, but who would do that??? I've seen this a number of times, and I just chuckle.

Kevin
To answer your rhetorical question, no one. But it does not fill me with confidence that it suggests it can park between cars that probably have 3 or 4 feet of separation.
Investors should diversify across many asset-classes so that whatever happens, we will not have all our investments in underperforming asset classes and thereby fail to meet our goals-Taylor Larimore

DanMahowny
Posts: 298
Joined: Sun Aug 06, 2017 8:25 pm

Re: Tesla Autopilot, worth it?

Post by DanMahowny » Sat May 26, 2018 10:57 am

Cycle wrote:
Tue May 22, 2018 10:22 pm
According to the national safety council, one has a 1/98 chance of dying in a car wreck in their lifetime.
Preposterous. Absurd. Ridiculous.

User avatar
HomerJ
Posts: 11208
Joined: Fri Jun 06, 2008 12:50 pm

Re: Tesla Autopilot, worth it?

Post by HomerJ » Sat May 26, 2018 11:01 am

squirm wrote:
Sun May 20, 2018 1:18 pm
Wife and I have been considering a model 3. One reason I'd go for the Tesla is the supercharger network. However when considering the options, I'm not sure if the autopilot is worth it. I'm very particular about driving, in that I always want to be in full control, so I'm not sure how autopilot will work with someone like me... The other issue is judging from YouTube videos it always seems to be a hit or miss whether the car will take a curve correctly or slow when it should....

I was thinking maybe it's best to rent one for a week to find out, however I have a feeling that's not a long enough time to get the feeling for it without it being a novelty item.
You've got plenty of time to think about it. I doubt you'll be able to buy one for 2-3 years if you don't already have a reservation.
The J stands for Jay

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Sat May 26, 2018 12:27 pm

Leif wrote:
Fri May 25, 2018 10:50 pm
Kevin M wrote:
Fri May 25, 2018 7:47 pm
Longdog wrote:
Fri May 25, 2018 7:29 pm
Wow, that would be a disaster if you attempted to "park" in that situation.
Sure, but who would do that??? I've seen this a number of times, and I just chuckle.

Kevin
To answer your rhetorical question, no one. But it does not fill me with confidence that it suggests it can park between cars that probably have 3 or 4 feet of separation.
I honestly never have paid attention to the separation in non-parking situations. I know that I've never been offered the option to auto-park without sufficient separation in an actual parking situation, so that still doesn't concern me.

However, I have been offered the option, and have accepted it, in situations where the separation was small enough that I probably wouldn't have attempted to park on my own, but usually in these situations I have abandoned the spot after parking, since I don't want to get dinged by someone in front of or behind me when they try to leave. Also in these situations I've wished for auto-unpark, since it's usually a bit difficult to get out of such a tight parking spot.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

squirm
Posts: 1306
Joined: Sat Mar 19, 2011 11:53 am

Re: Tesla Autopilot, worth it?

Post by squirm » Sat May 26, 2018 1:21 pm

HomerJ wrote:
Sat May 26, 2018 11:01 am
squirm wrote:
Sun May 20, 2018 1:18 pm
Wife and I have been considering a model 3. One reason I'd go for the Tesla is the supercharger network. However when considering the options, I'm not sure if the autopilot is worth it. I'm very particular about driving, in that I always want to be in full control, so I'm not sure how autopilot will work with someone like me... The other issue is judging from YouTube videos it always seems to be a hit or miss whether the car will take a curve correctly or slow when it should....

I was thinking maybe it's best to rent one for a week to find out, however I have a feeling that's not a long enough time to get the feeling for it without it being a novelty item.
You've got plenty of time to think about it. I doubt you'll be able to buy one for 2-3 years if you don't already have a reservation.
Yes, I know...This is might be a purchase that won't be until 3 years or so. I'm also reading a lot of issues regarding the phone key, personally I'd rather just have a fob. A fob works 100% of the time, unless the battery in it is dead.

squirm
Posts: 1306
Joined: Sat Mar 19, 2011 11:53 am

Re: Tesla Autopilot, worth it?

Post by squirm » Sat May 26, 2018 1:27 pm

After reading the responses, I think we'll do without the Autopilot. I like the suggestions to go for the dual motors and purchase the autopilot later as a separate update, even it costs a grand more. Honestly, it seems the correct way to use autopilot is on a fairly uncrowded open stretch of freeway...however driving in that situation is the least stressful. Realistically all I'd want it to do is to help me watch the car in front in stop and go traffic...that's the time when you're moving along and all of a sudden traffic stops quickly, then I start swearing. Braking is all I need, not steering.

squirm
Posts: 1306
Joined: Sat Mar 19, 2011 11:53 am

Re: Tesla Autopilot, worth it?

Post by squirm » Sat May 26, 2018 1:27 pm

squirm wrote:
Sat May 26, 2018 1:21 pm
HomerJ wrote:
Sat May 26, 2018 11:01 am
squirm wrote:
Sun May 20, 2018 1:18 pm
Wife and I have been considering a model 3. One reason I'd go for the Tesla is the supercharger network. However when considering the options, I'm not sure if the autopilot is worth it. I'm very particular about driving, in that I always want to be in full control, so I'm not sure how autopilot will work with someone like me... The other issue is judging from YouTube videos it always seems to be a hit or miss whether the car will take a curve correctly or slow when it should....

I was thinking maybe it's best to rent one for a week to find out, however I have a feeling that's not a long enough time to get the feeling for it without it being a novelty item.
You've got plenty of time to think about it. I doubt you'll be able to buy one for 2-3 years if you don't already have a reservation.
Yes, I know...This is might be a purchase that won't be until 3 years or so. Depends how much our current car will go. I'm also reading a lot of issues regarding the phone key, personally I'd rather just have a fob. A fob works 100% of the time, unless the battery in it is dead.

User avatar
TomatoTomahto
Posts: 7307
Joined: Mon Apr 11, 2011 1:48 pm

Re: Tesla Autopilot, worth it?

Post by TomatoTomahto » Sat May 26, 2018 1:56 pm

Longdog wrote:
Sun May 20, 2018 4:56 pm
Andy, I don’t think a normal driver can sit with his hand in his laps, relaxed and enjoying the freedom that auto-steering affords, and instantly recognize a catostraphic situation and take over from autopilot. That immediate context switch, in my opinion, is not possible, especially when you’ve come to trust its behavior in so many situations. I think to be able to do that requires keeping your hands on the wheel at all times. And then.... what’s the point?
I keep a (single) hand on the wheel at all times; I find that sufficient to take complete control in an instant if I need to. The point, for me, is that I’m much more relaxed than when doing the mundane tasks of driving (keeping a safe distance from the car in front, staying in lane, observing other drivers), and I am convinced that I’m a safer driver. Is that worth $5000 ( or whatever it costs)? Yes. Without any hesitation, yes.

My “highway AP triage” is:
NJ to Saratoga Springs NY: 95% autopilot
NJ to New Haven CT via 95: 90% autopilot, manual in construction areas
NJ to New Haven CT via Merritt, 80% autopilot, mostly just adjusting speed, but 100% AP when in bumper to bumper traffic

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Sat May 26, 2018 3:21 pm

I actually have to tug the wheel a bit now and then, even if not necessary to maintain lane control, to keep from receiving the warning to keep my hands on the wheel. This seems to be more necessary in my Tesla with AP2 than the one with AP1. Maybe that's a testimony that AP2 actually is a little better than AP1 now.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

seawolf21
Posts: 217
Joined: Tue Aug 05, 2014 7:33 am

Re: Tesla Autopilot, worth it?

Post by seawolf21 » Sat May 26, 2018 4:29 pm

Kevin M wrote:
Sat May 26, 2018 3:21 pm
I actually have to tug the wheel a bit now and then, even if not necessary to maintain lane control, to keep from receiving the warning to keep my hands on the wheel. This seems to be more necessary in my Tesla with AP2 than the one with AP1. Maybe that's a testimony that AP2 actually is a little better than AP1 now.

Kevin
Better how?

crystalbank
Posts: 52
Joined: Wed Nov 01, 2017 12:21 am

Re: Tesla Autopilot, worth it?

Post by crystalbank » Sat May 26, 2018 5:08 pm

I would skip it. Not just the 'safety issues', but I wouldn't wanna deal with the perpetual software updates and it's new quirks after every update. I can barely keep up with my computer and phone updates. I would keep it as close to basic no frills spec as possible and get a AAA membership just in case.

User avatar
Kevin M
Posts: 9664
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tesla Autopilot, worth it?

Post by Kevin M » Sat May 26, 2018 5:39 pm

seawolf21 wrote:
Sat May 26, 2018 4:29 pm
Kevin M wrote:
Sat May 26, 2018 3:21 pm
I actually have to tug the wheel a bit now and then, even if not necessary to maintain lane control, to keep from receiving the warning to keep my hands on the wheel. This seems to be more necessary in my Tesla with AP2 than the one with AP1. Maybe that's a testimony that AP2 actually is a little better than AP1 now.

Kevin
Better how?
What I'm saying is that perhaps I do minor corrections more with AP1 than with AP2, so naturally exert enough torque on steering wheel to not get the hands on warning. You can steer the car manually a little bit without terminating AP. So perhaps I'm doing this less with AP2 because I don't feel the need to do so as often, so not naturally exerting enough torque on the steering wheel to avoid the hands on warning.

Or, it could just be that the AP2 programming requires a little more torque a little more often.

I've just gotten in the habit of wiggling the steering wheel slightly every now and then, and seem to do that more in the one with AP2.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

hicabob
Posts: 2659
Joined: Fri May 27, 2011 5:35 pm
Location: cruz

Re: Tesla Autopilot, worth it?

Post by hicabob » Sat May 26, 2018 5:41 pm

crystalbank wrote:
Sat May 26, 2018 5:08 pm
I would skip it. Not just the 'safety issues', but I wouldn't wanna deal with the perpetual software updates and it's new quirks after every update. I can barely keep up with my computer and phone updates. I would keep it as close to basic no frills spec as possible and get a AAA membership just in case.
I wonder when AAA will start having an electric car charger in their trucks as they do now with spare gasoline?

User avatar
TomatoTomahto
Posts: 7307
Joined: Mon Apr 11, 2011 1:48 pm

Re: Tesla Autopilot, worth it?

Post by TomatoTomahto » Sat May 26, 2018 7:06 pm

crystalbank wrote:
Sat May 26, 2018 5:08 pm
I would skip it. Not just the 'safety issues', but I wouldn't wanna deal with the perpetual software updates and it's new quirks after every update. I can barely keep up with my computer and phone updates. I would keep it as close to basic no frills spec as possible and get a AAA membership just in case.
Which "new quirks after every update," specifically, are you referring to? I have had my software updated probably 100 times since I've owned my Tesla. I don't recall any of the "perpetual software updates" being a step backward; the majority of the updates are NBD (no big deal) and occasionally there's one that's really useful.

It's not like your computer and phone upgrades, which can be tedious. I set my updates to take place at 2AM and wake up to new software.

So, what's your experience that warrants referring to "new quirks after every update?" Quirks. Every.

ETA: it just occurred to me that you might not even own a Tesla, and are just supposing that the upgrade model of your PC/phone is how onerous the upgrades are for a Tesla. Perhaps you own one of the millions of Chryslers that have to be recalled for a software upgrade. Now THAT'S tedious. Being one of 4.8 million whose cruise control might not disengage when requested to would be closer to tedious than having my car improved overnight, while I sleep, without having to take it in to the shop. So, do you have personal experience of a quirky Tesla upgrade?

crystalbank
Posts: 52
Joined: Wed Nov 01, 2017 12:21 am

Re: Tesla Autopilot, worth it?

Post by crystalbank » Sat May 26, 2018 7:56 pm

TomatoTomahto wrote:
Sat May 26, 2018 7:06 pm
crystalbank wrote:
Sat May 26, 2018 5:08 pm
I would skip it. Not just the 'safety issues', but I wouldn't wanna deal with the perpetual software updates and it's new quirks after every update. I can barely keep up with my computer and phone updates. I would keep it as close to basic no frills spec as possible and get a AAA membership just in case.
Which "new quirks after every update," specifically, are you referring to? I have had my software updated probably 100 times since I've owned my Tesla. I don't recall any of the "perpetual software updates" being a step backward; the majority of the updates are NBD (no big deal) and occasionally there's one that's really useful.

It's not like your computer and phone upgrades, which can be tedious. I set my updates to take place at 2AM and wake up to new software.

So, what's your experience that warrants referring to "new quirks after every update?" Quirks. Every.

ETA: it just occurred to me that you might not even own a Tesla, and are just supposing that the upgrade model of your PC/phone is how onerous the upgrades are for a Tesla. Perhaps you own one of the millions of Chryslers that have to be recalled for a software upgrade. Now THAT'S tedious. Being one of 4.8 million whose cruise control might not disengage when requested to would be closer to tedious than having my car improved overnight, while I sleep, without having to take it in to the shop. So, do you have personal experience of a quirky Tesla upgrade?
It's great that you're very happy with your car and my previous post clearly rustled some jimmys. :D

I'm not necessarily criticizing that Tesla's can be autoupdated so easily. But my criticism lies within the fact there from my research Tesla seems to have the attitude of a software company i.e. ship it now and fix it later in updates. Traditional car manufacturers have extensive QA processes and are very conservative about shipping something incomplete. They are not always immune recalls and almost every manufacturer issued recalls at some point. My 'quirks' comment is related to AutoPilot. Do your own research in Tesla forums and there are numerous posts about changes in AutoPilot after software updates and not all users are happy with the changes over time.

Another example - Correct me if I'm wrong but some Teslas (may be all?) don't have rain sensors but instead rely on a software driven machine learning algorithm to detect rain when driving (via cameras) and Tesla issues software updates to fix/improve reliability. Some people are pretty ok with it, but I would rather just have a rain sensor like every other car on the street and call it day.

Post Reply