-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Autopilot Still in Beta? Edmund's Review
- Thread starter N5329K
- Start date
I see thank you!
It’s ashamed that it doesn’t do the same at 90mph. I can see why this is a hand on when using autopilot. At the same it sound very nerve racking at traffic lights.
I consider most automotive to be in Beta at this point at least Tesla is up front about it on the AP and FSD features anyways.
One of the big ABS manufactures for semi trucks found an issue with their ABS system that at low speeds it would think the wheel had locked up and release the brakes this system was used a lot on school buses and garbage trucks. Of course they didn't have OTA software updates and were having to have hands on hardware to fix the issue ( I can't remember if it was replacing the module or reprogramming it). The system had been in the field for some time before they figured out what was causing all of the accidents until then they just blamed the drivers.
Look at how many recalls there are. Really most of this should be standard easy stuff but they keep making mistakes (this applies to all manufacturers). And it isn't like they are making new mistakes it is frequently the same one over and over again. How many ignition switch wiring recalls have to be done before they realize to use larger wire and heavier contacts to start with. How many times we had multiple recalls on the same car for the same issue. I remember a few throttle linkages that would stick they had a recall and a year latter they had to recall again as the new parts had the same issue.
I went ahead and ordered FSD. I am hoping they will start having some features above EAP released in the next year or so. I don't expect FSD for several years if ever without buying a new car. I plan on keeping my car 10+ years if I was planning on having it less than 5 years I probably would skip FSD and just get EAP.
One of the big ABS manufactures for semi trucks found an issue with their ABS system that at low speeds it would think the wheel had locked up and release the brakes this system was used a lot on school buses and garbage trucks. Of course they didn't have OTA software updates and were having to have hands on hardware to fix the issue ( I can't remember if it was replacing the module or reprogramming it). The system had been in the field for some time before they figured out what was causing all of the accidents until then they just blamed the drivers.
Look at how many recalls there are. Really most of this should be standard easy stuff but they keep making mistakes (this applies to all manufacturers). And it isn't like they are making new mistakes it is frequently the same one over and over again. How many ignition switch wiring recalls have to be done before they realize to use larger wire and heavier contacts to start with. How many times we had multiple recalls on the same car for the same issue. I remember a few throttle linkages that would stick they had a recall and a year latter they had to recall again as the new parts had the same issue.
I went ahead and ordered FSD. I am hoping they will start having some features above EAP released in the next year or so. I don't expect FSD for several years if ever without buying a new car. I plan on keeping my car 10+ years if I was planning on having it less than 5 years I probably would skip FSD and just get EAP.
At a wild guess, maybe the 90 mph limit on emergency braking is that above that it's going to fast to detect a crash in time. But in the U.S. there are no public roads where it's legal to go that fast, even allowing for a 5 mph leeway before cops stop you (depending on location). I know I'll never be driving that fast. And if you're on a track you probably do want the car using emergency braking.
What I cannot understand is why it only slows you down by 25 mph, rather than apply the brakes until a collision is no longer imminent, or until a low set speed is reached.
Remember when you sent out 'patches' to correct a bug or flaw? These were free. We called them bugfixes, but sales called them patches for obvious reasons.
An Update added new features, and you normally had be under service contract to get Updates.
Yes, everything in the manual and on the box worked. Sure there were bugs, but the code was tested as well as possible, your power users got an advanced copy of updates.
If you were working on a new feature, and the cutoff date was 11/1, and it wasn't stable enough to be proud of, it simply did not ship until the next time, no matter how close you were. "But I can fix it in a day or two!" fell under the category of "In the life of every project, there comes a point where you must shoot the engineer in the head and ship something."
Electricfan
Active Member
Well, no can't agree with resigning myself to "that's just the future". Your TV won't kill you if it doesn't work properly. I'm hoping some good comes from the recent Model X tragedy, and Tesla is held accountable for the death. I don't care if it means all AP systems are disabled. Your suggestion that we must resign ourselves to literally sacrificing our lives so Tesla can make improvements in its eternally "beta" software does not work for me at all, not even a little bit. You might be right, though. I'm old, and maybe the future is for the young. But by my lights there is still something called "right and wrong", and putting out a system that you advertise will automatically drive the car, without making is safe, is just wrong.
That makes sense, maybe later they’ll have faster reaction time. lol the only road I know you can go 90mph is going to El Paso, TX. The good thing about that road is that it has clear marking and hardly anyone on the road. However, there’s is a part of that route where cellular and radio connection is little to none.
Tesla says that the driver is still fully responsible and that their systems right now are only driver assist. How is that any different than cruise control? If you aren't paying attention cruise control will let your car speed off a cliff. Tesla hasn't put out a system that it says can drive a car yet. How do you weigh it when the system fails and it kills one person but during that same time the system worked on other cars and saved 10 people? Is it still bad because it caused 1 death or is it great because overall it saved 9?
What if there's a system that will kill one person per one billion vehicle miles, compared to human drivers killing 1.25 people per 100 million vehicle miles? Do we prohibit the system because it will sometimes kill people?
A harder question: What if there's a system that, if used properly, reduces fatalities by 20%, but used improperly increases fatalities by 20%? Do we second-guess the users and say, "People are too stupid to use it properly, so we'll ban it."?
In the present case, we don't even know which, if either, of these analogies fits. Drivers get distracted regularly, sometimes by something unexpected, sometimes because they make bad decisions. EAP will prevent some number of accidents when a person is momentarily distracted. But it won't prevent all. And it will kill you if you treat it as though it were FSD. I come down on the side of believing that these sorts of driver assist features will make the highway slightly safer. But we need to beware of the fact that for the time being, "autopilot kills driver" will always make the news, routine accidents (involving conventional cars) do not, because there are too many of them, and "Tesla car saves distracted driver" will never make the news.
WileyTheMan
Peanut Gallery Member
That's not how it's advertised, but consumer perception always seems to think that.
I am buying EAP on my 3 primarily so when my wife drives her 60m commute in heavy traffic, she can use it as TACC. She can relax a bit during the trip but neither take her hands of the wheel nor stop paying attention. It will give her an additional safety measure to enhance her reaction time if something unexpected happens. This is worth $5k to me.
RayK
Active Member
Auto breaking feature is supported in only one very specific instance. You are sitting in the driver's seat with a totally dead Model 3 on your hands (note: the following command does NOT work on S or X to my knowledge). No response from the touchscreen, the phone app or even the card key. For some reason even both of the emergency door latches do not work. So, how do you get out of the car? Simply say "auto break the glass". Voila! You can now climb out any one of the doors (door windows all break at the same time; windshield, roof and trunk remain intact).
Sorry, I just couldn't resist after a day. I let it go this long in order for @SP_H to receive the proper response to the query. I'm usually not the spelling police but it does bug me when it goes unchecked.
Electricfan
Active Member
Sorry, but I could not disagree more strongly. Using the word "autopilot" IS advertising it that way.
Well Autopilot also makes it sound like the car can fly otherwise why do I need a pilot auto or otherwise. If you look at Autopilot on airplanes Tesla is using the term correctly as on airplanes it is constantly monitored.
Braking vs breaking. Brake is what is used to slow down.
Neither will your AP kill you if you use it the way it is intended. If you don’t use common sense, are careless and not pay cursory attention - it is no different from using a chain saw or a dumb cruise.
Last edited:
Cars kill people. I doubt we'll ever be able to get away from that. The goal is to make them safer. Padded dashboards, then seat belts, then airbags and anti-lock brakes. AP is one more step in that direction, but I do think that "autopilot" was a poor choice of names. "Steering assist" might have been a better choice. We're probably going to see a few more deaths before drivers finally get it that this is not FSD and they need to take over when conditions are sub-optimal.
omgwtfbyobbq
Active Member
Criticizing the safety aspects of EAP's beta features is fine, but I think they should explain the differences between the features that are still in beta, Autosteer and auto lane change, and the features that aren't, TACC, Autopark, and Summon. Owners can choose to use all of EAP's features, but they should understand that Autosteer and auto lane change are in beta and require more attention from the driver.
Someone can wait until Autosteer and auto lane change are out of beta, but I wouldnt be surprised if EAP was more expensive at that point. Just like FSD will likely be more expensive if and when it's working.
omgwtfbyobbq
Active Member
To be fair to Tesla, their TACC seems to be better than the TACC most other manufacturers offer and I'm guessing a big chunk of that cost is going towards getting Autosteer and auto lane change out of the beta stage.
Help, what is the trick to setting cruise control?
Drove 65 miles today from Newport Beach to Encinitas today.. 99 percent of the freeway driving on autopilot...
I think the « autopilot » name is more than appropriate....
I think the « autopilot » name is more than appropriate....
Seems like Tesla should make available this warning for people considering purchasing the EAP options so they know if they're willing to use it or not. This is the warning that appears once when you first activate autosteer, and in it, it clarifies why they call it "Beta." This particular screenshot might be older, but something very similar exists for the 3.
Google Navigation (not maps) is still "Beta" AFAIK. It's been in Beta since I first started using Android in 2010. It's hardly Beta in the traditional software sense, but like EAP, Google is calling it Beta so people don't get complacent with it.
Google Navigation (not maps) is still "Beta" AFAIK. It's been in Beta since I first started using Android in 2010. It's hardly Beta in the traditional software sense, but like EAP, Google is calling it Beta so people don't get complacent with it.
Similar threads
