Model 3 EAP Experience

[Mika]

So I've had the Model 3 for a few days now, and have taken a couple opportunities to try out EAP on the freeway. I've never driven or even ridden in an S/X with AP1 or 2, so trying out the 3 has been my only AP experience so far.

My first experience was very positive, a lengthy drive home from Fremont where (after a brief calibration) I was able to use AP for most of the trip, probably 25 miles or more. A lot of it was slow bay area commute traffic, and almost all of it was completely straight, so while this wasn't a big challenge, it handled it with aplomb. I was impressed at this point.

In the following days, I've attempted to use it on some more challenging freeway, specifically 101 as it travels through central Marin. The stretch of freeway I've tested on has a lot of moderate curves and hills. So far, I've had autosteer disengage on virtually every curve when traveling 60mph or above - at some point during the curve it will start to hug one side or the other of the lane, possibly do a couple abrupt jerks to try to center itself, and then eventually give up and have me take over. This is in broad daylight, if that effects performance. After a few attempts where the car can't stay in its lane for more than 20 seconds, I've now given up on it until further updates.

Is this pretty typical of AP2 cars? Useful only in slow and/or straight stretches of road?

 

[RVD98072]

this is pretty much typical of AP2....

 

[voip-ninja]

Doesn't sound like it's worth $5,000 (to me anyway) unless someone sits in traffic all the time during their commute.

 

[hckyplyr]

They will lower the price once 80% of the model 3's dont buy it. A person buying a $35000 ($25000) after rebate will not pay 5k for EAP.

 

[run-the-joules]

Doesn't sound like it's worth $5,000 (to me anyway) unless someone sits in traffic all the time during their commute.

That's me, and that's why I'm buying a Tesla.

 

[omgwtfbyobbq]

Doesn't sound like it's worth $5,000 (to me anyway) unless someone sits in traffic all the time during their commute.

It should work fine on relatively straight freeways/highways/roads even if traffic's moving. But I agree that if the majority of someone's driving is in a city and/or twisty roads/highways, they couldn't get much use out of EAP.

 

[FURY]

That is one of the reason thatthe AP1 Model S is a preferred choice for many(free superchargins a big one also).

The corollary is whether Tesla can make AP2 even equal to AP1. A year ago I thought that would be likely...until I saw those two new hires fro APPLE, who were high level autopilot persons. That was in Jan 2017. FF, but not so FF if the end user wants at least AP1 functionality, to this January and I am really concerned that the AP 3, which I predicted, Jan '17 was the next and likely enhancement over AP1.

Since TES:LA is pushing out Model 3's with such a persistent "BETA" piece, is very, very disappointing to me. Sadly, I suspect that AP3 will happen later (near end 2018, at earliest) Whenever that happens, it is a possibility that the cars before that will be as is, being "BETA' n'all...

Thank you very much

FURY

 

[RVD98072]

If it's really twisty it doesn't work very well (it kind of does but it's scary). If you drive kind of slowly (like <40 mph) it's not too bad on most curves but I'm still hesitant.

On straight roads it works well and in traffic it's awesome. Just using it in traffic is worth the price of admission to me.

 

[voip-ninja]

It should work fine on relatively straight freeways/highways/roads even if traffic's moving. But I agree that if the majority of someone's driving is in a city and/or twisty roads/highways, they couldn't get much use out of EAP.

For me the summon or self-parking features would be something nice to have, but not for $5,000. I would not be able to get the value out of EAP for the typical driving that I do based on the descriptions of what it is capable of.

It would be nice if Tesla let you demo the feature for 30 days and then make the decision on if you wanted to purchase it.

 

[d21mike]

traveling 60mph or above - at some point during the curve it will start to hug one side or the other of the lane, possibly do a couple abrupt jerks to try to center itself, and then eventually give up and have me take over. This is in broad daylight, if that effects performance.

What is the speed limit on that road? I am not familiar with that area but if the speed limit drops below 60 for safety reasons (like sometimes when changing freeways it drops to 45 from 65). Sometimes I will try to see how far I can go on AP when on a trip where you switch between freeways and you can auto lane change to get into the correct lane but before hitting the curve I will lower the TACC to match the speed limit in the curve before getting there.

 

[Hertog_Martin]

Part of the reason I want to buy a Tesla despite it stretching my budget quite a bit is the Autopilot. But whereas at first I thought Tesla was miles ahead of the competition, it now seems that it's not and has a hefty price tag as well.

As I'm in Europe I won't get it before years end, any chance that there will be both a hardware and a software update by then?

I mean, Elon has bigger plans surely for Autopilot than what is depicted in the OP

 

[insaneoctane]

Has Tesla done the "free AP weekend" type promotions before? If the technology is worth it, it will prove itself in either a free weekend or 30 days after taking it home....Seems like marketing 101 to me.

 

[WileyTheMan]

Has Tesla done the "free AP weekend" type promotions before? If the technology is worth it, it will prove itself in either a free weekend or 30 days after taking it home....Seems like marketing 101 to me.

They certainly could do it. But you would have to actually have a marketing department to make Marketing 101 happen.

 

[Az_Rael]

It would be nice if Tesla let you demo the feature for 30 days and then make the decision on if you wanted to purchase it

Has Tesla done the "free AP weekend" type promotions before? If the technology is worth it, it will prove itself in either a free weekend or 30 days after taking it home....Seems like marketing 101 to me.

Yes, they have done this in the past. Gave everyone with an AP1 car free AP for 2 weeks.

 

[SMAlset]

For me the summon or self-parking features would be something nice to have, but not for $5,000. I would not be able to get the value out of EAP for the typical driving that I do based on the descriptions of what it is capable of.

It would be nice if Tesla let you demo the feature for 30 days and then make the decision on if you wanted to purchase it.

My husband has only tried auto steer feature once I think on a long distance drive, not sure what version of AP 2 we were on but it was still kind of rough at that point on lane changes. Now he does really like the EAP's car-length following setting and uses it all the time and the speed setting feature. I mostly do in town driving so don't use it but while on a long distance trip did. Kind of took me a little bit to get use to it (maybe trust it would be a better term) but I did find it useful and it was reliable. Right now neither of us would be comfortable with any hands off kind of driving so we only got the EAP on our MS. We will add it to our Model 3 when we order.

 

[Hertog_Martin]

Is Tesla collecting data btw when you have AP on?

If they use the data to train the AI then it can enhance rapidly when a quarter million Teslas are driving around.

 

[Mika]

What is the speed limit on that road? I am not familiar with that area but if the speed limit drops below 60 for safety reasons (like sometimes when changing freeways it drops to 45 from 65). Sometimes I will try to see how far I can go on AP when on a trip where you switch between freeways and you can auto lane change to get into the correct lane but before hitting the curve I will lower the TACC to match the speed limit in the curve before getting there.

This was on highway 101, posted limit would be 65 I believe. I would normally drive this stretch at 70-80, traffic permitting. It's not like these are sharp curves, just large, gradual bends, and it can't handle them at anything close to the speed that traffic is flowing. It's definitely a little disappointing.

Anyway, sounds like my experience isn't an anomaly, I was kinda hoping someone would say my calibration is off and I should be seeing better results. Though auto steer might be pretty limited right now, at least TACC seems very smooth and useful.

 

[dmode]

I have used AP2 for over a year now (since Dec 2016). I will not buy another car without similar features. I have a long commute to work, in a lot of stop and go traffic. AP2 is really good in those situations. I also do a decent number of road trips (20,000 miles in 1 year), and AP is really useful when you are driving on freeways outside of urban areas (e.g., I-5 in CA). So yeah, if you have a long commute / commute in heavy traffic and if you do frequent roadtrips, AP is highly recommended. However, if you mostly use surface streets and twisty highways, not sure if AP will be useful

 

[tpatana]

That's me, and that's why I'm buying a Tesla.

Me too.

 

[SMAlset]

Is Tesla collecting data btw when you have AP on?

If they use the data to train the AI then it can enhance rapidly when a quarter million Teslas are driving around.

Some owners are participating in the sharing of their road data from the cameras. If you wanted to do so, you acknowledged giving your permission. Not sure when this started or how large of a group the request went out to.