Poll: Is FSD Worth the Cost?

[Garlan Garner]

Seriously? Not only are you misquoting me, but then you're using that to rant about level 5 autonomy, which I said nothing about.

Back on topic, Tesla has stated for several years that FSD is designed to conduct short and long distance trips with no action required by the person in the driver's seat. That's not level 5, just really good level 2, maybe level 3, with the driver having to take control at times.



What changed over the past year or two is the statement about the system's safety being at least twice as good as the average human driver. That was in older versions of the autopilot page, but is not in recent versions.

https://web.archive.org/web/20170209190258/https://www.tesla.com/autopilot

Indeed I'm serious and I'm replying to you....not quoting you.

Tesla is currently designing their beta FSD product to conduct trips with no action required by the person in the drivers seat. That is indeed Level 5. I don't care if the trip is short or long....its level 5 when no driver interaction is required.

Now does that mean that they are currently selling level 5 FSD? No. Are people not reading what they are buying - expecting level 5 autonomy? Yes. Absolutely they are.

Tesla is fully describing what the current version of beta FSD does right now - and will be improved upon in future updates - as it states clearly on their website.

 

[PRLM3]

But I find NOA to be an extremely frustrating experience because of how it manages the lane I'm in. It wants to change lanes whenever the car in front of me is going 2 MPH slower, at which point it cuts off the guy next to me, and stays parked in front of him for the next couple miles until it decides to move over. Taking exit ramps automatically is nice, but overall I don't find NOA to be helpful, which adds no value to me.

Funny how everyone's experience is so different but I guess not surprising. FWIW, I find full EAP with NoA to be absolutely amazing. I prefer to leave auto lane changes off and am set to mad max. I cap the max speed at +15 over limit (may cheat on that but bumping right scroll wheel - killer feature). I can do a 4 hour trip up or over FL and almost never, I mean never, have to override the software. Since v10 was released, it reads the road and performs incredibly well. I particularly like the fact that it recognizes the left lane as the passing lane and will move out after a while all on it's own.

Another nice upgrade feature was the smarter lane changing -- it still will prompt me to change lanes at odd times not infrequently, but as of 10.0, even when it does that -- it knows it's stupid, marks the lane dividers and adjacent car red, and waits for the vehicle to move out of the way, then executes the lane change. I was very anxious the first time I tried this, but have seen it execute this flawlessly multiple times and am totally comfortable with it. For instance, Max wants to pull into the left lane but there is a car coming up on your left, and the sensor sees it. It has already signaled to you it wants to change lanes -- even if I signal go ahead and change, because the lines and oncoming car are red, it waits for the car to pass me, then moves into the lane. Y'all, that's pretty dope.

The other killer application for EAP with or without NOA is the bumper to bumper, full stop, crawl, maybe get to 20, crawl stop. Repeat. Save your blood pressure: forget about changing lanes, set EAP, and chill. Text, check an email -- yes, stay in the game and pay attention -- but the experience overall is a game changer.

Auto Summon is still largely a novelty -- but I have to park in a garage at work not infrequently with insanely narrow spots and worse -- massive polls inches from your bumper. So I use "regular summon" all the time here to pull my car out so I can actually get in without contorting my body or shoving it through a space smaller than it wants to go through! In that setting which is not uncommon for me - it's hugely practical, and yes, it's usually hysterical to watch someone freak out!

Obviously the trade off on price for feature is a hugely personal thing so there's no right or wrong here per se. I paid 5k for EAP and have held off on the last 2.5 for FSD as I think for the time being I have the sweet spot on value. I would definitely do the 5k again for EAP though - no question, as every time I get in and particularly when I hit the highway I get real value. Will I pull the trigger on the 2.5 k for FSD at some point? For stop signs and traffic lights (reliably) -- probably. I already use it for the work commute (15 minutes / no highway) minimally. (Truth be told, tying the tie, checking a text, in 35 mph traffic with a car in front of you -- it's way safer than humans imho after 14 months.) We'll see what Elon pulls off in 2020, perhaps it's v11.0?

 

[daver3]

Good afternoon everyone,

After 10 months with EAP, I have mixed feelings on the whole thing. I think autopilot is fantastic, and use it every day.

But I find NOA to be an extremely frustrating experience because of how it manages the lane I'm in. It wants to change lanes whenever the car in front of me is going 2 MPH slower, at which point it cuts off the guy next to me, and stays parked in front of him for the next couple miles until it decides to move over. Taking exit ramps automatically is nice, but overall I don't find NOA to be helpful, which adds no value to me.

I was really excited for Enhanced Summon to be released. But after awkwardly blocking parking lots a couple times, I realized in current form it's not useful for anything other than a mostly empty parking lot. I would much rather it be cautious than risk hitting things, but if it has to be THIS cautious to be safe, I don't find it useful.

At this point, FSD isn't yet available, but after seeing how NOA and enhanced summon have been implemented, I have a tough time being optimistic. If are making another purchase today, I would opt for the standard autopilot and nothing more. I'm curious to know what you guys think? Is it worth forking over the money with the expectation that these features and FSD will evolve and be great? Or do you think it's unlikely they will add $7000 in value to your driving experience?

I'm pretty much with you, I like AP and use it all the time, I don't like NOA for the same reasons you state and Summon is mostly a parlor game. I used it at the club with all the golf group watching and it backed out beautifully and started forward, then a golf cart navigated around it to the right and both golfers did a huge double take because there was nobody in the car. It drove up to the steps where I was waiting and I went out to get in, just as my friend in the golf cart came around to comment. He moved on to tell anyone he could talk to about the incident. However, I've tried it in multiple other parking lots and it does not do well. . .but it does get attention.

 

[Schly]

I'm pretty much with you, I like AP and use it all the time, I don't like NOA for the same reasons you state and Summon is mostly a parlor game. I used it at the club with all the golf group watching and it backed out beautifully and started forward, then a golf cart navigated around it to the right and both golfers did a huge double take because there was nobody in the car. It drove up to the steps where I was waiting and I went out to get in, just as my friend in the golf cart came around to comment. He moved on to tell anyone he could talk to about the incident. However, I've tried it in multiple other parking lots and it does not do well. . .but it does get attention.

Good afternoon everyone,

After 10 months with EAP, I have mixed feelings on the whole thing. I think autopilot is fantastic, and use it every day.

But I find NOA to be an extremely frustrating experience because of how it manages the lane I'm in. It wants to change lanes whenever the car in front of me is going 2 MPH slower, at which point it cuts off the guy next to me, and stays parked in front of him for the next couple miles until it decides to move over. Taking exit ramps automatically is nice, but overall I don't find NOA to be helpful, which adds no value to me.

I was really excited for Enhanced Summon to be released. But after awkwardly blocking parking lots a couple times, I realized in current form it's not useful for anything other than a mostly empty parking lot. I would much rather it be cautious than risk hitting things, but if it has to be THIS cautious to be safe, I don't find it useful.

At this point, FSD isn't yet available, but after seeing how NOA and enhanced summon have been implemented, I have a tough time being optimistic. If are making another purchase today, I would opt for the standard autopilot and nothing more. I'm curious to know what you guys think? Is it worth forking over the money with the expectation that these features and FSD will evolve and be great? Or do you think it's unlikely they will add $7000 in value to your driving experience?

FSD is not about what we see now, which is pretty shaky in my opinion.

Based on my current experience with EAP, I would guess that FSD is over five years out.

But here’s the thing. Tesla is nearly done changing the entire code base for FSD and EAP. We have seen none of it. So, to compare what we’ve been using to what is coming SOON, is not an appropriate comparison.

Tesla will push out a version that’s already 95% complete that we have experienced NONE of, so thinking that they’re going to just improve the current version of the software incremental until it’s FSD is false.


He magic that is coming is something we have barely had a taste of yet.

Seeking Alpha can be a bit of a hype machine, but this is an extremely well written article explaining what’s in store for us and why we don’t really see it coming.

https://seekingalpha.com/article/43...&utm_campaign=nl-must-read&utm_content=link-0

 

[MostlyHappy]

Good afternoon everyone,

After 10 months with EAP, I have mixed feelings on the whole thing. I think autopilot is fantastic, and use it every day.

But I find NOA to be an extremely frustrating experience because of how it manages the lane I'm in. It wants to change lanes whenever the car in front of me is going 2 MPH slower, at which point it cuts off the guy next to me, and stays parked in front of him for the next couple miles until it decides to move over. Taking exit ramps automatically is nice, but overall I don't find NOA to be helpful, which adds no value to me.

I was really excited for Enhanced Summon to be released. But after awkwardly blocking parking lots a couple times, I realized in current form it's not useful for anything other than a mostly empty parking lot. I would much rather it be cautious than risk hitting things, but if it has to be THIS cautious to be safe, I don't find it useful.

At this point, FSD isn't yet available, but after seeing how NOA and enhanced summon have been implemented, I have a tough time being optimistic. If are making another purchase today, I would opt for the standard autopilot and nothing more. I'm curious to know what you guys think? Is it worth forking over the money with the expectation that these features and FSD will evolve and be great? Or do you think it's unlikely they will add $7000 in value to your driving experience?

Thanks for the post and thread. It is a question I have often asked myself - I have standard autopilot which I use a lot on my 75 mile commute of which 55 miles is on 3 lane motorway. Is it really worth paying an extra 15% for FSD? I find autopilot mostly great, but like others hate that you have to disengage it to overtake. Makes me lazy and want to just stay in the middle lane - something I hate in other drivers so really should not do myself. But I do find it occasionally unreliable, particularly in merging traffic, lots of spray (the UK has had the wettest winter on record) and for some reason it occasionally gets into a panic and slams on the brakes. I also find it annoying that you have to fight it the whole time. Yeah I get that you are supposed to supervise, but keeping constant turning pressure on the wheel is a major detriment to the driving experience. I also find that the speed limits picked up from Google maps are not always reliable. There are also a few funnies. In the UK we have pedestrian crossings with zig-zag lines on the approaches. It amused me feeling the self steer try to follow them. I drive with a pretty light touch and when overtaking ease gently back into the lane of the vehicle just overtaken. Sometimes I get the agitated command to take control immediately because AP thinks I am taking too long. I can see the benefit of being alerted when seemingly drifting out of lane - it is often a symptom of a driver falling asleep, but I wish that it would learn from my driving style.

I am wondering whether if there was an option for EAP without FSD, how many people would take it. If the price was not too steep, I probably would based on what others have said, but it would be nice to get a months free trial first.

 

[P33M]

Believe since last year m3 come with so called basic autopilot, without NOA and Lane change,

 

[trak0r]

I paid (or will be) for FSD on my M3, mainly because psychologically it feels like the self driving stuff, in all its forms, is the car’s party piece. It’s a Tesla, everyone expects it to be driving itself.

I’ve rationalised it to myself that it’s like paying for the Touch Bar on the new MacBook Pros. It feels like it’s USP. It’s gimmicky, it has limited use (I barely ever use it in anything but it’s most primitive application), but I like that it’s a shiny feature not seen on other laptops, or even older MBPs. Could I rationally explain paying extra for it? No. I feel the same way about FSD.

I’m in the UK so FSD is even more crippled over here thanks to the EC directives than it is over there, and we’re not seeing the latest stuff like traffic lights and being a foreign market probably won’t for some time.

I still paid (£5800) for it though. Just don’t ask me to explain why.

I think most who buy it probably did it for similar reasons. Sucks that you cannot use even the basic stuff in UK. I specifically bought the car without any advanced features that didn't do what their name suggests. I bought because I wanted the raw sheer performance of power train since I track the car at various events. The car is setup very well and helps me place 1-2 in events. Much better the ICE hardware.

We all have our reasons to buy the car, but for the money and what you get, it is not worth springing for AP or FSD IMHO today. I will spend the $10,700 for FSD AFTER Tesla proves it can reliably/legally drive me from New York to San Francisco while I'm sleeping Or watching movies on that big screen ;D

 

[mazers]

Thanks for the post and thread. It is a question I have often asked myself - I have standard autopilot which I use a lot on my 75 mile commute of which 55 miles is on 3 lane motorway. Is it really worth paying an extra 15% for FSD? I find autopilot mostly great, but like others hate that you have to disengage it to overtake. Makes me lazy and want to just stay in the middle lane - something I hate in other drivers so really should not do myself. But I do find it occasionally unreliable, particularly in merging traffic, lots of spray (the UK has had the wettest winter on record) and for some reason it occasionally gets into a panic and slams on the brakes. I also find it annoying that you have to fight it the whole time. Yeah I get that you are supposed to supervise, but keeping constant turning pressure on the wheel is a major detriment to the driving experience. I also find that the speed limits picked up from Google maps are not always reliable. There are also a few funnies. In the UK we have pedestrian crossings with zig-zag lines on the approaches. It amused me feeling the self steer try to follow them. I drive with a pretty light touch and when overtaking ease gently back into the lane of the vehicle just overtaken. Sometimes I get the agitated command to take control immediately because AP thinks I am taking too long. I can see the benefit of being alerted when seemingly drifting out of lane - it is often a symptom of a driver falling asleep, but I wish that it would learn from my driving style.

I am wondering whether if there was an option for EAP without FSD, how many people would take it. If the price was not too steep, I probably would based on what others have said, but it would be nice to get a months free trial first.

EAP without FSD was $5000 when it was available for configuration. Would that cost be acceptable?

 

[Bobtbadger]

At the risk of being attacked I've decided that I should express some of my concerns about my experiences with FSD on my six-month-old P3D. I'm in the UK so current European legislation is in effect which is, no doubt, dumbing down the car's capabilities but I have real concerns about Tesla's current implementation of the hardware and here's why: we've had some pretty extreme weather in the UK over the last couple weeks including, unusually, some snow. I've noticed that the cameras, radar and ultrasonic sensors seem to be particularly affected by this type of weather. Last week I was driving on a 200 mile trip on the motorway with auto steer engaged (I don't really like NOA as it consistently does unexpected things). Within two minutes of the snow starting I started to get warnings saying that various sensors were blinded/unavailable and auto steer disengaged. The snow stopped about five minutes later but despite the weather turning to rain, the front end sensors (radar etc) didn't come back online at all. I stopped the car and cleared the frozen snow from the front. I started on my journey again and all the self driving functions were available. It then started to rain quite heavily and, again within minutes, self driving disengaged. I've also consistently had messages about cameras becoming blinded in the sun. Normally this is the two pillar cameras but the other day I got the same message from the fender camera. So here's my question: how will the car deal with this when FSD is finally enabled? I watch the video of the self driving day on YouTube in its entirety. I found it to be really excellent but, as far as I can remember, nobody talked about the issues I'm raising above. My car has problems on wet days, on snowy days and on sunny days. Am I missing something? Or is there a need for other types of sensors? Elon was pretty robust on this topic stating that once you've cracked vision you've cracked the problem but if my glasses steam up on a cold morning I can't see where I'm going. I get that there are a number of redundancies built in but my car quits FSD well before vision is lost. So, like a number of people around the boards, I think my car will need additional sensors before it can genuinely full self drive...... discuss

 

[Garlan Garner]

At the risk of being attacked I've decided that I should express some of my concerns about my experiences with FSD on my six-month-old P3D. I'm in the UK so current European legislation is in effect which is, no doubt, dumbing down the car's capabilities but I have real concerns about Tesla's current implementation of the hardware and here's why: we've had some pretty extreme weather in the UK over the last couple weeks including, unusually, some snow. I've noticed that the cameras, radar and ultrasonic sensors seem to be particularly affected by this type of weather. Last week I was driving on a 200 mile trip on the motorway with auto steer engaged (I don't really like NOA as it consistently does unexpected things). Within two minutes of the snow starting I started to get warnings saying that various sensors were blinded/unavailable and auto steer disengaged. The snow stopped about five minutes later but despite the weather turning to rain, the front end sensors (radar etc) didn't come back online at all. I stopped the car and cleared the frozen snow from the front. I started on my journey again and all the self driving functions were available. It then started to rain quite heavily and, again within minutes, self driving disengaged. I've also consistently had messages about cameras becoming blinded in the sun. Normally this is the two pillar cameras but the other day I got the same message from the fender camera. So here's my question: how will the car deal with this when FSD is finally enabled? I watch the video of the self driving day on YouTube in its entirety. I found it to be really excellent but, as far as I can remember, nobody talked about the issues I'm raising above. My car has problems on wet days, on snowy days and on sunny days. Am I missing something? Or is there a need for other types of sensors? Elon was pretty robust on this topic stating that once you've cracked vision you've cracked the problem but if my glasses steam up on a cold morning I can't see where I'm going. I get that there are a number of redundancies built in but my car quits FSD well before vision is lost. So, like a number of people around the boards, I think my car will need additional sensors before it can genuinely full self drive...... discuss

I wonder how FSD and NoA will be once it exits beta.

don't you?

 

[Garlan Garner]

Looks like some of the competition is giving up the fight to compete with Tesla's FSD and Autopilot. That's not good.

Tesla's Hardware 3 computer frightens legacy auto after Model 3 teardown: 'We cannot do it'


I wonder why the competitions' tone isn't like the tone of this thread? hmmmmm

 

[holmgang]

Looks like some of the competition is giving up the fight to compete with Tesla's FSD and Autopilot. That's not good.

Tesla's Hardware 3 computer frightens legacy auto after Model 3 teardown: 'We cannot do it'


I wonder why the competitions' tone isn't like the tone of this thread? hmmmmm

It takes a bit of liberty to combine the quote of "one stunned engineer from a major Japanese automaker" and "Industry insiders expect such technology to take hold around 2025 at the earliest" to project it as a the consensus opinion of competitors. That is a very wide brush.

That article also points out to another long-standing fact: manufacturers don't need to build autonomous driving capabilities. Just they like everything else in their cars from time immemorial, they take it from suppliers and integrate it. Could be from intel/mobileye, bosch, or whoever else. Let specialists do the detailed work and tie it to the vehicle.

 

[holmgang]

Seeking Alpha can be a bit of a hype machine, but this is an extremely well written article explaining what’s in store for us and why we don’t really see it coming.

https://seekingalpha.com/article/43...&utm_campaign=nl-must-read&utm_content=link-0

its from a random blogger , whose credibility i cant see proved more valuable than a random anonymous schmo (like you or me)

https://seekingalpha.com/author/trent-eady#regular_articles

https://seekingalpha.com/instablog/48129566-trent-eady/5022226-all-tesla-predictions

 

[doghousePVD]

Also, when a Freeway forks, or when there's a badly marked exit, with NoA the car knows which lane to take to keep on its route.

Sometimes/rarely. In New England there are quite few exits that NoA screws up on. Miscounting the number of lanes, left exits, and exit-only lanes are fairly common errors.

 

[Knightshade]

Stuck? You don't have to turn anything on.

I have FSD.

When I turn off NOA....I essentially have the current version of AP.

No, you don't. You have EAP.

Which does a lot more than the current AP does.

For example you can still use your turn signal to have the car change lanes for you- current version AP owners can't.

Come on over. Experience it for yourself.

or go to the track where Drive Times got his 2.8. He wouldn't mind showing you his 2.8 run. He was the one who inspired me to go out and try it for myself.

https://electrek.co/2019/11/14/tesl...cceleration-dips-below-3-sec-software-update/

Uh... the guy in the video there got 2.998.

Not 2.8.

Why do you keep citing sources that prove you wrong dude?


Nor did you really get 2.7- you raced someplace with broken equipment.

NOBODY has gotten any times CLOSE to the ones you claim you did.


Unless you actually ran 2.997 and like with the drag times video are dropping the 2 middle digits for some weird reason.

EAP and FSD NEVER said that if you purchased the package that IT WOULD drive somewhere with no input from the driver. NEVER.

That's flat out false.

Tesla said exactly that for over 2 years selling FSD prior to March 2019.

 

[TLLMRRJ]

It takes a bit of liberty to combine the quote of "one stunned engineer from a major Japanese automaker" and "Industry insiders expect such technology to take hold around 2025 at the earliest" to project it as a the consensus opinion of competitors. That is a very wide brush.

That article also points out to another long-standing fact: manufacturers don't need to build autonomous driving capabilities. Just they like everything else in their cars from time immemorial, they take it from suppliers and integrate it. Could be from intel/mobileye, bosch, or whoever else. Let specialists do the detailed work and tie it to the vehicle.

Exactly. Don't forget about that $1000 app that runs on your cell phone, plugs into a Toyota, and drives just about as well as AP with one camera and nothing else.

No shortage of marketing bs in this area of technology.

 

[doghousePVD]

I heard from a birdee that FSD does something with lane changing.

Can someone go out and try to have EAP lane change and let us know if that works?

EAP definitely does lane changes, as well as parking and summon. I am not aware of any functions in FSD that isn’t included with EAP except for some visualization.

AP is not the same, so FSD adds some functionality.

 

[Garlan Garner]

EAP definitely does lane changes, as well as parking and summon. I am not aware of any functions in FSD that isn’t included with EAP except for some visualization.

AP is not the same, so FSD adds some functionality.

AP tries to center itself when an exit comes up that isn't clearly marked - and then quickly moves to re-center itself when the exit or split goes away.

When you put a destination in FSD / NoA the car won't center itself at all if an exit or split comes up. It will follow the route to get to the destination.
As a matter of fact FSD/NoA will turn on the turn signal about 100 feet before a split in the expressway occurs to let the driver behind know - "Hey- I'm going this way". and it won't even attempt to center itself at the split.
The turn signal will stay on for about 4 clicks and turn off.

That is an uncommunicated feature of FSD that AP does not have.

 

[Zeronet]

I bought FSD, but i wish i didn't. I only use TACC (a lot) and don't won't it to steer itself because it's constantly nagging me about keeping hands on the wheel, than i rather steer myself. I'm right handed and keep only my right hand on the wheel, but it doesn't seem to detect that properly. Not going to use any hacks and try to fool it, but because i'm right handed i don't want to use left hand. Where are the sensors? How do they work?

I had trouble with this for six months when I was convinced that I needed the squeeze the wheel harder to get the nagging to stop. My grip muscles got a lot of exercise. But I finally realized that all it wants is a little turning jerk. Not enough to drop out of assist, but just a little jerk. It can be done with either hand.

 

[Mobster]

Lots of good opinions here and shared experiences. I respect everyone's point of view.

I bought my first Model S in 2012 (early adopter?)
I bought my second Model S in 2016 and made sure FSD was included. The hype in the showroom was fantastic backed up by Elon's tweets. Good car. However....

I have grown tired of looking forward to something that now appears to never happen. Not with the "F" in FSD. I grow tired of asking for my AP 2.0 to be upgraded to HW3 and am told "No. We have a lot to do and you are probably on a list" Sigh.

I would at least like to use the 8 cameras for other things. Autopilot right now is very unpredictable. I do really like most of what it does other than when it tries to kill me.

No, I would not recommend spending the money for FSD. Years go by. Children grow and move out. I retired from work. And.....FSD continues to be a "bet" creating lots of stir on these forums. (look at it as a "startup"? investment) No. No.