Ordered an X. Declined FSD. Is this a mistake?

[d21mike]

It seems those that are completely against FSD is because they believe there will be NO Incremental improvements over EAP that require FSD. They believe that is is ALL FSD or NO FSD. Am I wrong? I think there will be incremental improvements and expect them to start happening in 2018.

Example: Would STOP SIGN and TRAFFIC LIGHT Detection be part of EAP or FSD? I say FSD.

I use EAP pretty much every time I get in the car. If it could detect and stop at the stop sign or honor traffic lights it wold be a HUGE improvement for me. Worth the $3k I paid at time of purchase.

NAV support on Freeways it supposed to be part of EAP. On Ramp to Off Ramp. But I would assume that would not be the same on surface streets.

I think there is a lot of opportunity for improvements to AP that require FSD and that is the main reason I decided to purchase it. The cost now of $3k vs current $4k after purchase is a minor reason for me. But the $4k is not a guarantee especially if they need to add more hardware.

 

[Peteski]

It seems those that are completely against FSD is because they believe there will be NO Incremental improvements over EAP that require FSD. They believe that is is ALL FSD or NO FSD. Am I wrong? I think there will be incremental improvements and expect them to start happening in 2018.

Example: Would STOP SIGN and TRAFFIC LIGHT Detection be part of EAP or FSD? I say FSD.

I use EAP pretty much every time I get in the car. If it could detect and stop at the stop sign or honor traffic lights it wold be a HUGE improvement for me. Worth the $3k I paid at time of purchase.

NAV support on Freeways it supposed to be part of EAP. On Ramp to Off Ramp. But I would assume that would not be the same on surface streets.

I think there is a lot of opportunity for improvements to AP that require FSD and that is the main reason I decided to purchase it. The cost now of $3k vs current $4k after purchase is a minor reason for me. But the $4k is not a guarantee especially if they need to add more hardware.

Well FSD is just what it states it is i.e. Full Self-Driving. Anything less than that is by definition not FSD. I personally think Tesla will implement pretty much anything that enhances EAP without requiring additional hardware. Or maybe they will put a line in the sand and start offering some of those future EAP enhancements only to those who paid for FSD as a token gesture for not actually providing proper FSD. But there has been no sign of that to date. FSD seems to be all smoke and mirrors with no sign of any meaningful progress toward implementation. Could be years away, long after we have all replaced our current cars. So for me it was a no-brainer to decline the option.

 

[Sawyer8888]

Wise decision @acal. FSD continues to be marketed as an option after purchase for an additional $1K; no more, no less.

If you buried 300,000 pennies into FSD with the hopes it gives you anything more than the "$1,000" in savings, you're probably just looking for ways to sleep better at night. Early access to some L5 capabilities? Maybe, but I think that is highly unlikely.

If you're looking to help advance the technology by giving Tesla an interest-free loan and that makes you feel good, then by all means... got to do what makes you happy.

If FSD becomes reality, all AP2 cars will have the opportunity to have their hardware upgraded if needed for no additional cost, whether owners paid the initial $3K or subsequent $4K. If Tesla changes their mind and decides to charge those paying $4K an additional premium for more hardware, then that would be an awfully bad move on Tesla's part IMO. I certainly would not purchase FSD for more than the quoted $4K. I would be put off and look at other brands.

For those expecting to save $1000, as time passes, that "investment" loses a little more every day. If financed, you're paying interest on that money, not to mention the initial sales tax. Then there's inflation. A dollar today is not worth what was worth last year. Better off having it in your pocket now - or better yet, in an investment that has the potential of making you money. Want to help Tesla while also helping yourself? Buy their stock. (see your financial advisor for the risks involved).

 

[Ben W]

It seems those that are completely against FSD is because they believe there will be NO Incremental improvements over EAP that require FSD. They believe that is is ALL FSD or NO FSD. Am I wrong? I think there will be incremental improvements and expect them to start happening in 2018.

Example: Would STOP SIGN and TRAFFIC LIGHT Detection be part of EAP or FSD? I say FSD.

I use EAP pretty much every time I get in the car. If it could detect and stop at the stop sign or honor traffic lights it wold be a HUGE improvement for me. Worth the $3k I paid at time of purchase.

NAV support on Freeways it supposed to be part of EAP. On Ramp to Off Ramp. But I would assume that would not be the same on surface streets.

I think there is a lot of opportunity for improvements to AP that require FSD and that is the main reason I decided to purchase it. The cost now of $3k vs current $4k after purchase is a minor reason for me. But the $4k is not a guarantee especially if they need to add more hardware.

I agree with you; I expect L3 freeway onramp-to-offramp autonomy (geofenced to a known set of freeways/interchanges/offramps) to be the first differentiating feature between "FSD" and non-FSD, and I expect to see it arrive in 2019 or 2020.

If additional hardware is required (besides an upgraded CPU/GPU/SoC, which is a given), I wouldn't be at all surprised to see the $4k price point go up higher. But the $3k up-front buyers should certainly receive these upgrades for free.

General stop sign and traffic light detection is actually a vastly more complicated problem, since it requires 99.99999999% (ten-nines) reliability: I wouldn't expect this capability to be available (even geofenced) til 2022-2023. Math: a typical driver sees ~100 stoplights/stop signs per day, and there will soon be about a million Teslas driving around. With even 99.999999% (eight-nines) reliability, that's about one missed stop sign / stoplight per day, for the fleet as a whole. And of course, running a stoplight can easily cause injury or death. Reducing this error 100-fold to 99.99999999% would still result in 3-4 incidents per year for the entire fleet, which is perhaps acceptable, but still far from perfect.

What really surprises me the most is that Tesla offered the $3k FSD option at all to 3-year leaseholders, when there was practically no chance that FSD would be meaningfully enabled in that window. Tesla should expect a ton of $3k refund requests when those leases expire.

Disclaimer: I did pre-buy FSD with my Model 3, but I also expect to own it for eight years or so, and I'm pretty sure FSD will be mostly or completely implemented by then. I also strongly believe in the overall mission and would like to support Tesla in that effort, as well as hold their feet to the fire to get FSD working as soon as realistically possible.

 

[Peteski]

I think the fundamental problem with FSD is that it will be very difficult to ever get away from the hands-on-the-wheel, driver responsibility mantra. You are never going to be able to truly relax and sit back from the wheel. Simple scenarios like debris in the road will be a concern, even pothole avoidance. The more semi-autonomous the driving becomes (e.g. stop light recognition, automated on/off ramps), the more potential there is for drivers to become more and more complacent and it will only take a couple of high profile accidents to make those systems revert to "hands-on" operation as we have already witnessed.

When I drive with AP2, it's pretty obvious that the car has very limited awareness of the complexity of its surroundings. It can't see bumps, potholes or other small debris in the road ahead, or even if it can it doesn't make any effort to avoid them. It can't even see all that far ahead compared to a human driver. Real world FSD still seems like a distant pipe dream to me, although AP2 is in itself quite impressive and useful in many scenarios providing you pay attention to its limitations.

 

[BestRadar]

i dont think its worth paying for something that isnt out yet and may not be out for awhile

If I could do it all over again I would not buy the option. I bought my car 11/2016 with the promise that it was coming. 1 1/2 years later I have nothing for my investment. I really feel for the people that leased their vehicles and 1/2 the 3 year lease is up already. Tesla really should stop taking money from people on a hope that this will at sometime become a reality. The thoughts of my car dropping me off at the front door somewhere and then going and finding a parking sport have long faded from my mind.

 

[ItsNotAboutTheMoney]

No. It wasn't a mistake.

_If_ it ever arrives, you'd likely think that it's worth $4k. In the meantime, you have kept $3k.

 

[arubinst]

I agree with you; I expect L3 freeway onramp-to-offramp autonomy (geofenced to a known set of freeways/interchanges/offramps) to be the first differentiating feature between "FSD" and non-FSD, and I expect to see it arrive in 2019 or 2020.

I’m sorry but what you’re describing here is an EAP feature. This would only diferentiate it from AP1 and would be the first E (enhanced) feature of EAP.

 

[boonedocks]

If I could do it all over again I would not buy the option. I bought my car 11/2016 with the promise that it was coming. 1 1/2 years later I have nothing for my investment. I really feel for the people that leased their vehicles and 1/2 the 3 year lease is up already. Tesla really should stop taking money from people on a hope that this will at sometime become a reality. The thoughts of my car dropping me off at the front door somewhere and then going and finding a parking sport have long faded from my mind.

Ditto DID purchase it on my 3-12-17 S100F but I did NOT buy it on our 12-23-17 delivered X75D after seeing absolutely no signs of FSD in the foreseeable future.

 

[Peteski]

I’m sorry but what you’re describing here is an EAP feature. This would only diferentiate it from AP1 and would be the first E (enhanced) feature of EAP.

View attachment 296773

That's a very good point and no way Tesla could back out of offering such a feature to EAP owners. FSD still seems like a million miles away. I'll pay for it when it's a real thing that I can try, but EAP is not even finished yet!

It makes me wonder how Tesla are going to deal with all those FSD buyers who are going to trade their cars back with the feature never implemented. Are they going to offer a full refund for FSD or only its residual value? I would bet good money that FSD will still be unavailable when my car is traded in 2022.

 

[Sully's8]

I think FSD is further away than some people think. EAP is still far from perfect, but at least usable for some driving. I wouldn't even consider paying for FSD at this non-existent stage and very much doubt it will be available in the next 4 years while I own my current X. If it comes during that time and looks good and is actually legal in the UK then I might consider it on my next car. Until then I'm happy with EAP which is showing good signs of improvement lately.

These discussions crack me up. Long way to go on FSD, smart not to buy and might be smart not to be an early adopter when it is available. . .read on.

As for EAP. . . . I use it regularly in bumper to bumper traffic if I catch a rush hour window between the Loop and the burbs. I admit I've thrown it on at higher speed for a second if I'm fumbling for my phone or whatever, but very rarely. It is beautiful at low speed, on the interstate, in low risk (ie non-construction zone) multi lane bumper to bumper traffic...where the risk is a fender bender. . I actually find myself being very productive yet still aware of what's going on on the road. I no longer stress or dread leaving later in the morning for the office, or right at 5 at the EOD. This is the best reason to get EAP.

I bought the car to drive it when it has room to perform, letting the computer do so is anathema to the whole value of the car to me. So when there's room, at highway speed, I LOVE being in control of the car. No sense spending money on a vehicle only to concern oneself with whether or not the computer can drive it. Total silliness in my opinion.

When FSD happens (and I am thoroughly convinced it will- likely in my kids' generation), whenever it is, I will wait until it's literally perfected. My life's worth a lot more on the cost / benefit scale than "surviving" early adopter FSD experiences.

There's already been too many tragedies with EAP. Be smart people.

 

[Ben W]

I’m sorry but what you’re describing here is an EAP feature. This would only diferentiate it from AP1 and would be the first E (enhanced) feature of EAP.

Well, not exactly. What I described (or intended to describe) was onramp-to-offramp driving that is completely autonomous, with no driver oversight needed for the duration. This is subtly but fundamentally different from onramp-to-offramp EAP which would still require driver attention. It's the difference between L2 and L3 autonomy.

No doubt onramp-to-offramp L2 (part of the EAP package) will come first, but onramp-to-offramp L3 may not be far behind. So it could still be a significant differentiating feature for the FSD package within the next couple years.

 

[Uncle Paul]

People do not seem to be taking consideration of all the incremental enhancements that have been released OTA for all the AP-2 vehicles. They bring the FSD closer to reality.

Tesla could have released these enhancements to only the FSD buyers. That would have given them something extra for their money, but instead Tesla has decided to roll them out for everyone with AP-2 hardware.

Tesla is making steps towards FSD every day, but people do not seem to realize this or give them credit.

My year old Model X's autopilot and grown so much better in that time. No more ping ponging, better maps, quicker response, more cameras on line, auto windshield wipers, auto headlight dimming, more accurate trip routing, enhanced side route options, better road congestion displays, etc.

I do not believe that Elon was wrong, just overly optimistic.

 

[S4WRXTTCS]

People do not seem to be taking consideration of all the incremental enhancements that have been released OTA for all the AP-2 vehicles. They bring the FSD closer to reality.

Tesla could have released these enhancements to only the FSD buyers. That would have given them something extra for their money, but instead Tesla has decided to roll them out for everyone with AP-2 hardware.

Tesla is making steps towards FSD every day, but people do not seem to realize this or give them credit.

My year old Model X's autopilot and grown so much better in that time. No more ping ponging, better maps, quicker response, more cameras on line, auto windshield wipers, auto headlight dimming, more accurate trip routing, enhanced side route options, better road congestion displays, etc.

I do not believe that Elon was wrong, just overly optimistic.

The reason they're not taking those into consideration is those were done to reach feature parity with HW1 of AP. The fact that it didn't really add anything new meant that no one really acknowledged what a monumental achievement it really was. It was mostly ignored in a sea of groans about it being late, and not as good, etc.

Tesla couldn't have released those enhancements to only FSD buyers because Tesla had promises of things they'd deliver to EAP owners. Things AP1 owners like myself have been enjoying for years.

FSD is so all or nothing that I really don't see what can be introduced for it where it would appease the 20% of owners that have it without pissing off those that have EAP only.

 

[Ben W]

FSD is so all or nothing that I really don't see what can be introduced for it where it would appease the 20% of owners that have it without pissing off those that have EAP only.

Would you consider onramp-to-offramp FSD to be a potential fit for this? It will likely be ready years before surface-street FSD, and will make a huge quality-of-life difference for those who regularly make long commutes on the highway.

The functional difference would be that onramp-to-offramp EAP (Level 2) would continue to require driver attention at all times, while onramp-to-offramp FSD would not. Undoubtedly this will require computer hardware upgrades to achieve, which would be part of the $3k FSD package, but not part of ordinary EAP.

 

[S4WRXTTCS]

Would you consider onramp-to-offramp FSD to be a potential fit for this? It will likely be ready years before surface-street FSD, and will make a huge quality-of-life difference for those who regularly make long commutes on the highway.

The functional difference would be that onramp-to-offramp EAP (Level 2) would continue to require driver attention at all times, while onramp-to-offramp FSD would not. Undoubtedly this will require computer hardware upgrades to accomplish, which would be part of the $3k FSD package, but not part of ordinary EAP.

A limited Level 3 like what the Audi A8 is supposedly capable would be something that I could definitely see FSD owners getting. I can't see that pissing off any EAP owner. In fact it would likely get a significant percent of EAP owners to purchase the FSD package. I certainly would because low speed bumper to bumper traffic is really annoying.

I'm hoping Audi will have success in enabling that feature in the US to set the stage for Tesla to have it as well.

 

[Jonathantuba]

I have a model X on order. I originally ticked FSD, but then changed my mind as I drove around and thought how likely FSD is to happen anytime soon.

I live in the UK where it is common for roads to be single track with passing places, or on residential roads full with so many parked cars that they have become in effect single lane. Deciding if to go or wait is very much a matter of judgement looking at how the car the other way is driving. Does he look like he will stop, or appears in rush and I better pull over? Often drivers gesture to each other. How is a FSD Tesla ever to cope with this situation? At best it will just stop and beep for the driver to take over.

At intersections and traffic lights, it can often be confusing to even an experienced driver. How is FSD going to decide what light applies to it and obey and what lane it should get in, with often aggressive drivers of other cars pushing in?

I do not see how FSD can ever work safely unless ALL the cars on the road are FSD and they then behave predictably and the cars can communicate between each other as which goes first in single lane sections. Instead of traffic lights, indicators need embedding into the road for the cars to get instructions of if to proceed, or wait at junctions. The present road system was never designed for FSD and a lot is unsuitable.

FSD will I hope come in the next 10-20 years, but I can’t see any time soon. The best we can hope for is ever improved EAP to use driving on major roads.

So after thinking this all out, I changed my specs to EAP and no FSD. I am sure I will change cars before FSD ever becomes a reality on anything but possibly very limited routes.