Not a chance. Long term the price will drop like every other advancement in automotive technology. Just a question of when and how fast.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Not a chance. Long term the price will drop like every other advancement in automotive technology. Just a question of when and how fast.
I am on the fence as well, not sure if there is enough usage for me to pay $5000 up-front. However, the additional $1000 if activated in the future sucks!
So, my question is: will i pay taxes if I activate EAP later? That way, at least I can take some of the bite of the additional $1000 out if I change my mind and decide I want EAP after all?
I have heard that if I activate it at a Tesla service center that is located in a state that doesn't charge taxes on software and/or states that don't have sales taxes at all (Oregon for e.g.), i would not have to pay taxes on the $6000?
So, my question is: will i pay taxes if I activate EAP later? That way, at least I can take some of the bite of the additional $1000 out if I change my mind and decide I want EAP after all?
I have heard that if I activate it at a Tesla service center that is located in a state that doesn't charge taxes on software and/or states that don't have sales taxes at all (Oregon for e.g.), i would not have to pay taxes on the $6000?
WhiskySTL
Member
That's a really interesting point, if so. It makes sense.
Out of curiosity, I just checked, and my state does not appear to charge tax on software. However, of course, our sales tax is nowhere near 20%, so... it'll still be better to buy it up front.
But that does do something to lessen the financial impact for those would would rather wait to buy the software.
I don't think so. I see two choices people will have in the future with regard to fully autonomous vehicles.
1. Buy your vehicle and pay a recurrent subscription for autonomous capabilities
2. Utilize self driving taxis on a pay-per-use basis.
Both of which will be more than $8000 over the course of the average length of new car ownership (8 years).
Self-driving car software is more complicated than any end-user facing software ever devised. You'll have to pay for server farms to store and process data. Data network fees for V2V and V2G. Armies of software developers to tweak and tune the software further and further along. All these cost money. As much as I hate to say it, recurrent fees is the better way to pay for this.
ProphetM
Member
I will probably not be getting EAP, simply because the car will already be pretty expensive without it. It'll be a stretch for me to afford a Model 3 so I'm going to pay for the long range and premium package so I can get it as soon as I can and then wait until later to buy EAP.
Glamisduner
Active Member
That's not a functional autonomous solution then. If it keeps needing tuning etc it's Beta software. At most a functional fully autonomous solution would require navigation map updates, but even if the map is out of date all that should result in a navigation failure. All the rest of that is not a core requirement of autonomous driving.
Sandiegodoug
Active Member
James Anders
Member
I guess savings will very greatly among individuals.. if you live in Texas where electricity Is super cheap and you drive a gas guzzler then you will for sure save a fortune (even if you don't drive too much)
If you drive in a eco state like CA like me and drive a car with an acceptable gas mileage you are screwed.
If you drive in a eco state like CA like me and drive a car with an acceptable gas mileage you are screwed.
Anyone wo thinks a privately owned 50k Model 3 will be competitive with whatever utilitarian vehicle Google comes up, once cars can drive themselves is delusional. So the Tesla network won't really save you anything. So why pay more? Because it's the only option for FSD?
Not really. Since there are many different OEMs working on self driving cars, from traditional car manufacturers to new startups, from suppliers like Bosch to tech giants like Google, Tesla won't have a chance to be the only one to sell FSD cars.
It often seems that Tesla would be ahead, but IMO they aren't. Making the cars upgradable in the future is a cool idea, but it makes it seem as if Tesla would be very close to FSD, which they aren't. Tesla's system really isn't that good right now, they just allow it to do more than others allow their systems.
Nissan's ProPilot for example doesn't have lane change on highways, but probably not because they can't implement it, but rather because they think its too dangerous if it doesn't properly work.
Tesla has a different approach. They allow more, but want their drivers to double check. Which is ok, too. But it leads to more confidence and probable abuse. I for example never take the hands off my AP1 Model S, but I know people that do, which can lead to accidents and unnecessary headlines.
Long story short: at some point there will be many cars offered with self driving capabilities, which will drastically lower the price to obtain one.
I'm an autopilot pessimist, and do not plan to buy it unless the cost drops dramatically. But to be fair to Tesla:
1) The sudden divorce with MobilEye was devastating, and a massive setback to Tesla's autopilot plans. They had to switch to their nascent internal programme with new hardware, which wasn't at all ready; they're just now starting to catch up.
2) Companies like Google are doing their self-driving tech with the inclusion of LIDAR. Which is a wonderful mapping tech. It also cost $75k per unit when Tesla first got into self-driving. Today it costs "only" $7,5k per unit. Tesla deliberately chose to handicap themselves by not including LIDAR in order to ensure that the self-driving tech that they do develop will actually be affordable, and so that they can include the capability for it on every vehicle. They do pay a price for this, of course. Photogrammetry tends to be prone to misstitching problems when creating 3d models, and radar (which they haven't been using for very long, and which they assign a low trust rating to as a consequence) "sees" the world in an unusual manner (aka, a piece of aluminum foil is blazingly bright, but a piece of plywood is transparent; humans are ghostly translucent figures)
I'd take it as an encouraging sign that they're now collecting significant data from user reports of autopilot failures to reproduce the problems. It suggests to me that on their development builds that they think they've gotten the "easy" problems solved (the ones that they can readily recreate themselves). But Tesla has to be very cautious about deploying new AP builds, because if they ever screw up on a build and it significantly increases the accident rate, the amount of liability that they could face for that would be huge. Hence how Tesla tends to phase-in new cameras and sensors slowly, initially with low priority given to their readings, and only slowly ramping up the trust in them with time.
Last edited:
One thing, BTW, that I have mixed feelings about is that Tesla's desire for increased data is going to run into conflict with their desire to shift the mobile services bill to consumers (aka, no free data connectivity). If they're more and more using my data to improve their mapping services (aka, the more they start downloading my radar / imaging / driver input data so that they can build a better model of the world around them and figure out how everything "should" look), they better not be expecting me to pay the bill for that. As far as I'm concerned, Tesla is free to download gig after gig of data from my vehicle, so long as they pay for it. (Or download it over wifi, that's fine too).
1) But MobileEye continues to exist and they have at least a year head start on Tesla.
2) IMO it's a mixture. Audi can do L3 Autonomy in some situations with cameras and lane keeping plus TACC is almost universally known. The LIDAR vehicles do a lot better and while they are more expensive, not sure where you got the numbers from though, that cost will come down.
On your point 3, which you din't call 3): I totally agree, being conservative is necessary. But any other OEM is even more conservative. Maybe Tesla might have the right mix of conservatism and progressiveness, who knows. Essentially I disagree with the assumption that Tesla is the pioneer in self driving.
I'm simply explaining that the fact that Tesla has fallen behind is perfectly understandable, given the circumstances, and hardly indicative of some trend that will continue indefinitely. I'm not disputing that Tesla has fallen behind, at all
I do think that Tesla will eventually catch up and exceed its competitors, for the simple reason that they're prioritizing it more and are more flush with investment cash.(That said, I still remain a pessimist on the technology in general)
Those numbers come straight from Google. Tesla could always incorporate LIDAR sensors later when they're cheaper (or if/when Tesla thinks that they can make them cheaper), as yet another data source, but even after the first order of magnitude price reduction, it's still 1-2 orders of magnitude too expensive.
Wow, ok that's impressive, I thought it would still be way more expensive. Another cost improvement like that and they can use it in basically any car.
Oh God I hope we don't need a LIDAR retrofit soon.
Even $750 would be pretty expensive to require in every car. And order-of-magnitude cost improvements are hard fought. Achieving one order of magnitude is something, but two? Let alone more than two? Not saying it can't / won't happen, but I won't stay up at nights waiting for it.
Note that I haven't brought up how LIDAR domes are rather awkward to incorporate into vehicles, the way they are at present. Your car basically needs to wear a "hat" for it to be able to see to all sides, sticking up well out of the top to avoid self-obstruction. Draggy and ugly. You could compensate this to some extent by incorporating multiple lidar domes in different parts of the car, but then that's just multiplying the cost problem. A lower-profile LIDAR (though still something that sticks out) could give you long-range-only LIDAR, but then you'd still have to rely on cameras / radar to see things that are closer. Perhaps Tesla could incorporate a long-range-only LIDAR underneath a glass roof. But I'd hope it wouldn't be annoying with noise from the spinning, or that they find a way to make a cost-effective non-spinning version....
LIDAR just has so many problems for self-driving cars. Great data, but low on the "practicality" aspect. Radar is fundamentally different because phased array antennas are cheap to make and, perhaps more importantly, self-steering ("aiming" done purely in software). Ultrasonic sensors are cheap because they're such simple devices. And cameras are cheap because we produce them by the billions.
Last edited:
I don't think you'd really care how the car looks like, if it's something you just hire to drive you from A to B. And I don't even think you need the "hat". LIDAR is based on sending out a laser and then detecting it's reflection. I see no reason why it has to be a hat on the car. I's probably easier to retrofit, but I think a car made with LIDAR in mind could just have the sensors and lasers built in the bumpers.
And let's be honest here, there is no reason why LIDAR should be expensive at all. Why should it be more expensive, than a camera?
IMO it has one problem: cost. And since we are arguing on a sub-forum dedicated to a vehicle that was thought to be impossible to build not too long ago, for exactly that same reason, it seems to me cost really isn't that much of a problem.
Cost has gone down 90% without even making LIDAR a mass market product, now how cheap will it be in 5 years? Probably pretty cheap, I'd say.
Since when? Anything from Audi that`s slated for L3+ and actually is in production has a Lidar on board (so far only the new A8).
And concerning general LIDAR costs:
<1000 bucks per unit from 2018 onward and definitely superior for cities,pedestrians/ cyclists etc.
To get that kind of precision data from ultrasonic, camera and radar.......
Well, let`s just say there`s a reason why every development car which targets lvl 4+ autonomy has lidar sensors so far.
That´s another reason why i`m really sceptical concerning EAP and FSD with the current Tesla hardware.
Imho real autonomy will take another few years and LIDAR sensors will be a dime a dozen by then which is why I think Tesla`s route to autonomy will hit a dead end if they don`t change their approach..
Last edited:
