Tesla: Recalled touchscreens were meant to only last 5-6 years
motorauthority.com
motorauthority.com
My worry with a large number of modern tech heavy cars will be that while the centralisation of a lot of these controls and features into one panel may be convenient to the user, when these inevitably fail (as in this case) - what can the owner do when the parts are no longer available, or if the manufacturer is out of business? Mechanically the car may be sound - but if the system responsible for the interaction between the vehicle and the owner is borked and impossible to replace does that mean the car is a complete write off?
I think back to some of the beaters my friends owned when they started driving, with broken HVAC switch gear, burned out audio head unit wiring harness, etc... while mechanically sound (enough to be road legal...), issues like this didn't hinder or prevent driving the car, and they continued in service for years in this state.
Actual average age of cars on the road in the US is around 12 years. If you are not in the rust belts, it is common to still see cars from the nineties. They are usually driven by people who could not afford a newer car or could not afford major expensive or hard to get parts.
Right, but how many of the people that expect their car to last more than 10 years are buying new cars? The signalling I get from Tesla is that they don't care a whole lot about the used market (as evidenced by this issue).
I think consumers don't realize how much they actually care about the used market. If you buy a $60,000 car that you sell for $20,000 when you're done, you paid $40,000 to drive the car. If you buy a $60,000 car that you can't resell because it's dead or too risky for buyers, you paid $60,000 to drive the car.
Unfortunately, that cost doesn't manifest until years later. Roughly now, based on how long Teslas have been around and their expected lifetime number. I would not be surprised to see Tesla start losing orders if they can't support the used market.
I think consumers care very much about the value of what is typically their 2nd most valuable asset. They see it when they discuss trading in a car for a new one. Or when shopping for a used car. The information is very readily available.
The people not caring about it consciously prioritizing luxury over resale value.
In the case of an electric car, which is simpler in a lot of ways, it seems more likely that parts will just be replaced with third-party compatible equivalents; which might naturally lead to a Ship of Theseus situation...
It’s not heat but only radio/gps thankfully. Looking for a drop in replacement has been challenging as it’s not clear how one will work with the back up cam and steering wheel controls.
I’ve just started looking however
IMHO, the only reason people tolerate that is because the culture was habituated to it in the 90s/00s, when personal computers were improving so quickly that you'd likely want to replace yours even if it wasn't broken.
Cars are so expensive and the culture has been habituated to them having increasing life and reliability that I don't think a car company could pull off shortening their usable lifespan without burning their reputation. Until full self-driving capability gets to a deployable state, I don't really see any technological advancement that could ease people into a faster refresh cycle.
Replacing broken parts was less common. No one really expected to just maintain the machine to the same specs for several years.
Musk's reply sounds like their AI will automatically turn the left/right on and you have to over ride with a touch screen - which to me sounded like a center console... Either way based on the many posts/videos on their current ML I don't think either is a good idea.
yeah I dont drive let alone a tesla and i still have no idea what those shiny (glass?) circles do lol
Anton Yelchin, the actor that played Chekov in the recent Start Trek films died because his Jeep Grand Cherokee has a "joystick" style control for the Park gear selector. It wasn't obvious that it was properly engaged. He got out, the car rolled down an incline, and pinned him against a brick wall until the life was squeezed from his body.
Elon Musk likes to joke that cars are two-ton death machines, but this stuff is no joke.
https://www.theverge.com/2016/6/20/11978044/anton-yelchin-je...
https://www.theverge.com/2016/6/22/12007862/fca-jeep-grand-c...
b) To engage the brake, you have to push the lever, and then it moves back to its original location, like a joystick. This means that there's no positive tactile feedback that the brake is engaged
To quote:
"The problem is that the "Monostable" design doesn’t provide any meaningful feedback about what gear you’re in — it returns to the center position after each shift. To completely confirm if you’re in drive or park or reverse, you have to look at either the LEDs on the shifter (often covered by your palm) or the digital display in the instrument cluster. This has confused thousands of people, led to over a hundred injuries, and now potentially a death. And it’s all because of a design that prioritizes screens over switches."
The 2014-2015 Grand Cherokee was recalled because of 304 vehicle rollaway incidents, 117 crashes, and a death.
There is no excuse for this kind of "too clever" design.
People drive while tired. People drive in emergency situations. People drive while distracted by screaming infants.
You can't blame dangerously bad user interface design on the users. Essential safety settings should be designed so that the operator falls into the pit of success instead of stumbling into the pit of deadly failure.
A motor vehicle is a heavy machine which requires proper operation because it can cause serious injury and death. This is no different than operating a forklift. People need to start taking responsibility for their actions. If folks don't read the operating manual and can't successfully operate a motor vehicle, they have no business driving.
That's a set of goalposts that move automatically to render any argument against it invalid.
No matter how counter-intuitive, error-prone, or difficult the control interface of an automobile is made to be, you can always say: "Well, if you can't figure it out, don't drive!"
Would you fly on a plane designed with this attitude? "It'll drop out of the sky if you accidentally bump anything, without a noticeable warning, but pilots that can't handle flying shouldn't be in the cockpit anyway!"
But seriously: would you get on a 737 MAX without the MCAS fixed?
Would you get in a 737 MAX with an unfixed MCAS a decade after the MCAS incident, when people have forgotten? With a new pilot that had never heard of the two specific crashes?
Or would you insist in flying in a plane designed not to crash into the ground in ways that's counterintuitive for the pilots to deal with?
Similarly, would you let your relatives drive a 2014 model Jeep Grand Cherokee that hadn't been recalled and had the parking brake changed to be automatic? Would you trust your Grandmother to check the tiny little light every time that will stop her dying, or would you take it to the shop for her so that she doesn't have to?
Ford had to recall and fix the 2015 Lincoln MKC. That had a column of buttons in the order P-R-N-D-S-I, the last two being Sport and Ignition. Ignition was the start/stop button, and if pressed while the engine was running, immediately stopped the engine. This tended to happen when someone was trying to go to "sport" mode and accelerate. Worse, it looked just like the other buttons.
[1] https://www.law.cornell.edu/cfr/text/49/571.102
[2] https://www.reifflawfirm.com/push-button-ignitions-gear-shif...
It is very common view, that modern cars are much more safe, comfortable, ecologically sane, faster, and even reliable in first 1-2 years, than old ones, but they will break up and require very expensive repair (engine re-sleeving or re-boring with full rebuild) much, much sooner than older ones.
What is interesting to me, that USA users write about new cars that they expect 150K+ miles from them and it was not so in the past. Completely opposite to what users think here.
Centralization of controls into a single touchscreen is most certainly not convenient to the user, because touchscreens suck [1]. What is convenient is the thoughtful deployment of physical controls in ergonomic locations.
[1] Touchscreens are really only a good choice in situations where there's extreme space constraints, or as an enhancement to existing controls for very specific interactions where they make sense.
All commands must be around the wheel. An alternative that could work is voice commands, if the car can really understand the driver. Good luck to foreign drivers, tourists with rental cars or people living abroad, unless they can configure their own language.
A touchscreen is a dangerous interface for the myriad of simple mundane systems like heat up/down, AC on/off, fan more/less, sound more/less/next.., heated seats, windows... things that you may need very much want to do while driving.
For a car? Are many people driving around in a car from 2000 or earlier? When I search for cars that old or older it's 0.4% of the used market. Clearly the vast majority of cars do not last this long and I think you have an unreasonable expectation.
This sounded unlikely to me, so I checked a local site and got 4.9% >20 years old in the used market. A lot of people drive their old cars beyond the point that they are not worth re-selling, so I'd expect that to be an underestimate of how many cars are that old.
The article linked suggests the average age of a car on the road in the USA is 12 years - if you're then selling cars that are designed to obsolete in 6 years, that has potential consumer law implications.
In rural areas (read lower income) it's not uncommon to see late 80s, 90s vehicles out and about. Some are in good shape, others not so much...
Outside of the US, some countries have heavy import taxes on new vehicles, meaning that the secondary market dominates and older vehicles are more likely to be overhauled than junked.
My newest car is a 2009, purchased in 2018 for $4,000. Everything works except the tire pressure sensors (the batteries in the wheel sensors have died; I'll replace them at next tire change).
My oldest car is a 1969 but that's a toy, so I don't really count it. My oldest car that I use for transportation is a 1983. The air conditioning has failed on that one, but I live without it.
We're flying airplanes that are 30-40 years old. 18 wheel trucks are routinely driven a million miles. I think we need to adjust our expectations that cars need to be replaced every 5-10 years because they're no longer reliable. That's not a wise management of natural resources.
Because they clearly don't often last that long in practice.
Are you saying you wished they lasted longer? Ok yeah that'd be great... but I mean they don't at the moment you can see that for yourself, so expecting it is an unreasonable expectation.
You can expect that cars should last a hundred years if you want... but you're going to be disappointed.
Tesla does not need those screens to last 100 years; but it’s not unreasonable to expect at least 15 - 20 years.
1: https://www.cnbc.com/2020/07/28/25percent-of-cars-in-us-are-... 2: https://carsurance.net/blog/cars-sales-statistics/
Depreciation has little relevance to the majority of used-car buyers who need a vehicle, not an investment. They drive it until it can no longer be driven.
What is more relevant to the lifespan of a car is durability and availability of parts. For example Honda shared parts between the Civic, CR-V and Integra so they are relatively easy to obtain even going back to late 90s models.
"...about 80 percent of all Toyota vehicles sold over the past 20 years are still on the road today." [1]
Used cars are an entry point into an automaker's brand, and they do indeed want their cars to last as long as possible. Furthermore, reliability is a pretty large factor in resale value, and by extension, lease rates.
[1] https://pressroom.toyota.com/america-best-selling-car-camry-...
That does not take into account older cars being exported to eastern Europe.
We were raised to follow this practice and it's hard to imagine doing anything else.
My current car is 20 years old and I have owned it for about a decade. All the other cars I previously owned were also about 15+ years old. I'm not interested in buying a car with a shorter lifespan.
Your service points on an EV are going to be tires and brakes. If the car battery is rated for 1 million miles, how many years is that?
The central control system is a Touchscreen which will be exposed to a variety of heat/cold/Sun exposure how long should it last?
Not everyone can afford to buy a car every 7 years or WANT to do so.
https://www.autotrader.com/car-shopping/buying-car-how-long-...
I don't think you can use this statistic on its own to say how many live >20 years can you?
My previous vehicle was a 1984 Ford F350, which I purchased for $800 in 2017 and then subsequently drove from Seattle to North Carolina with only one breakdown (that I could fix roadside).
With proper maintenance and repair, most vehicles can last several hundred thousand miles and 30+ years. The lack of representation of these vehicles on used car search sites is that most of these inexpensive vehicles are sold private-party. Look at your local Craigslist site for examples.
How many times did you replace the dashboard before ending up with a dash blanket?
Hope you're doing well man. With love from Texas <3
A quick look at eBay shows a nearly identical F350 listed at $21,500, with >190k miles, so clearly someone sees value in vehicles of that vintage.
Share of new cars was 6.9% in 2019, ~5.0% in 2020, and scrappage rate was 5.1% in 2019 (source: https://news.ihsmarkit.com/prviewer/release_only/slug/bizwir...).
If the average in/out flow per year is ~5%, average life is 20 years. If 6%, ~17 years.
There’s a reason old 4Runners and Tacomas and Land Cruisers are worth so much, and barely come up for sale in decent condition.
The sad thing is I can barely find some parts for that 21 year old car that is mostly in perfect running order, but several small buttons don't work anymore, most of the display (showing basic info like outside temperature, fuel consumption etc), rear windshield blade, so at some point good running cars are abandoned because of lack of parts, not because they really need to. Replacing with newer cars is not just the financial cost, but the associated energy and pollution costs that manufacturing processes have.
For electric cars I have huge life expectations, 30 years or more with just batteries and tires as consumables, but the car body and electrical engine should last a very, very long time. There is no reason to accept less than that.
Lots of useful parts for all kinds of cars going back decades:
If not living in the USA, add: https://www.viabox.com/
It had 300k km on it and ran just fine. Newer cars run forever with basic maintenance. I sold it for $1k and the buyer planned to give it to their teen as a commuter, so I assume it’ll be on the road a while yet.
[1] https://www.tesla.com/ns_videos/2019-tesla-impact-report.pdf (page 8)
Yes.
> Are many people driving around in a car from 2000 or earlier?
A fair number, though not a lot because an accident or neglected maintenance will often kill them first. Though the average age of a car on the road has been climbing quite a bit recently, so 20+ years being common probably isn't unreasonable to see during the lifespan of cars sold today. Well, maybe not for the Teslas, but...
> When I search for cars that old or older it's 0.4% of the used market.
That's because there is little demand outside of collectibles, so they mostly aren't worth selling; if you've got a car that old, you are probably going to drive it until it's ready for the scrapyard, not reselling it.
I've wondered about that myself. Because comments I've made that an electric drive train should have an expected service life of 30 years seems to draw a lot of ire from somewhere. Probably the automotive equivalent of found the google engineer.
Same goes for appliances, which are becoming encrusted with features and less reliable.
A modern engine and chassis, etc. but with minimal, tactile, and easily serviceable controls.
An EV with great range (e.g. 300mi), good performance (doesn’t have to be a sports car), and classic controls.
You could go real minimal on computers. You’d just need a microcontroller to monitor speed, temp, charge, battery health, etc. Basic cruise control of the 1990s type would be nice but not essential. A radio with a USB plug for a phone to play music and good speakers is all you need for an entertainment system.
Basically I’m describing an electric 90s Civic with mid range options.
Oh and no cellular modem or spyware.
Put all the money into drive train, battery, and reliability. Design it to have easily swappable and upgradable batteries and to last for 25-35 years.
You’d have one customer at least.
You could probably make a pickup version with killer torque and be like Tesla for red states. Rednecks will be sold on the electric thing once they see what instant full torque means.
But, ignoring engine adjustments, what I'd love in a car is things like using the most common lights; and making the lights easy to access. Because changing the lights is a normal and necessary thing to do.
Don't make it take 30-45 minutes to change the (12V) battery, because changing the battery when it dies is a normal and necessary thing to do.
Basically, design around ease of maintenance, more than making it look cool and hip. I'm not necessarily going to do all the maintenance, but if important things are easier to get to, it saves me money, because it saves my service shop time.
I'd love it if it wasn't 2 cm off the ground, so you could get under it to change the oil; but I'm not going to win that, because I'm exaggerating a bit, but low clearance helps fuel efficiency, and fuel efficiency is required. I'd also love less of the under body panels to help with airflow that get banged up and then fall off or drag and cause terrible noise.
I strongly believe this will happen soon and as a consumer I can’t wait for it. Especially once it turns into a gold rush like the D2C mattress space and everyone is competing to see how much money they can lose to acquire a customer.
I'd love to see more such "dumb" options in the TV and auto markets but I suspect we overestimate the demand for such things because we HN readers are largely tech-savvy. I think the ideal would be dumb devices with open APIs so that third-party smart features could be added, but that's of negative benefit to manufacturers (lack of vendor lock-in, potential for brand damage due to poorly-implemented third-party features, etc.), so I'm not optimistic about that happening.
It's a dumb mistake, but not exactly an uncommon one. I made the same mistake early on and I have contract engineers that propose inappropriate storage solutions to me still on projects. Some very insistently. I think a whole lot of people that work on the embedded side of engineering treat storage as a magic perfect box.
(For instance GM has a strict list of test scenarios for their onstar integration)
https://cadillacsociety.com/2019/10/28/second-lawsuit-filed-...
Sophisticated, user-facing computing is one of the newer parts of cars and so we don’t really know what to expect yet as far as longevity relative to the car itself. However, we should not necessarily see the need to replace computing devices, or parts of them, as prescribing the life of the entire vehicle.
If it was a relatively accessible memory card, it'd be fine for it to have a service lifetime.
They didn't initially design it to fail so quickly, a software update accelerated the wear of the card.
So there's lots of things going on.
People would similarly get pissed off if a $10 mechanical part that was embedded in a $300 mechanical part consistently failed in 3 years. Especially if the failure was due to a change in a fluid spec or something.
I realize there's anger but I'm trying to understand how much of a leap this is beyond a mechanical failure that people don't like, and view as an engineering failure, but nevertheless expect and endure when owning cars of a certain age and mileage.
If you don't have the information, that's okay. I'm looking for it (unsuccessfully so far.)
https://forums.tesla.com/discussion/177785/2016-model-x-touc...
https://youtu.be/85OysZ_4lp0 (2015 Nissan Tsuru vs. 2016 Nissan Versa)
https://youtu.be/5dQbSn94rUQ (2015 and 2019 Jeep Cherokee driver-side small overlap IIHS crash test comparison)
https://youtu.be/ul_uiq-HJPM (2016 and 2019 Honda HR-V driver-side small overlap comparison)
https://youtu.be/C_r5UJrxcck (the most extreme comparison - 1959 Chevrolet Bel Air vs. 2009 Chevrolet Malibu)
I doubt it, with all the electronics and stuff car makers are looking for <10 years duration for new cars.
Their basic warranty is 4 years, and the drivetrain warranty is 8 years. If you buy a used vehicle direct from Tesla they will add on 1 year to the original warranty, but that's it.
That isn't much longer than the warranties you can get on gas-powered cars, and their used warranties seem a bit worse than what you can get from "certified used" programs from other car companies.
In theory, an electric car should last much longer than a gas car, but Tesla's warranty isn't any longer. I bet the price tag to replace Tesla battery packs out of warranty is shocking.
Most items have 1-2 year warranties, but we expect most of them to last much much more than that.
The pressure for a company to have everyone paying them monthly is just to high. You’ll get pulled into their “ecosystem”, where it’s harder to switch. Why sell a car once per 10 years to a dealer who marks it up and has the interaction with the customer. Electric cars already have less maintenance (theoretically). Tesla doesn’t want you going away from paying them, no company does, and they are all fighting to prevent you from not paying them.
It’s not because people consider it a durable good. It’s because people can’t afford or aren’t willing to pay the amount for a new car if an alternative is available.
Many people aren’t going to shell out for a $35k Tesla that only lasts 10 years / 200,000 miles and is worth near zero at the end of 10 years if they can get a Toyota that will last 300,000+ miles, 20+ years, and is still worth a few thousand dollars.
Cars are the second most expensive purchase most people make in their life, they literally can’t afford the luxury of a new one willy nilly (for cars brands aiming to be mainstream brands).
At first I thought "gas wins" because there will be no power grid. But then I thought "maybe electric wins" because getting more gas might be impossible where as getting electricity I can get from solar cells, water wheels, hand cranks, etc... Maybe I takes 2-3 days of charging for every N kilometers of travel but that's better than zero when you run out of gas.
Maybe Mad Max is showing the age of its source material. All their cars should be electric. They'd even go faster!
But if you have to pick one, no question, electric. The supply chains for refining and delivering gasoline far, far exceed the supply chain for charging an electric: Sun, solar panels, inverter, car. Probably some batteries on the house side in there, but if you're charging during the day when the sun is out, those are more or less skipped if you have enough panel.
Mad Max has plenty of problems. "Oh no, guzzoline is in short supply, let's use it as fast as we possibly can!" is not a particularly useful way of approaching resource limits. On the other hand, it makes for some absolutely spectacular cinema! I love that Fury Road was almost entirely practical effects - the giant vehicles driving across the desert look absolutely right, because... well, they were giant vehicles driving across the desert. I understand the Doof Wagon (gigantic wall of speakers on wheels) tended to get stuck an awful lot, though.
At worst a hybrid is basically a gas car with naive virtue signaling. At best it’s a slightly more fuel efficient gas car.
People say similar things a lot in seriousness, but I don't think they know what they mean by "simpler".
Is EFI simpler than a carburetor?
Yes, carburetor is more reparable, if you know what to do, but EFI is much simpler.
It is different attributes.
But it's not obvious to me what simpler for an end user has to do with more reliable. I mean, my smartphone usually dies in about a year or two. Actual phones in olden days didn't.
So I still don't get why electric cars would be considered simpler in a sense that's relevant to reliability in expectation.
Carburetors need adjustment and maintenance, but some cars with ECUs and EFI from the 80s are getting old enough that the electronics are actually failing. So how do you define "simple" and "reliable"?
Exact sizes and moving parts ARE problem.
Yes, EFIs from 80s (35+ years) are deteriorating due do deteriorating of conformal coating and soldered joints. But how much times jets and membranes of actively-used carburetor should be replaced in 35+ life span? Especially, if gasoline is only so-so? Why replacing parts of carburetor is Ok and replacing EFI control box is not?
You could not buy new EFI, but could buy repair kit for carburetor? It is sad, but it is not problem of technology per se and, again, is not about simplicity, but about vendor lock-in. It is another (but valid!) argument.
And I have opposite experience with modern electronics: I'm still using my first (!) smartphone. As main one, not as backup. I'm using my workstation around the clock for 6+ years (I'm not turn off it for nigh, it works, really, 95% if these 6 years non-stop), my home NAS is built on platform which is 7+ years old (yes, HDDs are much newer, but HDDs are MECHANICAL!). My LCD monitor is 11+ years old (!) and still in perfect condition, and my PREVIOUS monitor is still used by my wife without problems...
It is all anecdotal evidences, but I've replaced a lot of sprocket and chains on my bicycle over last 6 years. I need to replace sprockets and chain on my motorcycle each season.
Yes, if only one transistor on my MoBo dies, I'll need to replace MoBo (and CPU, as it is old and you can not buy MoBo for it now) as one piece, I could not replace transistor as chain or brake pads (grr, really, I can, but it is not very common), but still my computer is much more reliable than my (expensive? almost top-of-the-line) bicycle and my motorcycle, and my EFI-equipped motorcycle is more reliable than previous one, with complex carburetor with many fine passages, calibrated jets, special membrane & stuff.
(Edit: typos and grammar).
That is terrifying, and this sort of thing is why I don't seriously consider Teslas now I'm looking to buy an EV.
Before I believed all model 3 cars should have at least the option of a dashboard to show status in the direction you are looking.
I thought the Model S/X would be higher cost so they wouldn't try to cost-reduce everything like the model 3.
There are also renderings with a traditional wheel buried in the HTML.
The outrage drives dozens of articles and a hundred million clicks, making the car the most talked about new vehicle for the next several weeks.
“You won’t believe this one ridiculous thing about the new Tesla!”
Mission accomplished. I would bet $100 that the Model S ships with a physical way to activate turn signals on the steering apparatus.
Aside from turn signals, I think they will try to do away with the physical stalk to “gear shift” and it will ultimately be an improvement.
As-is the right hand stalk is overloaded with functions for cruise and autopilot, on top of gear shifting, which makes the current incarnation sub-optimal in several specific usage scenarios. So it doesn’t surprise me they are looking to change how it functions.
I ended up with an EV that is just a normal car that happens to be electric, and it's great. I'm in the market for another EV now, and despite the fact that the Tesla X is technically the best solution for what I want to do (towing) it's not even in the running because of all the rest.
Sometimes tactile button and switches can't be replaced by touch screens.
The touchscreen is not a problem. They did have a screen yellowing issue due to not enough UV applied to cure a screen adhesive.
The real problem is the EMMC wearing out. And the root cause is that they're logging so much data to it continuously.
They could have chosen to log less data.
Tesla did reduce the excessive logging and even implemented a notification when the car suspects it is failing. I think they even extended the warranty, but of course reducing the logging will push some failures outside of the warranty period.
This recall thankfully fixes this problem permanently for those who could be affected.
But then again car touchscreens are some of the slowest pieces of tech around it seems.
A screen in a car needs to be designed to take into account the environment, and it looks like in this case it wasn't.
The article says that in paragraph four:
The government agency identified the source of the problem as the touchscreen's NVIDIA Tegra 3 processor and its integrated 8GB eMMC flash memory chip. When the chip reaches "lifetime wear," it can no longer maintain file-system integrity, causing failure of touchscreen functions, according to the NHTSA
Will that Tesla S be the same in 2045 to 2090, or even in 2120? It's looking like the answer is no. The only ones left will be mounted on platforms in car museums. There will not be auto shows featuring real Tesla cars, because the electronics will be dead and impossible to fix. No reasonable part replacement will be possible because of digital signatures.
We'll see some car bodies that get hollowed out and then draped over the innards of newer cars. Somebody already got an unsupported Tesla S running again by installing a gasoline engine, even cutting through the floor to install a transmission tunnel.
Bah. Just strip out the electronics and replace them with something newer and more open. Keep the body and the motors, replace the batteries and the computers.
That's impossible. Maybe an AI could guess left/right turns, but signals are also used to signal merges; if I need or want to merge, how could the car possibly know, and know which direction? In a normal vehicle being operated responsibly, the only indication that a merge is going to occur is the driver engaging the signal.
(Though I admit, many a driver doesn't start to signal their merge until their car has half left the lane. A signal then is often useless: I already know they're merging because the car is drifting recklessly into another lane.)
If it's intended to be totally autonomous, why does it even have a steering wheel?
The lack of drive selector is troubling.
I wonder if tesla will get no more unintended acceleration lawsuits -- or if they will start to lose each one.
He said that Tesla is well known for using general consumer grade parts (he mentioned the touchscreen as an example) instead of automotive parts (which have typically much longer expected livetimes). That allows the to push the price down but also to iterate faster.
The worrying bit is not really that Tesla does it, but that the mandated strategy at that manufacturer is to only compare themselves with Tesla (all others don't count), so I think we will see a drive to lower quality from all manufacturers to emulate tesla. This is a well know premium manufacturer.
My prediction is that the RoHS regulations that brought us lead-free solder will go down in history as the cause of massive environmental contamination around the world. In other words, exactly opposite what they thought they would achieve.
The revamped Model S and Model X still have big central touchscreens (now in a landscape orientation, rather than the previous portrait-oriented screen), but Tesla went even further by eliminating the turn-signal and gear-selector stalks. Tesla CEO Elon Musk said the car will simply "guess" which way the driver wants to go based on sensor data and navigation maps. Tesla is expected to include backup controls as well, leaving drivers to use the touchscreen for selecting drive, and steering-wheel controls for turn signals.
Just does not sound like a good idea at all.
Our van glove compartment repair was quoted at $300 for a broken plastic latch...
Its not the touchscreen its the emmc. Yeah that sucks but its not like they're bricking old cars with expensive or unserviceable parts.
It could have been a cartridge or easily serviceable module.
Especially if they are arguing that it's normal for the memory to fail in 6 years. I can't imagine all of my center console buttons failing after six years and requiring a whole new center console to get up and running again.
Yeah, I know Tesla sells things besides cars. But I think their reputation will be based on things they sell to the public, and I think most people will judge them on their cars.
I would say they're going to have a hard time of it.
I honestly think they were hoping to make that up in volume. As it gets cheaper to make cars at scale, some portion of the savings goes to fixing up the issues with the cars they already shipped. They haven't been able to scale their pipeline, though.
> I think Tesla is going to earn a similar reputation as other car manufacturers as time goes on.
Long term, I don't think it's going to go well. "Move fast and break things" is a bad principle to apply to a car. I can use DDG or Bing if Google doesn't work, but I can't summon a Kia when my Tesla won't start. I don't think they can use that velocity to create features consumers want enough to deal with the instability. They might strike gold elsewhere though where the instability is less of an issue. It might be okay for self-driving semis to break down, if they're cheap enough that you can afford extras to use when some break.
But it's a good candidate for someone looks for a reliable car. Someone hit me when the car on auto pilot. My hands are on the wheel but it's really hard for Tesla to take back control. The car was shaking and I have to apply a force to the drive wheel to force it went straight.
Tesla as a car feel very fraigle :-(.
but on the other hand, driving it is really fun.
Owning both a Model S and X, I’ve seen Tesla first hand attempt to evade covering warranty work on these models.
Friendly reminder to opt out of arbitration when you buy your Tesla (when the VIN is issued).
And, I assume, knowing what that would mean later.
https://www.thedrive.com/tech/27989/teslas-screen-saga-shows...
* The memory should have lasted 6 years not 3.
* The NHTSA is unreasonable when it requires the system to last the expected lifetime of the car (12 years).
It might just be ass covering, but currently the official stance from Tesla is that a car that falls apart after 6 years should be enough for anyone, because anything else is too hard.
What’s the point of EV going green if the life expectancy is 1/4 of a normal petrol car.
My Toyota is 12 years old now. Still runs and drives like new with basic maintenance.
https://www.forbes.com/sites/jimgorzelany/2017/01/04/passing...
The irony is that hybrid/electric vehicles are being owned longer than fossil-fuelled vehicles before being sold.
This is definitely a Tesla mistake, probably simply owing to lack of corporate experience and long-term planning. Could even be poorer access to (specialized auto-grade) parts owing to their small scale in the past.
But worse than the intentions is the fact that Tesla doesn't treat these things are design flaws that should be addressed even out of warranty. Is it right hat they feel its ok to profit from these flaws?
If you drive a car with the Tegra CID and current software, you get the distinct impression that they really wish all those cars were dead.
They probably also do want cars with lifetime supercharging dead (which are primarily MCU1 vehicles).
If you search around, you will also find countless posts of people being denied warranty work for their Tesla's or fixes only being billed as good will after the customer fights back.
Every company I've ever worked at has done this analysis and also set up metrics to catch a component at the system level causing heavy flash wear by excessive logging, caching or other write heavy operations.