Tesla Full Self Driving (FSD) info - interesting stuff!

[max_torque]

Whatever your personal opinion of Tesla, you have to admit they employ some good engineers an let them do good engineering, see the long talk on the new FSD system, including details of the ASIC developed to handle the data streams at an incredible rate:

http://youtu.be/Ucp0TTmvqOE?t=4331

[SilverSolder]

Very impressive presentation, far more detailed overview of the kind of AI that goes into making a self driving car - and very interesting hardware!

[Dubbie]

Some of those numbers are extremely impressive!

They really are much further along than I ever appreciated.

[maginnovision]

I think the actual metric is the cars driving themselves of which they don't currently do. GM is further along than them in production. You are actually allowed to let go of the wheel. Waymo seems ahead but as of now it's not exactly a public operation.

[Dubbie]

Did you watch the video?

Your comments suggest that you didn't.

[maginnovision]

I'm not interested in the bits that aren't relevant, just the end result. I've been reading all about it but I did not watch any videos so you are correct.

[wraper]

I'm not interested in the bits that aren't relevant, just the end result. I've been reading all about it but I did not watch any videos so you are correct.

All your post is not relevant to actual situation. They even made full self driving demo rides (signs, traffic lights, junctions) for attendees. By the end of the year, it is supposed to be released for actual customers.

[wilfred]

I think the actual metric is the cars driving themselves of which they don't currently do. GM is further along than them in production. You are actually allowed to let go of the wheel. Waymo seems ahead but as of now it's not exactly a public operation.

Someone wake me up when they don't have a steering wheel or controls of any sort. According to family folklore my grandmother was confident if my grandfather stayed too long after the council meeting at least his horse would know the way home. That's the car I want.

[Dubbie]

Watch the video Wilfred, your point is discussed at length. ~3h:25m

[extide]

Wow, only watched about 30 mins so far, will have to catch the rest later but I love love all the details on the SoC itself, die size, fab node, floor plan, etc. This must be the project that Jim Keller* was working on. I absolutely love the details of chip designs and stuff, and am surprised they have come forth with so much info, definitely very cool! They obviously designed this chip for running a very specific workload and optimized the absolute crap out of it. The NN accelerator is very lean. Does exactly what they need and nothing else. I wonder what GPU IP they used, based on the floorplan it looks like a MALI or maybe PowerVR design, (and maybe they reveal it later in the video, I haven't seen it all yet), but it would be very interesting if they designed their own GPU IP. I kinda doubt they did because that's not really the focus here, the neural network accelerator obviously is and that is definitely all custom, no point in designing a GPU IP when an off the shelf one is fine for what they need. I like the the quad-core clusters of A72's here, they look to have quite a bit of cache as well. I wonder if they are using an off the shelf ARM interconnect there or if they have rolled their own for that. I'd guess it's a CCN-500 series although I'm not sure. I wonder if they will move to one of the newer designs in the future allowing fewer larger clusters (perhaps 2x 8core A76 clusters in next gen).

*Jim Keller was the brains behind some very successful chips like some of the Alpha designs in the 90's, AMD K7 Athlon, AMD K8 Athlon 64, Apple A4 and A5, and AMD Zen. He is now at Intel, purportedly working on their push into discrete graphics.

[maginnovision]

I'm not interested in the bits that aren't relevant, just the end result. I've been reading all about it but I did not watch any videos so you are correct.

All your post is not relevant to actual situation. They even made full self driving demo rides (signs, traffic lights, junctions) for attendees. By the end of the year, it is supposed to be released for actual customers.

If they release full self driving I'd be amazed. They had hiccups even with their fully setup mock drive. I hope for all tesla drivers that FSD does come out one day because they don't seem to be thinking. Seriously they've been saying FSD next year for 3 years. What I don't know is what's different this time? Why was the FSD hype train event the day before a terrible quarter financial document release. Obviously I don't need the answer for that.

In all seriousness though I could never get behind tesla and their FSD program when the CEO says LIDAR is a crutch, developers using it are doomed and at the same time they don't seem to be doing anything the others aren't. I definitely do not think FSD will come out this year or next but maybe they'll upgrade all the cars they're obligated to by the end of next year. Then the next computer will come out and it'll start again. From my POV the hardware itself is about as interesting as a crypto miner ASIC. It's cool but I don't want it. From what I recall it's faster than the old Nvidia system they had but still slower than the newer Nvidia systems. Time will tell if that pays off but the reduced energy consumption probably is required for their cars.

[Dubbie]

and at the same time they don't seem to be doing anything the others aren't.

They went over at length all the things they are doing that others aren't.
Training NN with a constant stream of real automatically annotated data is a pretty huge thing. nobody else has that. Like they said, it's the long tail of rare events that cause problems.  Look at the lady crossing the road carrying her bike that was killed. Having a NN that is trained well enough to deal with odd scenarios like that requires absolute torrents of real world data.

I'm not addressing your question of if this is the time that level 4-5 self driving will really happen, time will tell on that count. But you can't deny that they are doing some extremely cutting edge stuff here.

[maginnovision]

and at the same time they don't seem to be doing anything the others aren't.

They went over at length all the things they are doing that others aren't.
Training NN with a constant stream of real automatically annotated data is a pretty huge thing. nobody else has that. Like they said, it's the long tail of rare events that cause problems.  Look at the lady crossing the road carrying her bike that was killed. Having a NN that is trained well enough to deal with odd scenarios like that requires absolute torrents of real world data.

I'm not addressing your question of if this is the time that level 4-5 self driving will really happen, time will tell on that count. But you can't deny that they are doing some extremely cutting edge stuff here.

I think that's a poor example though, Ubers car never should have hit that lady. It simply wasn't setup to stop. It knew she was there and it knew it was going to collide it just wasn't allowed to do anything. Tesla may also come into the scenario where they overfit leading to extreme examples of WTF(I'd argue it already has happened many times but you don't have to) or underfit and also fail to react.

I don't really know if I trust Tesla to be honest about everything either, it's not their strong suit. For anything they're truly doing that nobody else does they deserve credit, assuming it's helpful. I also like that Waymo has their simulator where they can cook up a scenario, run it and train against it. You actually get to build a sort of similar thing in the autonomous vehicle program at Udacity that culminates in control of a real car. For such a complicated problem where a single death could literally end the Tesla story I think they are being very stubborn assuming they know better than everyone else. The whole autonomous day was nothing more than a publicity event and I really hope they don't try and push out some half cooked solution. Every company working on this problem is doing cutting edge work though, Tesla is literally no different in that respect and nobody likely has a really good idea of where everyone else is at. Regardless of what happens I think it's really interesting to follow and look forward to hearing more from every company involved in solving this problem but I don't expect Tesla is years ahead like they think they do. They may just not be aware of what they don't know.

Just to add... If anyone has a link to an overview of everything they say they're doing I'd like to read it but I have no interest in a long video.

[Dubbie]

You may have missed it but they talked at length about the reasons why simulations are a poor substitute for real data. When you see the two compared, it’s very obvious. The real world images have so much more variety and complexity.

You also mentioned that they may not know what they don’t know yet. I think they are past this stage. Their system can run in shadow mode where a human driver is driving the car and the computer is constantly making predictions about what it would do next if it was driving. It can self check simply by waiting a minute and seeing what actually happened. This is vastly different to running things in a simulation as there is a vast depth and richness of scenarios and behaviours that could never possibly be replicated in a simulation.

I think at one point they mentioned that to have realistic enough other cars in the simulator, you’d almost have to solve the self driving problem!

[maginnovision]

You may have missed it but they talked at length about the reasons why simulations are a poor substitute for real data. When you see the two compared, it’s very obvious. The real world images have so much more variety and complexity.

You also mentioned that they may not know what they don’t know yet. I think they are past this stage. Their system can run in shadow mode where a human driver is driving the car and the computer is constantly making predictions about what it would do next if it was driving. It can self check simply by waiting a minute and seeing what actually happened. This is vastly different to running things in a simulation as there is a vast depth and richness of scenarios and behaviours that could never possibly be replicated in a simulation.

I think at one point they mentioned that to have realistic enough other cars in the simulator, you’d almost have to solve the self driving problem!

If they release FSD this year I'll believe what they said if not I'm going to say they're just trying to tell everyone they're right and everyone else is wrong. The fact is Tesla's all vision system has already resulted in 2 decapitations and many fatal accidents. The only saving grace they have now is they get to pass the blame to drivers. "Should have been paying attention, should have managed it themselves". They won't have that to fall back on if they call it FSD. I guess we'll see what happens in the future. Just don't forget to question them when they say they are the experts and their demo is no more impressive than a years old Google demo.

[wraper]

If they release FSD this year I'll believe what they said if not I'm going to say they're just trying to tell everyone they're right and everyone else is wrong. The fact is Tesla's all vision system has already resulted in 2 decapitations and many fatal accidents.

Moot point. Still less accidents than with human drivers in average. Not to say those were negligent drivers abusing the system. Nothing will be 100% perfect ever, and Tesla makes the safest cars you can currently buy. Basically all naysayers like to call upon anecdotal evidence and avoid hard statistics.
One death in Tesla somewhere in China - big news. 100 people die in car crashes in US every day - who cares.

[extide]

Quote from: maginnovision on Today at 07:29:53 am

If they release FSD this year I'll believe what they said if not I'm going to say they're just trying to tell everyone they're right and everyone else is wrong. The fact is Tesla's all vision system has already resulted in 2 decapitations and many fatal accidents. The only saving grace they have now is they get to pass the blame to drivers. "Should have been paying attention, should have managed it themselves". They won't have that to fall back on if they call it FSD. I guess we'll see what happens in the future. Just don't forget to question them when they say they are the experts and their demo is no more impressive than a years old Google demo.

So, I think you need to be a little bit realistic as far as the safety incidents go -- I mean it will never be zero, or at least not for a long time. It will however be much better than a human driver and that's an improvement we will see right away. All of this same sentiment will apply to any other manufacturer, too.

As far as what Tesla is doing that nobody else is ... you have to talk about that ASIC. I don't know if anyone else is building a fully custom ASIC for this purpose, but but I can't imagine that very many if anyone else is. The ASIC is extremely efficient, although not that fast ultimately. It has 2 neural network processors that do 36TOPs apice in each chip, for a total of 72TOPs per chip. For example the TU102 chip used in the GTX 2080 Ti can do 228 TOPs. However TU102 does that with 754mm^2, 18.6B transistors, and 250W. The Tesla ASIC is only 260mm^2, 6B transistors, and like 30W. That Tesla ASIC also has 12 fairly large ARM cores AND an entire GPU on that same die as well.

(All TOP numbers are for INT8)

[wraper]

For example the TU102 chip used in the GTX 2080 Ti can do 228 TOPs. However TU102 does that with 754mm^2, 18.6B transistors, and 250W. The Tesla ASIC is only 260mm^2, 6B transistors, and like 30W. That Tesla ASIC also has 12 fairly large ARM cores AND an entire GPU on that same die as well.

(All TOP numbers are for INT8)

Those figures are completely incomparable FWIW.

[extide]

How so, they are FMA dot products, just like the tesla chip is doing. Smaller matrices, though, but I don't believe that makes them not comparable as those numbers are including each add and each multiply for each position in the matrix. (16x16 vs 96x96)

[Bud]

. By the end of the year, it is supposed to be released for actual customers.

By the end of Last year Musk was supposed to release space tourism. I am still looking for a space tickets sales office. The problem is this idiot smokes too much.

[SilverSolder]

There was a recent thread hereabouts on why software projects are prone to be late.  It seemed the "fudge factor" ended up at about 1.8 ...  so if Tesla promises fully automated driving within one year, they might actually make it in two!

[wraper]

. By the end of the year, it is supposed to be released for actual customers.

By the end of Last year Musk was supposed to release space tourism. I am still looking for a space tickets sales office. The problem is this idiot smokes too much.

Please stop posting this nonsense misinformation in every thread remotely related to Musk  . Please provide any reference to your words. Oh you can't, because there isn't.

[james_s]

I wish they would focus on making innovative high performance and practical EVs. I could not care less about self driving, the whole point of owning a personal car is that I get to drive it, when I don't feel like driving, I take the bus which is a much more efficient way of moving people.

I think we are still at least 10 years away from fullly autonomous cars, possibly longer. Driving a car outside of carefully controlled conditions is just too complex of a task. Computers are good at the mechanical aspect but way back when I took drivers ed I remember the instructor emphasizing that driving is primarily a social activity and computers still utterly fail when it comes to nuanced social interactions. It will be trivial for people to troll autonomous cars and exploit their predictable behavior and I suspect a lot of people will make a game of it.

[Dubbie]

It will be trivial for people to troll autonomous cars and exploit their predictable behavior and I suspect a lot of people will make a game of it.

You could do that now!
You could put a brick in a paper bag and trick people into running over it. You could turn around signs to point the wrong way on dangerous corners. There are many ways to “troll” human drivers.

But guess what? People on the whole aren’t psychopaths and aren’t running around trying to cause car accidents by exploiting weaknesses in whatever way cars are currently driven.

[CatalinaWOW]

I wish they would focus on making innovative high performance and practical EVs. I could not care less about self driving, the whole point of owning a personal car is that I get to drive it, when I don't feel like driving, I take the bus which is a much more efficient way of moving people.

I think we are still at least 10 years away from fullly autonomous cars, possibly longer. Driving a car outside of carefully controlled conditions is just too complex of a task. Computers are good at the mechanical aspect but way back when I took drivers ed I remember the instructor emphasizing that driving is primarily a social activity and computers still utterly fail when it comes to nuanced social interactions. It will be trivial for people to troll autonomous cars and exploit their predictable behavior and I suspect a lot of people will make a game of it.

I actually like the self driving car concept.  It has been more than forty years since I lived and worked somewhere where riding the bus was an easy option.  The closest it has come in the recent past involved several miles of walking, and currently I have a ten mile walk to the nearest bus stop.  While I enjoy driving sometimes, most of the time it is a necessary evil.  Particularly on long vacation trips I would enjoy the ability to sit back and enjoy the scenery.  But even on the daily trips it would be nice to use that time for research, pleasure reading, thinking about the problem du jour or whatever.

[wilfred]

It will be trivial for people to troll autonomous cars and exploit their predictable behavior and I suspect a lot of people will make a game of it.

You could do that now!
You could put a brick in a paper bag and trick people into running over it. You could turn around signs to point the wrong way on dangerous corners. There are many ways to “troll” human drivers.

But guess what? People on the whole aren’t psychopaths and aren’t running around trying to cause car accidents by exploiting weaknesses in whatever way cars are currently driven.

A nice juicy bug in the LIDAR sensor ought to do it.

[james_s]

I just don't see it happening for a long time. It can be done with great difficulty in carefully controlled environments but out in the real world there are too many edge cases to consider. A friend of mine predicts that at some point one of these companies will get 99% of the way there, release a bunch of self driving cars to the market and within a short time there will be several big incidents that kill a bunch of people and that will be game over for a while. I tend to agree.

Regarding trolling, there is something inherently more appealing about messing with machines it seems, there is no shortage of examples of people hacking stuff and inconveniencing people just for giggles. I don't think there will be a lot of people maliciously trying to hurt people but I would certainly expect to see jammers that cause the cars to stop or people putting stickers on speed limit signs to make the cars slow down. It doesn't matter that most people don't do these things, it takes only a few who do. It's almost impossible to merge in many areas, I suspect the machines will not be aggressive enough and will frequently get stuck trying to merge onto busy freeways. Then there is all the accidental stuff like debris or spills in the road that obscure lane markings or snow, that will be a blast watching a bunch of self driving cars try to navigate a snow storm.

[SilverSolder]

Then there are the inevitable young guys that will find ways to hack their AI cars to snake its way through busy downtown traffic at breakneck speed, with millimeters of safety distances for passing and merging!  [Hmmm sounds kind of fun!]

Elon, if you are reading, this is how we need self driving cars to perform:

[wraper]

Then there are the inevitable young guys that will find ways to hack their AI cars to snake its way through busy downtown traffic at breakneck speed, with millimeters of safety distances for passing and merging!  [Hmmm sounds kind of fun!]

Elon, if you are reading, this is how we need self driving cars to perform:

And how they are supposed to hack it? They spent some serious effort to ensure that self driving processors won't run firmware not signed by Tesla. That's discussed in the video.

[Marco]

release a bunch of self driving cars to the market and within a short time there will be several big incidents that kill a bunch of people and that will be game over for a while. I tend to agree.

Tesla could crash at any moment now because of autopilot, that it has only killed drivers as of yet is a miracle ... first innocent is going to tank them, if it's a first responder in the US it could easily end up with Tesla management going to jail.

[Marco]

Training NN with a constant stream of real automatically annotated data is a pretty huge thing. nobody else has that. Like they said, it's the long tail of rare events that cause problems.  Look at the lady crossing the road carrying her bike that was killed. Having a NN that is trained well enough to deal with odd scenarios like that requires absolute torrents of real world data.

Pattern recognition only gets you so far, at some point a system has to start reasoning about and modelling the world ... a non recurrent neural network will never have enough training. We have no idea to train a recurrent network to be as good at problem solving as a dog, let alone have it drive a car.

[nctnico]

I think we are still at least 10 years away from fullly autonomous cars, possibly longer. Driving a car outside of carefully controlled conditions is just too complex of a task.

Driving a car is also a complex task for many people. However one of the problems to overcome is not just driving but also positioning. For driving in a city self driving cars need a much better way than GPS to get their position. Still there is also other technology on the way. The EU has standarised the method cars should use for vehicle to vechicle communication. For example: Volvo is going to release a system where cars warn eachother for the road conditions ahead. That is like looking around a corner. Seems very useful to me. On some highways there are relatively sharp bends in which you can't see what is ahead of you. I try to use the brake lights of other cars as an early warning system but that only goes so far.

[wilfred]

Elon, if you are reading, this is how we need self driving cars to perform:

On a racetrack perhaps. Why can't they develop a F1 car and enter it into the race? Win an F1 race, just one, and I might start to believe. It has to be easier than driving in traffic, for a machine.

[m98]

On a racetrack perhaps. Why can't they develop a F1 car and enter it into the race? Win an F1 race, just one, and I might start to believe. It has to be easier than driving in traffic, for a machine.

It would be relatively trivial, but won't be allowed for F1 and won't bring anyone enough money to do it. But there are some autonomous cars in Formula E:

[maginnovision]

They've sort of done autonomous race cars before in Fe. Last one I saw was drastically slower than a human but pretty good. I have made a scale one which isn't that fast but I know someone who made one that can go faster than either of us can drive.

[Rick Law]

I wish they would focus on making innovative high performance and practical EVs. I could not care less about self driving, the whole point of owning a personal car is that I get to drive it, when I don't feel like driving, I take the bus which is a much more efficient way of moving people.

I think we are still at least 10 years away from fullly autonomous cars, possibly longer. Driving a car outside of carefully controlled conditions is just too complex of a task. Computers are good at the mechanical aspect but way back when I took drivers ed I remember the instructor emphasizing that driving is primarily a social activity and computers still utterly fail when it comes to nuanced social interactions. It will be trivial for people to troll autonomous cars and exploit their predictable behavior and I suspect a lot of people will make a game of it.

I think there is some truth in driving being a social activity.  Understanding how people behave helps to avoid inter-human conflict.  Accidents at times are just such conflict that both want to be at that lane at that position at that time.

One particular day when I was teaching my kid how to drive, I saw something and told her: (distilled since I can't show how I was pointing with my hand) watch out - that yellow car on your right that just passed you, he is going to cut in front of the car ahead of you and then cut over again all the way to the far left lane.   About 15-30 seconds later, that yellow car did exactly that.  My kid was prepared and didn't slam into the car in front of her when it was braking hard to avoid the yellow car.

Based on the layout of cars around me, and based on how this yellow car driver behaved starting from how he was in my rear-view mirror, I knew exactly how he was going to act before his car shown any signs of doing it.  I think anyone with 10 years or so of driving experience would be able to do the same.  No different than most girls would know "that boy is going to hit on me" way before he tries to throw the first pass.

This types of road experience and behavior(s) of other drivers may be "handled" by self-driving cars in the future, but it is not there today.  It (in my opinion) is still only living in a world that small surprises will throw it into a big electro-pyschosis in its CPU brain.

As to "I wish they would focus on making innovative high performance and practical EVs."

While self-driving cars may be the future, for now, I think they need to focus on making a reliable car.

According to Consumers Report (which many in the USA who grew up before the internet would swear by) in October, Tesla-branded cars in general had dropped to third-worst in reliability, with only Cadillac and Volvo lower^[1].

Reference
[1] LA Times, Feb 2019 article: "Tesla Model 3 can no longer be recommended, says Consumer Reports"
^{https://www.latimes.com/business/autos/la-fi-hy-tesla-model3-cr-unreliable-20190221-story.html}

[MT]

Very interesting video , but i wonder how the software/system will handle full snow/sand/rain storm. Perhaps just stop
as some human do and wait for better whether or let the human take over, not much neural network calculus needed for such a decision. Another issue are AU road trains, whenever a jumpy jumps in front of one the train just runs it over
else the whole train will tumble over so Tesla neural system needs to implement a "kill a jumpy" code. How will the system detect and avoid potholes from shadows that look exactly like potholes or black ice conditions from oil/water
etc and perfectly dry roads, or just combinations of all and everything, pothole+snows storm+moose+drunk driver.

[apis]

Self driving cars don't have to be better than human drivers in every way, only when it comes to safety, in other respects they just have to be "good enough". If the cars need operator intervention every now and then it wouldn't be a big problem, as long as it doesn't happens too often. As long as self driving cars are sufficiently convenient and cost efficient it is going to be the preferred option.

Public transportation with bus isn't necessarily more efficient than an on-demand taxi service. During rush hours when busses are full they probably are more efficient, but when the busses are mostly empty, a taxi service would be more efficient.

Self driving cars don't have to be able to drive in snow yet, it just means there are places and times where they can't operate currently, but there are enough places without snow that it isn't going to slow down progress.

[james_s]

No it really is a big, BIG problem if they need human intervention EVER. It is a well documented fact that when something rarely requires our full attention the brain quickly finds other things to focus on and then when something does happen we have completely lost situational awareness. This is the reason that they have technology in trains that requires the operator to take some action on a regular basis to make sure they have not zoned out. A self driving car absolutely must be able to deal with all sorts of edge cases because things happen. Debris on the road, spills, unpredictable or erratic behavior by other drivers, confusing signs, a single incident where one of these cars gets confused and veers off the road into a crowd of pedestrians could easily put the company out of business under the weight of lawsuits and bad publicity. It doesn't matter if ultimately fewer total people are killed by the driverless cars, people don't focus on that.

Personally I'd love to see a lot less distracting tech in cars, and if it were up to me I'd make manual gearbox proficiency mandatory to get a driver's license. I'm quite convinced that it makes people more engaged and attentive drivers and I loathe slushboxes to the point that I wish they had never been invented, they're silly devices for lazy people. The more assistive tech a car has, the more that tech becomes a crutch and enables people to do all manner of things except for driving the damn car. I ride the bus to work and I see it many times every day, people cruising down the highway playing with their smartphone, eating, putting on makeup, while blasting down the highway in a 4,000+ lb machine.

[tox3]

Amazing how involved Elon is in technology

Personally I'd love to see a lot less distracting tech in cars, and if it were up to me I'd make manual gearbox proficiency mandatory to get a driver's license. I'm quite convinced that it makes people more engaged and attentive drivers and I loathe slushboxes to the point that I wish they had never been invented, they're silly devices for lazy people. The more assistive tech a car has, the more that tech becomes a crutch and enables people to do all manner of things except for driving the damn car. I ride the bus to work and I see it many times every day, people cruising down the highway playing with their smartphone, eating, putting on makeup, while blasting down the highway in a 4,000+ lb machine.

In highway driving there is not a large difference between manual and automatic transmissions.
But people are really stupid and does all kinds of stupid stuff behind the wheel.

[CatalinaWOW]

Lots of hubris here.  I learned on a manual transmission, and have owned several over the years.  But I far prefer automatic.  Two reasons.  First, I drive a car to get somewhere, not for the driving experience.  Having to sort around the gears all the time is a nuisance I can do without.  Particularly in big city rush hour traffic.  The second reason is just as important, though few will admit it.  I will admit that an expertly driven manual transmission can perform better than almost all, if not all automatic transmissions.  But I am not an expert, in spite of thousands of hours behind the wheel.  The few situations where I can do something better than the automatic are far outweighed by the many where the automatic outperforms me.  Particularly in difficult but not often encountered situations.  Like starting from a standing stop up a steep hill in heavy traffic.  Or towing a boat up a ramp.  (The latter two were particularly fun in a Mazda RX-3 with manual transmission.  No low RPM torque to speak of.  Wonderful car otherwise.)

Totally different area of discussion.  I agree that at least in the short term the solution for automatic cars in bad driving conditions will be to stop safely.  Unfortunately they will probably be prone to the same problem as humans who frequently misjudge their capability to drive in the current conditions.  I am a fairly good bad weather driver and cannot count the number of times I have seen folks spin out, slide off roads or other things due to their over estimation of their capabilities.  And I have a couple of instances where good fortune was all that save me from the same fate.

[Rick Law]

No it really is a big, BIG problem if they need human intervention EVER. It is a well documented fact that when something rarely requires our full attention the brain quickly finds other things to focus on and then when something does happen we have completely lost situational awareness.
...

I've to disagree with James_S here.  Ready or not, people will use it to excess anyway.  Dated even before "adaptive cruse control" hit the market, most of us who drive to work daily would have seen others driving while with "over-the-top" distractions - reading while driving, shaving while driving, putting on eye makeup while driving... you name it...  I have seen a lady knitting while driving...  A half-baked FSD would probably be safer than reading while driving.  As to whether it is more beneficial to society for those drivers to receive a Darwin Award sooner rather than later is another discussion.

As to Tesla, whether FSD is a big deal or not should not be a Tesla's worry -- The moment Consumers Report ranked the Tesla brand as third worst in reliability on the market (reference in my last reply = reply35 in this thread), their problem should no longer be FSD.  Started that moment, their problem should be reliability.  For them to worry about FSD right now is like worrying about the cookie crumbs on the floor while the living room is on fire.

[apis]

No it really is a big, BIG problem if they need human intervention EVER. It is a well documented fact that when something rarely requires our full attention the brain quickly finds other things to focus on and then when something does happen we have completely lost situational awareness. This is the reason that they have technology in trains that requires the operator to take some action on a regular basis to make sure they have not zoned out. A self driving car absolutely must be able to deal with all sorts of edge cases because things happen. Debris on the road, spills, unpredictable or erratic behavior by other drivers, confusing signs, a single incident where one of these cars gets confused and veers off the road into a crowd of pedestrians could easily put the company out of business under the weight of lawsuits and bad publicity. It doesn't matter if ultimately fewer total people are killed by the driver less cars, people don't focus on that.

You are right that there are many things the cars must be able to handle without operator intervention, and as I said, especially safety related issues. But if in one out of a thousand rides the cars stop at a blocked road and can't figure out a new route to the destination it's not a big problem. It would require the passenger to press a "call operator" button (or the car will do it automatically) to have someone manually reroute the car causing a delay of maybe 5-15 min. When it comes to safety I think people will expect self driving cars to be much safer than human drivers, if they regularly mow down pedestrians that will definitely be a show stopper. But I'm also convinced they have the potential to be much safer than human drivers and, unlike Tesla, Waymo have very impressive safety statistics (they are miles ahead of anyone else in the self driving cars race and they are already taking passengers in Phoenix I've read).

[thm_w]

As to Tesla, whether FSD is a big deal or not should not be a Tesla's worry -- The moment Consumers Report ranked the Tesla brand as third worst in reliability on the market (reference in my last reply = reply35 in this thread), their problem should no longer be FSD.  Started that moment, their problem should be reliability.  For them to worry about FSD right now is like worrying about the cookie crumbs on the floor while the living room is on fire.

If anyone is wondering: https://www.cnbc.com/2018/10/24/tesla-reliability-slips-to-third-worst-in-us-consumer-reports-says.html

BMW has had historically bad reliability, and the highest maintenance costs, and yet they do well. Probably because its a luxury/status brand, and people are willing to put up with it. Tesla is in a similar situation now, so I don't see it being as big of a deal as you do.

https://clark.com/cars/most-and-least-expensive-cars-maintain-repair/

[maginnovision]

As to Tesla, whether FSD is a big deal or not should not be a Tesla's worry -- The moment Consumers Report ranked the Tesla brand as third worst in reliability on the market (reference in my last reply = reply35 in this thread), their problem should no longer be FSD.  Started that moment, their problem should be reliability.  For them to worry about FSD right now is like worrying about the cookie crumbs on the floor while the living room is on fire.

If anyone is wondering: https://www.cnbc.com/2018/10/24/tesla-reliability-slips-to-third-worst-in-us-consumer-reports-says.html

BMW has had historically bad reliability, and the highest maintenance costs, and yet they do well. Probably because its a luxury/status brand, and people are willing to put up with it. Tesla is in a similar situation now, so I don't see it being as big of a deal as you do.

https://clark.com/cars/most-and-least-expensive-cars-maintain-repair/

I think the BMW thing is dependent on some things. BMW NA gives you 4 years maintenance for new cars, you can pay for 6 years. If you don't keep the car longer than that you never pay maintenance except tires. With older vehicles customers usually go somewhere else and use much cheaper aftermarket parts. I'm also not sure reliability is bad for BMW. Most cars have less than 1 or 2 repairs in a few years and the rest is maintenance. The last gen n63 had a few issues same as the n54 but even then they usually never required more than an extra 1 or 2(Extra set of injectors, maybe a recall or carbon cleaning) repairs depending on how the customer used the vehicle. Typically covered under warranty or goodwill repairs. I don't have worldwide numbers but the ones from one of the biggest dealers in the USA says they aren't bad at all.

[Rick Law]

As to Tesla, whether FSD is a big deal or not should not be a Tesla's worry -- The moment Consumers Report ranked the Tesla brand as third worst in reliability on the market (reference in my last reply = reply35 in this thread), their problem should no longer be FSD.  Started that moment, their problem should be reliability.  For them to worry about FSD right now is like worrying about the cookie crumbs on the floor while the living room is on fire.

If anyone is wondering: https://www.cnbc.com/2018/10/24/tesla-reliability-slips-to-third-worst-in-us-consumer-reports-says.html

BMW has had historically bad reliability, and the highest maintenance costs, and yet they do well. Probably because its a luxury/status brand, and people are willing to put up with it. Tesla is in a similar situation now, so I don't see it being as big of a deal as you do.

https://clark.com/cars/most-and-least-expensive-cars-maintain-repair/

I think it is a bigger problem for Tesla because:

(1) Outside of the aura of high tech and being an early EV, which was true when Tesla was new in the market, they have no other distinction.  Now every premium car maker are making EV's.  It is no longer so unique.  So they are completing in a crowded market that is getting even more crowded.

(2) USA is their largest market but USA government subsidies for them is entering phased out stage (they reached the numbers where phase out begins this year), they will not be as price competitive as they once were - and this gap will grow since phase out means subsidies per car will decrease.  In 2018, it was $7500/car; 1H2019 (first half 2019) will drop to $3750 and 2H2019 drops to $1875.  I think 2020 is zero.  This implies they will have further price/cost pressure and they don't have much room to squeeze when their quality is already so stressed.

In Denmark (source = zerohedge.com)^[1], in 2016 when subsidies phased out for EV, Tesla sales drop 94%!  So this year is a very bad year money-wise even if their operation is running smooth.  For them to need to spend money on fixing manufacturing quality problem(s) spells big danger; perhaps existential danger.

May be I would be proven wrong and they can handle the price-pressure with lower than average quality.  Their being third from bottom just doesn't give me a lot of confident...

Reference:
[1] zerohedge.com article June 2017:"It's Confirmed: Without Government Subsidies, Tesla Sales Implode"
"Nobody was hurt more than Tesla: the company, whose sales were skyrocketing at the time, lobbied against the move, with CEO Musk warning during a visit to Copenhagen that sales would be hit. It wasn't clear if the warning was targeting the government, the people of Denmark, or his own bank account and shareholders, but he was absolutely correct: in 2015 Tesla sold a total of 2,738 cars in Denmark. In 2016 the number dropped by 94% to just 176 units."
Full article here:
^{https://www.zerohedge.com/news/2017-06-11/its-confirmed-without-government-subsidies-tesla-sales-implode}

[SilverSolder]

The situation in Denmark is a little special though, they have crazy high car sales taxes there (something like 180%) so any discount really matters!   I recall seeing a Mercedes in Copenhagen with a personalised number plate:  "PAID 3X"

[nctnico]

@Rick Law: Tesla also receives money from other car manufacturers to compensate for their lack of selling electric vehicles. This source of income may also get reduced over time. If you own Tesla shares then now may be a good moment to sell.

[wraper]

@Rick Law: Tesla also receives money from other car manufacturers to compensate for their lack of selling electric vehicles. This source of income may also get reduced over time. If you own Tesla shares then now may be a good moment to sell.

Have you not heard about new factory in China?

[nctnico]

@Rick Law: Tesla also receives money from other car manufacturers to compensate for their lack of selling electric vehicles. This source of income may also get reduced over time. If you own Tesla shares then now may be a good moment to sell.

Have you not heard about new factory in China?

Where they will have to compete with even better and more producers of electric cars. Over here we have a few electric busses driving around in the city. Made in China.

I've read an interesting article in one of the bigger newspapers of the NL. China is becoming a real power where it comes to automotive engineering for both electric and self driving cars. They have the biggest uniform market for cars in the world.

[Rick Law]

The situation in Denmark is a little special though, they have crazy high car sales taxes there (something like 180%) so any discount really matters!   I recall seeing a Mercedes in Copenhagen with a personalised number plate:  "PAID 3X"

Yeah, but the same thing happened in Hong Kong - Tesla sales went from 2000 in 2016 down to 32 in 2017^[1].

China, Denmark, anywhere else, Tesla as a maker of only EV cars needs to keenly aware that they are not yet price-competitive to non-EV cars.  They have to live in the two corners - (1) richer folks and (2) government assistance be it by regulation or financial assistance.

Corner number two (subsidies) is getting smaller.  So for EV-only makers, they will be increasingly depend on corner 1 - the richer folks.  For the richer folks where money matters less, quality (or the aura of it) is important.

Unfortunately, Model 3 is not a good thing for corner-1.  Can you imagine an "entry level Rolls Royce" for the average folks?  Mercedes tried that with the 190's in the 1980s.  It didn't go over too well.  I think that is the reason why the BMW 1 series is now history in the USA.  Toyota/Honda (and others) understood that, they started Acura and Lexus.  Instead of diluting the brand,  Toyotas are for the penny-counters and Lexus are well, for those with (or think they have) more money to spend.  Those in corner 1 really would prefer to be separated from the average folks - like first class in a plane with a curtain separating it from economy.

I think, Tesla needs to keep their Teslas with an aura around it - all the ooh's and aah's.  People will pay extra for that.  Start another brand for model 3 and sell it in a separate facility.  But they need money to carry that out, and money isn't what they have in excess...  Elon and his team better figure this out and get themselves on better footing...

Oh, regarding China - given China is a new rich country, that separation would be particularly important.  My life-experience tell me, new rich are more status conscious and would want the distinction more.  I could be wrong, and China's new rich would not be as status conscious as I anticipate - but if I am right and they try to sell Teslas as primium car but also Model 3 (or less) that is cheap (cheaper) in the same shop, that would be a disaster in China.

Reference:
[1] Financial Times magazine, 2018, February: "Tesla sales drop in Hong Kong after tax breaks removed"
"Touted as the city with the highest number of Teslas per capita by Mr Musk in early 2016, just 32 of the company’s vehicles were registered in Hong Kong between April and December, figures from the Transport Department showed. That is down from nearly 2,000 in the same period in 2016."
^{https://www.ft.com/content/2b8eb480-0a45-11e8-839d-41ca06376bf2}

[nctnico]

In the Netherlands the tax breaks on EVs are partly removed (no tax breaks on EVs over 50k euro). Sales of the Tesla Model-S and Model-X have dropped to insignificant numbers. The only model selling in reasonable numbers is the much cheaper Model 3.

[Rick Law]

In the Netherlands the tax breaks on EVs are partly removed (no tax breaks on EVs over 50k euro). Sales of the Tesla Model-S and Model-X have dropped to insignificant numbers. The only model selling in reasonable numbers is the much cheaper Model 3.

Looks like the are loosing the "premium image"...    Dinner plate please!  Their goose is cooked...

[maginnovision]

It's a premium image that people enjoyed getting a break on. Now it's an expensive vehicle with smaller breaks which keep shrinking and with every one sold makes it less exclusive. Time will tell where this goes.

[extide]

Mercedes tried that with the 190's in the 1980s.  It didn't go over too well. 

FWIW, The 190 series became the C class, which is not only the most popular model that Mercedes sells in the US, it's the most popular model out of all luxury brands.

[TerraHertz]

I find the topic of self driving cars fascinating. Not so much for the technology, but what it reveals about people's not-so-rational dreams and expectations.  Much the same as the apparent yearning for 'AI servants and industrial production', but with the added dimension of staking your life on the software controlling a high speed vehicle.

A few points:
* Any highly automated system with some kind of always-on net connectivity will be susceptible to remote hijacking. Don't even bother to talk about computer security, in a world where (for eg) hardware level backdoors are built into all contemporary Intel CPUs, all network routers, all cell phones, etc. Ask Michael Hastings about the pros and cons of remote vehicle operation.

* All complex software systems are prone to obscure bugs and design mistakes. Some are more stupid than others (eg Boeing's 'please fly us into the ground right now' brilliant idea.)

* The nature of the problem prevents 'perfection.' Road driving is a highly complex task, with infinite numbers of obscure marginal and tricky cases. Even alert, skilled humans get fooled sometimes and end up in accidents. Some of which are high speed and fatal. No level of AI is going to completely eliminate all situational f*ck-ups.  Self-driving cars ultimately involve a statistical risk evaluation: What added probability of death or maiming do you accept for the 'convenience' of not being in control of the vehicle yourself?

* What's the actual benefit of not being in control? What kind of person could actually nap, or concentrate on some productive task, while being carried at speed by an AI system? I definitely never could.  Idle conversation, maybe. But that's not something I'd choose to even want, let alone try.   I can relax in trains, because that infrastructure is very good at achieving an extremely low risk. Planes, because again quite low risk, and skilled humans are in charge (unless a new Boeing...) Cars on ordinary roads - never!

* At a political level, there are too many ideologues trying to impose restrictions on individual travel, and also track such travel. Taxing travel at a rate per Km, requires logging of travel. Automated cars can be legislated to support that. Tracking for social control, ditto. Schemes like China's 'social credit', also would find automated cars useful.  How about governments mandating things like rationed travel, selected days when you are allowed to travel, zones you are not allowed to enter, and so on? Automated cars enable all that kind of bullshit.

* Privacy. Given that Google, Amazon, etc 'home assistants' are being revealed to upload a lot of information about conversations in the home (and commonly making comical interpretation mistakes), do you want that in cars as well?

* Legal liability. I can't wait till this issue becomes prominent through increasing numbers of accidents in which no human adult was responsible.  I wonder if legal principles for dealing with AI in general, will be founded in road accident case precedent? That will go well, not.

Then there are the fundamental philosophical issues with increasingly general purpose AI systems (which is where self-driving cars will be forced to go, by virtue of the complexity of the problem.) There are many questions here. For instance, how wide is the window of workability, in which the car is smart enough to get the job done, but not smart enough to get in a snit about something you said to it, and refuse to do the job? (Or decide to suicide, with you as passenger.)

Meanwhile, I'm keeping my old, pre-engine-management-computer car.  Self-driving car technology is good for one thing - entertainment, in watching a fairly predictable developing social insanity.

[apis]

but with the added dimension of staking your life on the software controlling a high speed vehicle.

Why trust a car at all, it's a pile of complicated software and hardware. Most people stake their life on incredible complex hardware and software every day.

Even alert, skilled humans get fooled sometimes and end up in accidents.

No one thinks self driving cars will be perfect, but some people realise that neither are humans. In fact humans are **** poor drivers and that's the main reason why replacing them with a computer is not such a bad idea. Even the Tesla "autopilot" cruise control is claimed to already have much better accident statistics than human drivers.

It's safe to be a naysayer regarding this because it won't become widespread overnight, might take decades, but self driving cars are no doubt the future, mainly because it will be cheaper than human drivers and money rules.

I understand that some people might enjoy driving and not like the idea of a future where the average man might not have the option to drive, but other than that I don't really understand why anyone would be against robotic cars or other forms of automation. Just seems like ludditis to me.

[apis]

Waymo disengagement rate for 2018 was 1 disengage per 11,017 miles out of 1.28 million self-driven miles in CA 2018. That corresponds to about one disengage per year for an average driver in the US.

"By the end of 2018 we’d driven another six million miles, which means our self-driving cars have now covered 10+ million real-world road miles. The connection between our real-world miles coupled with our 7+ billion miles in our simulation is key to our improvement rate."

https://medium.com/waymo/an-update-on-waymo-disengagements-in-california-d671fd31c3e2



"At one disengagement per 11,017 miles, Waymo’s closest competitor is GM-owned Cruise Automation at 5,205 miles. Apple’s effort is last at a disengagement every 1.1 miles, though the company disagrees with the metric."

 

[nctnico]

* The nature of the problem prevents 'perfection.' Road driving is a highly complex task, with infinite numbers of obscure marginal and tricky cases. Even alert, skilled humans get fooled sometimes and end up in accidents. Some of which are high speed and fatal. No level of AI is going to completely eliminate all situational f*ck-ups.  Self-driving cars ultimately involve a statistical risk evaluation: What added probability of death or maiming do you accept for the 'convenience' of not being in control of the vehicle yourself?

Now you are assuming self driving cars will always be worse than human drivers. Automatic anti-skid systems (which exist for a long time) are already proving you wrong.

[Rick Law]

Mercedes tried that with the 190's in the 1980s.  It didn't go over too well. 

FWIW, The 190 series became the C class, which is not only the most popular model that Mercedes sells in the US, it's the most popular model out of all luxury brands.

You have a point there.  I am thinking how snob-appeal may affect sales to the rich clientele.  I don't think anyone will dispute that it has an impact, but it would be merely one of the many variables that would affect the financials.  No doubt some of the S class buyers would rather not the C class folks clogging up the shop, but when you add up all the dollars the seller/manufacturer make, the case is not as clear whether it is more profitable to focus on high end.  Your point is absolutely right and I have to adjust my thinking here.

Walking myself back from snob-appeal and related, the issue of quality remains.

I still stand by my point that for Tesla FSD is not as important as interior trim falling off new cars or wind-shield cracking by itself.  That two were both cited by Consumers Report as quality issues.

When the car is falling apart by itself, mean while, Tesla is hoping I would trust my life to that car auto-driving me to somewhere...  That I think is silly.

[wilfred]

Waymo disengagement rate for 2018 was 1 disengage per 11,017 miles out of 1.28 million self-driven miles in CA 2018. That corresponds to about one disengage per year for an average driver in the US.

Anything above zero means you need a steering wheel and an attentive driver. No steering wheel or no automation. A hybrid mix is the worst of all possibilities. It means you can't trust the automation or you can't blame the driver.

The family of someone struck by an auto needs to seek a just outcome and lawyers at 20 paces with a large corporation is going to be an issue. You can't treat it the same a a plane crash due to a firmware error. You can't take the cars off the road you can't spend millions on a crash investigation.

The only solution I can see is to have each vehicle carry a designated driver who will accept the legal responsibility for the firmware choices. Perhaps they insert a card to load ethical parameters to customise the firmwares programming. Which obviously means you can't set the car to circle the block whilst you pick up your dry cleaning. Nor can you use it to freight children to school or Saturday morning sports. Nor can you get dropped off at the door and have the car go and find a parking spot.

If autos are the solution then I don't think the problem is well understood. Buses and trains will solve congestion better. Clustering people closer to work and community facilities will work better. Children will be better off walking to school close to home or socialising with friends on a bus or train.

[apis]

Waymo disengagement rate for 2018 was 1 disengage per 11,017 miles out of 1.28 million self-driven miles in CA 2018. That corresponds to about one disengage per year for an average driver in the US.

Anything above zero means you need a steering wheel and an attentive driver. No steering wheel or no automation. A hybrid mix is the worst of all possibilities. It means you can't trust the automation or you can't blame the driver.

When it comes to anything safety-critical it would have to be zero (it will never be exactly zero but close and what is most important: better than a human driver), no argument there.

For the most part safety means don't crash into something or drive off the road, which you can do deterministically quite well using good sensors (lidar), good maps and traditional control theory. That's what Waymo is doing. Tesla has taken a very different approach, they use only cheap sensors (cameras) and then use machine learning algorithms to convert the video feed into a 3D representation of the world. Such machine learning based computer vision systems have been making really amazing progress the last decades. But as we have seen, the Tesla system sometimes makes fatal mistakes. The old school, but more expensive, Waymo approach is much more robust. Waymo using lidar means they don't have to convert a 2d representation into 3d, they get a very accurate 3d representation directly from the lidar sensor (and on top of that they add sensor information from cameras and radar).

Tracking objects and steering the car to avoid objects (short term) is a solved problem as long as you have an accurate 3d representation of the world. The most difficult parts are things like trying to predict other drivers behaviour, pedestrians intentions and things like that.

However, there are many corner cases that aren't safety critical but can still be hard for a car to deal with, e.g. because getting out of the situation requires technically breaking the law (driving somewhere you're not allowed to driver normally when the normal road is blocked). Would it be legal to program the car to break the law? But It's not so critical if the car can't find the route because the road is blocked, it can just drive somewhere where it can stop safely and ask the passenger or a tele-operator for help. Maybe there is an obstacle on the road it's not sure how to deal with (there was an event with a mattress on the road the self driving cars stopped for while human drivers just drove over it.) The cars will be programmed to be overly cautious and it might be annoying if they stop for something they don't have to stop for, but it won't be a major problem as long as it doesn't happen too often. I imagine these kind of minor problems could be solved in 5-15 minutes and if it only happens once every year then it won't be so bad. And one would expect the number of disengagements to keep getting lower as they improve the software and sensors.

The family of someone struck by an auto needs to seek a just outcome and lawyers at 20 paces with a large corporation is going to be an issue. You can't treat it the same a a plane crash due to a firmware error. You can't take the cars off the road you can't spend millions on a crash investigation.

The only solution I can see is to have each vehicle carry a designated driver who will accept the legal responsibility for the firmware choices. Perhaps they insert a card to load ethical parameters to customise the firmwares programming. Which obviously means you can't set the car to circle the block whilst you pick up your dry cleaning. Nor can you use it to freight children to school or Saturday morning sports. Nor can you get dropped off at the door and have the car go and find a parking spot.

I think we will mainly see taxi services for quite some time, not privately owned self driving cars. If one of the cars cause a problem it would be the taxi company who is responsible (and they in turn would hold the manufacturers of the cars responsible).

You never need to worry about parking, a car picks you up when you need it and drops you off at your destination. It will be like taxi, but without the driver.

If autos are the solution then I don't think the problem is well understood. Buses and trains will solve congestion better. Clustering people closer to work and community facilities will work better. Children will be better off walking to school close to home or socialising with friends on a bus or train.

Busses and trains can also be made self driving. Busses and trains are great as long as they are filled with people. But busses and trains also drive around almost empty for a large part of the day. Proponents of PRT (podcars) have always claimed that on demand services with smaller cars are more efficient than traditional mass transit systems. The problem with cars today is that everyone wants their own car, so we have one person driving around with 3 seats empty (or more) and most of the time the car is parked somewhere. We normally don't use taxi today because it's too expensive to pay a human chauffeur, but if you don't need a human driver that changes.

[apis]

For the most part safety means don't crash into something or drive off the road, which you can do deterministically quite well using good sensors (lidar), good maps and traditional control theory. That's what Waymo is doing. Tesla has taken a very different approach, they use only cheap sensors (cameras) and then use machine learning algorithms to convert the video feed into a 3D representation of the world. Such machine learning based computer vision systems have been making really amazing progress the last decades. But as we have seen, the Tesla system sometimes makes fatal mistakes. The old school, but more expensive, Waymo approach is much more robust. Waymo using lidar means they don't have to convert a 2d representation into 3d, they get a very accurate 3d representation directly from the lidar sensor (and on top of that they add sensor information from cameras and radar).

I might add that Waymo also uses machine learning, and Tesla has radar sensors as well as cameras (but no expensive LIDAR). But from what I know there is a large difference in their approach and Tesla rely much more on machine learning.

In the last question of this talk the Waymo engineer also mentions they use a hybrid approach, only using machine learning when it improves safety (even if they have to use the more expensive sensors).

"You want to be safe in the environment, so you don't want to make errors in perception, prediction and planning. And the state of machine learning is not at the point where it never makes errors."

https://youtu.be/Q0nGo2-y0xY?t=3774

[Marco]

The problem with radar is that for the moment it's just doppler for a huge FOV, not ToF distance based imaging... which makes it almost completely useless.

[maginnovision]

I'm pretty sure ti has mmwave that can give you size, speed, and direction of travel for detected objects. Seems ok to me. Of course all the inputs are fed into a filter to determine what's most likely happening you never accept everything at face value. Except for Tesla, they seem like they're ML or nothing.

[wraper]

For the most part safety means don't crash into something or drive off the road, which you can do deterministically quite well using good sensors (lidar), good maps and traditional control theory. That's what Waymo is doing. Tesla has taken a very different approach, they use only cheap sensors (cameras) and then use machine learning algorithms to convert the video feed into a 3D representation of the world. Such machine learning based computer vision systems have been making really amazing progress the last decades. But as we have seen, the Tesla system sometimes makes fatal mistakes. The old school, but more expensive, Waymo approach is much more robust. Waymo using lidar means they don't have to convert a 2d representation into 3d, they get a very accurate 3d representation directly from the lidar sensor (and on top of that they add sensor information from cameras and radar).

I might add that Waymo also uses machine learning, and Tesla has radar sensors as well as cameras (but no expensive LIDAR). But from what I know there is a large difference in their approach and Tesla rely much more on machine learning.

In the last question of this talk the Waymo engineer also mentions they use a hybrid approach, only using machine learning when it improves safety (even if they have to use the more expensive sensors).

"You want to be safe in the environment, so you don't want to make errors in perception, prediction and planning. And the state of machine learning is not at the point where it never makes errors."

https://youtu.be/Q0nGo2-y0xY?t=3774

Waymo approach is garbage frankly. They may have a very small number of disengagements when driving in areas with their high detail maps. But when on a large scale, it's a dead end approach. You will never have precise high detail maps for everything, not to say without errors. Also changes happen all of the time, keeping it up to date is a huge challenge on it's own. It's simply unfeasible to make up to date high detail maps for every area. Not to say when it comes to bad weather, LIDAR becomes an erratic piece of garbage.

[G7PSK]

I am begining to think that all Tesla related material should be moved to the Dodgy technology page.

https://youtu.be/sAQlLu5ttOk

[wraper]

I am begining to think that all Tesla related material should be moved to the Dodgy technology page.

Not again  . Just look statistics on how many cars catch fire and think again if you should have posted this. Not to say it's off topic.
https://www.google.com/search?biw=2048&bih=1166&tbm=isch&sa=1&ei=HavKXPzgK4aFmwXs0pG4DQ&q=car+fire+garage&oq=car+fire+garage&gs_l=img.3..0i24.17775.19340..19545...0.0..0.58.341.7......1....1..gws-wiz-img.......0j0i67j0i5i30j0i8i30.DwDjUrYwVWE

[Marco]

I'm pretty sure ti has mmwave that can give you size, speed, and direction of travel for detected objects.

It's possible, it's just not currently on the cars ... or it wouldn't be running into fire trucks.

[janoc]

I will just post link to this here:

https://jalopnik.com/automation-transformed-how-pilots-fly-planes-now-the-s-1834176244

It specifically discusses Tesla's approach to this and why it is a huge problem. There are decades of research on the impact of human factors when dealing with automation (e.g. in aviation) and, frankly, I am shocked how clueless some people are about this, despite this being an engineering forum.

Talking about "disengagement rates" - hello, when the car drives you into wall with a fully engaged automation because it got confused, does that count? Or when it drops the mess in your lap with 3 seconds to react - you will be dead before you realize what is happening, unlike in a plane where you usually have minutes and kilometers of empty air ... Talking how it is safer than human drivers - with only data for this being the oft-quoted claim from Musk using data from Autopilot which is supposed to be used only on highways - where the least number of crashes occurs, etc.

Or someone making the argument that because antiskid systems are better at preventing loss of control than humans are, it somehow follows that automatic cars will be better too (never mind the several orders of magnitude difference in the complexity of the problem ...).

This debate is pretty much the same every time Tesla is mentioned. The fanboys come out of the woodwork and will defend every nonsense that caused people to die in the past only because now it comes from Tesla, so laws of physics and human psychology somehow don't apply anymore.

There is also this:
https://techcrunch.com/2019/05/01/tesla-sued-in-wrongful-death-lawsuit-that-alleges-autopilot-caused-crash/

Yes, that wasn't a fully autonomous car and yes, the driver will most likely be found at least partially at fault - but if Tesla loses this, it will be pretty much game over for both their Autopilot and most likely the fully self driving cars in their current iteration as well. Every lawyer will jump on every single minor scrape hoping to sue bejeesus out of these companies. And they will have only themselves to blame because, as the plaintiffs wrote in the suit, "Tesla is beta testing its Autopilot software on live drivers".

I have nothing against Musk (e.g. what he achieved with SpaceX is amazing!) or Tesla but if we keep the debate to this style where every criticism is brushed away with "I don't hear you, go away, Luddite!", sticking head in the sand and/or mindlessly repeating hype about machine learning as some sort of magic mantra that will fix everything just give it enough data, people will die - in completely avoidable and preventable accidents.

[madires]

Another issue most people are not aware of is that the driver assistance systems and systems for autonomous driving can be fooled easily. This was demonstrated several times by security researches at hacker cons. There are a lot problems which need to be addressed before we may think about trusting a self-driving car.

[apis]

Waymo approach is garbage frankly. They may have a very small number of disengagements when driving in areas with their high detail maps. But when on a large scale, it's a dead end approach. You will never have precise high detail maps for everything, not to say without errors. Also changes happen all of the time, keeping it up to date is a huge challenge on it's own. It's simply unfeasible to make up to date high detail maps for every area. Not to say when it comes to bad weather, LIDAR becomes an erratic piece of garbage.

Priority one is no accidents (Waymo has no fatalities, Tesla 3).
Priority two is comfort.
Priority three is that there no annoying disengagements that cause delays. Waymo has the lowest number of disengagements per mile by a large margin. Tesla is still on the "cruise control" stage.

Google has already demonstrated they are able to map the parts of the world where they would want to drive in the foreseeable future, so the need for maps is not much of a limitation. Using maps is safer so why not use them if you can? Clearly they are able to deal with some changes in the map environment or else they wouldn't be able to drive as much as they do.

All sensors have both advantages and disadvantages, but lidar is without a doubt the best type of sensor you can have. Waymo drives in light rain at the very least. If they can't use the lidar in some extreme conditions they will be in the same bad situation as Tesla, having to rely on only cameras and radar. It's not like the lidar is the only sensor. Lidar + camera + radar > camera + radar, no matter how you look at it.

The most difficult part for a self driving car is to take the sensor data and create an accurate representation of the surroundings. That is also one of the most safety critical tasks (you don't want to miss any pedestrians on the road like Uber did, or road dividers or trucks, etc, like Tesla did). So it should be obvious that using the best possible combination of sensors to minimise the risk of confusion is the way to go.

Musk has said the following about lidar apparently:
“In cars, it’s freaking stupid. It’s expensive and unnecessary. And as Andrej was saying, once you solve vision, it’s worthless. So you have expensive hardware that is worthless on the car.”
Basically he's admitting that computer vision isn't good enough yet and that he is betting on that they will solve vision before everyone else.

Tesla has convinced people to collect the road data for them and people even pay a premium to be their safety drivers. If something goes wrong Tesla just blames it on the driver not being alert enough: "it's only cruise control". It's clever marketing if nothing else, but it's killing people. They get a lot of data for free. Machine learning on it's own isn't good enough yet, but Tesla is betting that they will get to good enough before the competition, and by then they will be sitting on more data and hardware with lower marginal cost.

But there is no reason to think other companies like Waymo won't be able to remove the lidar whenever it's safe to do so. It's not like Waymo and others are unable to use more machine learning and less expensive sensors whenever they choose to.

[G7PSK]

Might well be the wrong place,probabley the wrong page as well. There are 272.1 million cars on the road in the US last year of which 171,500 caught fire Tesla has built 300,00 cars and from what I can find online there are at least 2 fires a week relating to them. I will leave the math.

[wraper]

Google has already demonstrated they are able to map the parts of the world where they would want to drive in the foreseeable future, so the need for maps is not much of a limitation. Using maps is safer so why not use them if you can? Clearly they are able to deal with some changes in the map environment or else they wouldn't be able to drive as much as they do.

Yow fresh they generally are? 1 year, 2 years? I've seen plenty of situations when map came out yesterday, but does not match actual road which was rebuilt differently a few months ago. Not to say, it's not just satellite view that is required in this case.

[apis]

I have nothing against Musk (e.g. what he achieved with SpaceX is amazing!) or Tesla but if we keep the debate to this style where every criticism is brushed away with "I don't hear you, go away, Luddite!", sticking head in the sand and/or mindlessly repeating hype about machine learning as some sort of magic mantra that will fix everything just give it enough data, people will die - in completely avoidable and preventable accidents.

The main problem I see is a lot of people who say it's impossible but they don't have an informed explanation of why they believe so or any data to back it up. It's just heated opinions and there seems to be very poor understanding of how these cars work. People on the internet say it's not possible and in the meantime these cars are driving tens of thousands of miles uninterrupted in the real world, apparently unaware it's not possible. Seems pretty disingenuous to me.

I agree there's reason to be critical but it's also annoying that people assume all self driving cars are equal. Tesla is way behind the others in this field. If you want to assess the technology you should look at the technology leader which is Waymo not Tesla.

Mostly the criticism is on the lever "it's impossible, there are too many nuances and edge cases". Well, obviously not, they are already at a level where they can drive completely uninterrupted in California for what corresponds to a year for an average person in the US. (With zero fatalities of course). They've made the cars overcautious to guarantee no accidents, but that means there will be false positives (disengagements), so the difficult part is to get those down to an acceptable level without increasing the risk of serious accidents.

Self driving cars won't take over by 2019 as some (Elon) says, but there's nothing to indicate they won't get there eventually imo.

[wraper]

Might well be the wrong place,probabley the wrong page as well. There are 272.1 million cars on the road in the US last year of which 171,500 caught fire Tesla has built 300,00 cars and from what I can find online there are at least 2 fires a week relating to them. I will leave the math.

There were 2 fires in one week not too long time ago. But 2 fires a week is not true at all. Usually there are no fires for several months in a row, then each fire is a big news. No publicly available info of model 3 catching fire so far.

[apis]

Google has already demonstrated they are able to map the parts of the world where they would want to drive in the foreseeable future, so the need for maps is not much of a limitation. Using maps is safer so why not use them if you can? Clearly they are able to deal with some changes in the map environment or else they wouldn't be able to drive as much as they do.

Yow fresh they generally are? 1 year, 2 years? I've seen plenty of situations when map came out yesterday, but does not match actual road which was rebuilt differently a few months ago. Not to say, it's not just satellite view that is required in this case.

It's impossible to discuss this when the data isn't public. The details are still treated as secret sauce so we will basically just have to wait and see. But I would be amazed if they didn't think about this very obvious problem long before Google (and then others) invested billions in the technology, and the engineers and computer scientists who have created this technology aren't idiots either.

(The real world changes continuously so the map is outdated immediately when it's made in some sense, so the cars can obviously deal with some amount of change.)

[extide]

There were 2 fires in one week not too long time ago. But 2 fires a week is not true at all. Usually there are no fires for several months in a row, then each fire is a big news. No publicly available info of model 3 catching fire so far.

Funnily enough, even if they were catching on fire 2x a week, it would still be way below average.

According to his numbers, the national average is 1 fire/year per 1587 cars, and Tesla is at 1 fire/year per 2885 cars. (For Tesla, that's calculating 2 fires/week or 104 fires/year, which is definitely more than reality.)

[wraper]

Funnily enough, even if they were catching on fire 2x a week, it would still be way below average.

According to his numbers, the national average is 1 fire/year per 1587 cars, and Tesla is at 1 fire/year per 2885 cars. (For Tesla, that's calculating 2 fires/week or 104 fires/year, which is definitely more than reality.)

According to this it's more like 1 fire in 4 months in average https://www.autoblog.com/2018/05/11/a-list-of-tesla-car-fires-since-2013/?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAIAWmKnMzLZc1kifeclBxdibttYoe8QYTp9tZNQIGQYz1ien_c4gZdjnv8hd9WyUhQFvCGY1gLsz8dyN_OpXQ7S8X3CFHw-XSrn8trVVabPIBm3Ur70nMsGj3gd1D1UB7B---pxR4mX9fo9luFDUhmDENShOPflotJSmf4PE40wA
BTW Tesla made 350k cars in 2018 alone, 100k in 2017, 84k in 2016.

[nctnico]

Google has already demonstrated they are able to map the parts of the world where they would want to drive in the foreseeable future, so the need for maps is not much of a limitation. Using maps is safer so why not use them if you can? Clearly they are able to deal with some changes in the map environment or else they wouldn't be able to drive as much as they do.

Yow fresh they generally are? 1 year, 2 years? I've seen plenty of situations when map came out yesterday, but does not match actual road which was rebuilt differently a few months ago. Not to say, it's not just satellite view that is required in this case.

But how much of the total amount of roads is that? 0.0001%? You also have to understand that the map data + location is used to augment what the sensors bring in.

IMHO it is pretty useless to discuss all kinds of edge cases. People also drive off the road for many different reasons. Why should a self driving car be error free? In the end what counts is that less people get injured or killed in car accidents and that mobility improves.

[wraper]

Google has already demonstrated they are able to map the parts of the world where they would want to drive in the foreseeable future, so the need for maps is not much of a limitation. Using maps is safer so why not use them if you can? Clearly they are able to deal with some changes in the map environment or else they wouldn't be able to drive as much as they do.

Yow fresh they generally are? 1 year, 2 years? I've seen plenty of situations when map came out yesterday, but does not match actual road which was rebuilt differently a few months ago. Not to say, it's not just satellite view that is required in this case.

But how much of the total amount of roads is that? 0.0001%? You also have to understand that the map data + location is used to augment what the sensors bring in.

IMHO it is pretty useless to discuss all kinds of edge cases. People also drive off the road for many different reasons. Why should a self driving car be error free? In the end what counts is that less people get injured or killed in car accidents and that mobility improves.

I don't say it should be error free but relying on precision maps is a kind of dead end approach. You can get stellar results when in perfect environment but when it comes to real life, it will kinda suck. They report low number of disengagements but IMO it's a misleading number since they drive in thoroughly investigated areas.

But how much of the total amount of roads is that? 0.0001%

Like 1k-10k times more than that. If it was 0.0001% I doubt you would ever stumble on that in your life, yet I've seen it several times in one 300 km trip.

IMHO it is pretty useless to discuss all kinds of edge cases.

Edge cases is the most important part since it's what breaks proper operation.

[nctnico]

But how does a changed road suddenly becomes an edge case? Why would that be? Also I never wrote that a system has to rely on precision maps however precises maps are a good start to find out which road a car is driving on. Current navigation systems can easely get confused. Accurate maps and cm precision (terrestrial) positioning will help to let a self driving car know where it actually is.

[wilfred]

IMHO it is pretty useless to discuss all kinds of edge cases. People also drive off the road for many different reasons. Why should a self driving car be error free? In the end what counts is that less people get injured or killed in car accidents and that mobility improves.

An auto should be error free because when it fails it is necessary to look at who, how and why it had the error and if necessary look to seek justice for the victims (if any) of the error. I think it will be good if fewer people die or get injured overall. But since that doesn't matter to the victim or their family and friends who will want to hold someone responsible it can't work to justify the introduction of autos.

If people don't trust and feel safe in or near autos then they will not succeed. People are shocked when someone mistakenly presses the accelerator instead of the brake and drives into a crowd waiting for a bus or whatever. However, because they are also human and fallible they can and have to accept that such things will happen. There will be an investigation to find out what factors led to the incident like alchohol or drugs, mobile phone distraction and so on.

Those factors may not be what causes an auto to fail but people will also have a harder time getting their heads around what a firmware error or sensor failure is and how it might impact them in their daily interactions with an auto. People just will not accept a corporation (or a government) doing the math and putting profit ahead of lives, their lives.

The real problem I see with autos is not getting them to drive from A to B but dealing with the issues when they don't.

[Rick Law]

Well...  We shall see if he is right:

CNBC, May 2, 2019: "Elon Musk to investors: Self-driving will make Tesla a $500 billion company"
"... ... Musk confidently told investors on the call that autonomous driving will transform Tesla into a company with a $500 billion market cap, these people said. Its current market cap stands around $42 billion. He also said that existing Teslas will increase in value as self-driving capabilities are added via software, and will be worth up to $250,000 within three years. ... ..."

Full article:
^{https://www.cnbc.com/2019/05/02/elon-musk-on-investor-call-autonomy-will-make-tesla-a-500b-company.html}

[madires]

Emphasis on "up to". It could be also just an upgrade package for US$10k since all car manufacturers will offer self-driving capabilities in the future.

[apis]

I don't say it should be error free but relying on precision maps is a kind of dead end approach. You can get stellar results when in perfect environment but when it comes to real life, it will kinda suck. They report low number of disengagements but IMO it's a misleading number since they drive in thoroughly investigated areas.

Thoroughly investigated as in having detailed maps? All companies including Tesla (as far as I know) intend to launch taxi services in limited areas in the beginning. I have no doubt that Alphabet has the ability to provide detailed maps of whatever dense cities and interconnecting highways where they wan't to launch first. If outdated maps will be a problem it will be in remote areas so it won't be an issue anytime soon.

(While Waymo's cars can drive anywhere, including in complex city traffic, Tesla autopilot only works on highways, which is the easiest type of road.)

I believe Waymo use the maps to guarantee the car knows where it is on the road within a few cm, even without GPS, and also to help discern dynamic objects (the car needs to pay closer attention too) from static objects. That means they can dedicate more processing power to other problems, and the cars are more or less guaranteed to not confuse the road with an object on the road (as we have seen both Tesla and Uber had problems with sadly).

Waymo's cars have all the sensors that a Tesla car does, and on top of that lidar and better radar. Waymo make their own sensors in house.
https://medium.com/waymo/introducing-waymos-suite-of-custom-built-self-driving-hardware-c47d1714563

[apis]

This is something that's a piece of cake for Waymo and similar cars, but the Tesla autopilot thingie gets confused just because the lane marking isn't perfectly visible:


Tesla doesn't officially say it should either, they say it's only cruise control (unlike Waymo), but if you listen to Elon you'd think it's ready for full autonomous driving within a year.

[wraper]

This is something that's a piece of cake for Waymo and similar cars, but the Tesla autopilot thingie gets confused just because the lane marking isn't perfectly visible:

How about Waymo not working when raining but Tesla doing just fine?

https://youtu.be/KQo0KzbZNmc

[apis]

How about Waymo not working when raining but Tesla doing just fine?

It can only do that since it's not pretending to be fully autonomous. Showing a few minutes of driving in heavy rain with a driver prepared to take over any second doesn't prove it can do that over millions of miles without any mistakes. As far as I can tell, Tesla haven't shown they are anywhere close to the level of autonomy that Waymo and others who use Lidar and detailed maps are.

"To poke holes in Musk’s comments, you need look no further than the Society of Automotive Engineers (SAE) taxonomy for autonomous vehicles, commonly referred to as the SAE levels, which have become the global standard for defining self-driving. Most experts would categorize Navigate on Autopilot as Level 2 autonomy, meaning the vehicle can handle basic tasks, like acceleration, braking, and lane changes, but the human driver needs to maintain full attention of the road and be prepared to take control of the vehicle in a moment’s notice.
...
Not even highly automated vehicles like those operated by Alphabet’s Waymo, Ford’s Argo AI, or GM’s Cruise Automation would really qualify as “full self-driving” because they are only able to operate within a specific geographic location and under specific conditions, like good weather. Those companies typically keep safety drivers behind the wheel, with the understanding that their vehicles would be considered Level 4 capable under the SAE’s taxonomy."
https://www.theverge.com/2019/1/30/18204427/tesla-autopilot-elon-musk-full-self-driving-confusion

I like Elon, and he's doing cool stuff, but you have to take everything he says with a truckload of salt. Tesla is giving self driving cars a bad reputation imho. The one who deserve credit for being first and who have gotten the furthest with this technology is still Waymo.

The SAE classification isn't very good and maybe Tesla have a secret prototype they are developing in Singapore or somewhere else, who knows, but so far Tesla autonomous self driving is "all hat, no cattle" as far as I can tell.

[Dubbie]

Does anyone know the accidents per mile driven of Tesla Vs Waymo?

This to me would seem like a much more relevant statistic.

If tesla has driven 300K miles with 3 accidents, and Waymo has driven 50K with zero, this doesn't put Waymo ahead in safety in my mind.

[apis]

If Tesla's autopilot can't even handle a highway interchange reliably, which is a very basic task, they are clearly not anywhere near full autonomy, so trying to compare the number of fatalities is like comparing apples and oranges anyway. The best public data are number of disengagements in California which the companies are required to publish by law. According to that data Waymo are twice as good as GM who are second.

Tesla haven't even published any data because they aren't testing in California.
'Tesla said it “did not test any vehicles on public roads in California in autonomous mode or operate any autonomous vehicles, as defined by California law.”'

You can see the full table here:
https://www.therobotreport.com/waymo-autonomous-vehicles-apple/

[Dubbie]

Interesting.

Thanks for the data.

[CatalinaWOW]

The disengagements data is interesting on many levels.  For example Nissan managed to make very little improvement in the last year.  And Uber, who averaged a problem every half mile never had any business on public roads.  Not even close to ready for prime time.  It is really hard to assign any credibility to those who are going single digit distances without disengaging.  Plenty of fuel here for those who say that self driving is impossible, or at least impossibly far in the future.

On the other hand Waymo and GM Cruise are well into the danger zone.  Good enough to inspire complacency on the part of the human operator, but maybe not good enough to be fully in charge.

Finally, plotting the data shows that in general performance is well correlated with experience.  When experience reaches about ten times what Waymo has now you should expect about one disengagement in the life of an automobile.  Somewhere in that neighborhood of performance should be getting close to satisfying most folks.  Certainly in the same class as human operators.

The plot also clearly shows Apple and Uber as not good learners well out of family on the bad side, and Telenav and BMW on the other side of the scale.  All are low on the experience scale.  If Telenav and BMW can keep up that level of performance through a significant body of experience they could easily arrive at the finish line first.

[apis]

Interesting plot that.

Maybe it's worth pointing out again that disengagement doesn't mean the car would have caused an accident.

For example, as wraper has pointed out heavy rain is problematic because it blocks the lidar sensors. So heavy rain is likely one cause for disengagements in those statistics. But at the same time we see that Tesla autopilot drives in heavy rain without lidar. I.e. a car would likely be able to continue driving without any accidents relying only on the cameras and radars even in heavy rain. Heavy rain would mean an increased risk of accidents (the same is true for human drivers though), but an accident would still be unlikely, and the car might handle it by simply pull over and stop in a suitable location until the weather changes.

Similarly if the map is out of date, I would assume the cars will fail gracefully.

On the other hand, a disengagement might also occur because the car was about to run over a pedestrian and the safety driver had to step on the brakes to avoid it.

The latter is obviously a bigger problem than the former. Disengagements are a crude metric, but it's the best that's available since it's something the companies are required to report by law.

[wilfred]

Wouldn't a disengagement require a licenced driver at the wheel? I'm starting to see a problem with training the next generation of licenced drivers. If you're going to take over in a crisis you sort of need experience to be able to make the decisions. Hopefully the car will be better than an inexperienced driver.

I'm still unconvinced a non-driving driver will have any hope of remaining engaged with the traffic. They have to sit there concentrating as if they were driving and babysit a machine that they implicitly don't trust completely. I'd rather take a bus.

[maginnovision]

Also the idea that your car is just going to stop until weather gets better... Who would want that? I don't think the market for that would be very large.

[Dubbie]

Or it could stop while the passenger got themselves into a position to drive the rest of the journey manually.
That would make sense.

Obviously if there was no steering wheel this couldn't happen, but i imagine that by the time the steering wheel goes, the cars would be better drivers than any human.

[Marco]

AFAICS "disengagement" does not include when the vehicle just comes to a safe stop because it has no clue what to do next.

[wilfred]

What will an auto do when one of these bastards crawls across the lidar sensor. They love crawling across windscreens and freaking out drivers. But in the words of Douglas Adams they are mostly harmless. F'ing big though.

(short video, warning it is vertical)

[apis]

What will an auto do when one of these bastards crawls across the lidar sensor. They love crawling across windscreens and freaking out drivers. But in the words of Douglas Adams they are mostly harmless. F'ing big though.

It's like a tiny eight-legged cat, cute.

The cars have several lidars and several cameras and radars, they don't depend on just one sensor, so it would probably handle it much better than an arachnophobic human driver.

[apis]

AFAICS "disengagement" does not include when the vehicle just comes to a safe stop because it has no clue what to do next.

I believe it's anything that interrupts the planned driving task. It's not a very good metric, but it's the best that's publicly available.

[windsmurf]

CR finds that the new automatic lane-changing feature is far less competent than a human driver

“It’s incredibly nearsighted. It doesn’t appear to react to brake lights or turn signals, it can’t anticipate what other drivers will do, and as a result, you constantly have to be one step ahead of it.”

“This isn’t a convenience at all,” says CR’s Fisher. “Monitoring the system is much harder than just changing lanes yourself. Using the system is like monitoring a kid behind the wheel for the very first time. As any parent knows, it’s far more convenient and less stressful to simply drive yourself.”


https://www.consumerreports.org/autonomous-driving/tesla-navigate-on-autopilot-automatic-lane-change-requires-significant-driver-intervention/

[soldar]

I've read an interesting article in one of the bigger newspapers of the NL. China is becoming a real power where it comes to automotive engineering for both electric and self driving cars. They have the biggest uniform market for cars in the world.

I posted this in another thread:

Three years ago the market for passenger vehicles was (in millions):

- USA: 7.5
- EU: 14.5
- China: 21.5

My guess is that their share has grown since then.

[windsmurf]

Elon Musk shakes up Tesla Autopilot team after growing frustrated with the group’s progress
https://bgr.com/2019/07/09/tesla-autopilot-team-elon-musk-shakes-up/

Musk's decision not to implement LIDAR in Teslas may be backfiring.

[Towger]

Another spin on the same story: https://www.theregister.co.uk/2019/07/10/teslas_autopilot_losing_devs/

[windsmurf]

The Tesla Model 3 with Autopilot driving, with driver that fell asleep:

[orion242]

https://www.businessinsider.com/automatic-braking-cars-still-hit-pedestrians-aaa-study-2019-10

[Gyro]

It looks as if the latest autopilot isn't too fussy about whether the driver is present or not!

Tesla's Autopilot 'tricked' to operate without driver

The Autopilot feature in Tesla vehicles can be tricked into operating without a driver, an influential consumer magazine in the US has found.

Consumer Reports engineers looked into claims that Autopilot can operate without a driver present.

They tested the Model Y on a closed track and concluded the system could be "easily tricked".

It comes days after a fatal Tesla crash in Texas. Police believe no one was in the driver's seat.

https://www.bbc.co.uk/news/technology-56854417


If you don't sufficiently defend against something, you can be sure that some moron is going to do it. 

[wraper]

If you don't sufficiently defend against something, you can be sure that some moron is going to do it. 

If you are dumb enough, you can easily make any car drive without anyone being present on driver seat, no autopilot needed. Which seems to be this particular case.

[Marco]

Even the biggest nitwit knows a normal car can't drive itself, I can forgive people for people assuming a car with Full Self Driving beta pretty much can.

That's why level 3 is a mistake. Level 3 autonomous systems should only be allowed with professional safety drivers supervised by systems so incredibly annoying (ie. no eyes on the road -> immediate safety buzzer) only a paycheck would convince you to abide by them. No automatic lanekeeping either outside of emergencies, if you drift slowly out of a lane and it intervenes you should know it and not like it ... all automation which allows you to take eyes off the road and stop active steering should be so annoying you don't want to have it kick in, until the car can actually drive itself. Level 4 or bust.

[wraper]

Even the biggest nitwit knows a normal car can't drive itself, I can forgive people for people assuming a car with Full Self Driving beta pretty much can.

That's why level 3 is a mistake. Level 3 autonomous systems should only be allowed with professional safety drivers supervised by systems so incredibly annoying (ie. no eyes on the road -> immediate safety buzzer) only a paycheck would convince you to abide by them. No automatic lanekeeping either outside of emergencies, if you drift slowly out of a lane and it intervenes you should know it and not like it ... all automation which allows you to take eyes off the road and stop active steering should be so annoying you don't want to have it kick in, until the car can actually drive itself. Level 4 or bust.

FSD beta is only available to a small number of individuals. Those who abuse it, get banned from using it.

[Marco]

Using it is abusing it, it's legal but it shouldn't be. The level of annoyance necessary to make this systems safe to test simply isn't there. The level of annoyance necessary to make their old level 3 automation safe to use isn't there either.

They are creating new significant failure modes (ie. more collisions with static objects, due to the inability of radar to detect them and visual identification creating too many false positives to be usable). That's unacceptable to me, they should be making existing failure modes less likely, not creating new ones.

[SilverSolder]

If you don't sufficiently defend against something, you can be sure that some moron is going to do it. 

If you are dumb enough, you can easily make any car drive without anyone being present on driver seat, no autopilot needed. Which seems to be this particular case.

Unfortunately, stupidity seems to be a nearly infinite natural resource! 

[james_s]

Any automated system that requires a human operator to be available to take over immediately in the event that the automation cannot handle a situation is fatally flawed by design. It is a well known fact of human psychology that if you take away the requirement for the brain to focus on something, it will find something else to focus on instead and will not be ready to take over. If you have to maintain a level of focus that would enable taking over then the system is completely useless.

If I'm driving a car, I just want to DRIVE the car. The less gadgets and gimmicks between me and the road the better.

[Gyro]

No, I can't think of anything more boring or annoying than having to pay close attention to the road without physically having anything to do. Especially when I might be called to action at any moment.

Things like a nice warning that I'm drifting out of lane are good, it prompts me that my attention is wandering.

[SilverSolder]

No, I can't think of anything more boring or annoying than having to pay close attention to the road without physically having anything to do. Especially when I might be called to action at any moment.

Things like a nice warning that I'm drifting out of lane are good, it prompts me that my attention is wandering.

That is actually a good safety feature, I have seen several terrible accidents in my lifetime that were due to people falling asleep behind the wheel...  The worst one, a whole family was wiped out... 

[Marco]

It's fine, as long as it's utterly annoying. People have to desperately want to avoid it kicking in. Safety through annoyance, until level 4 (never ever) it's the only way.

Tesla's design of it's automation is designed to sell cars while taking massive risks. If there's a cop or emergency responder between the Tesla and the next police car or firetruck they rearend they're going to be the new Uber, it's a real existential risk to Tesla which they are taking for competitive advantage. Risky business.

[james_s]

In which case it's completely useless, because it just ends up being more annoying than actually driving manually.

[Marco]

In which case it's completely useless, because it just ends up being more annoying than actually driving manually.

You don't tend to manually drift into an occupied lane.

It can nudge you back annoyingly when it detects a danger without taking over normal steering and lulling you into sleep so you can hit a firetruck.

[james_s]

You don't need self driving for lane departure warning. Personally I don't want either one, I want a simple car with a proper manual gearbox so I focus 100% on *driving* the car when I'm behind the wheel. All these gadgets and tech cars have now is just crutches that enable people to focus on anything but driving.

[wraper]

Personally I don't want either one, I want a simple car with a proper manual gearbox so I focus 100% on *driving* the car when I'm behind the wheel.

That's certainly a pure pleasure when you are slowly moving in congestion.

[Ed.Kloonk]

No, I can't think of anything more boring or annoying than having to pay close attention to the road without physically having anything to do. Especially when I might be called to action at any moment.

Things like a nice warning that I'm drifting out of lane are good, it prompts me that my attention is wandering.

It's a bit like having to sit in the passenger seat with an elderly parent driving. Some of you would know what I'm talking about. Often you just want to drive the car yourself. But what if the driver claims competency and pulls rank anyway?

I think this is the way the car companies will lead with their automation. They do already, by proxy, bully the owner into believing the the car knows better. Will come a day when the car will deny you intention to drive. And you won't be able to tell it to fuck off and sit in the other seat.

[SilverSolder]

Personally I don't want either one, I want a simple car with a proper manual gearbox so I focus 100% on *driving* the car when I'm behind the wheel.

That's certainly a pure pleasure when you are slowly moving in congestion.

OK, we need a few ingredients:  a nice place to go, a proper stick shift, a babe sitting next to you, convertible roof, 1KW stereo...   25 years off your age wouldn't hurt either... 

[james_s]

Personally I don't want either one, I want a simple car with a proper manual gearbox so I focus 100% on *driving* the car when I'm behind the wheel.

That's certainly a pure pleasure when you are slowly moving in congestion.

Slowing moving in congestion is a drag no matter what. I've been driving a manual for over 20 years, every now and then I have to drive something with a slushbox, it makes no difference in heavy traffic but any other situation the manual is hugely superior.

[BrianHG]

Elon Musk shakes up Tesla Autopilot team after growing frustrated with the group’s progress
https://bgr.com/2019/07/09/tesla-autopilot-team-elon-musk-shakes-up/

Musk's decision not to implement LIDAR in Teslas may be backfiring.

Tesla's by my count has less than 1/100th the required processing power to run full self driving as reliably as a human using just cameras.

Also, their method/scope of training also way too narrow focused just so they can get something barely functional with such small narrow minded neuro-net processor.

Full self driving will soon one day come, but, I think with Elon's narrow scope of the true development and processing requirements, he will end up falling to a newer generation/company who will bring the real thing to fruition unless he changes his business practices and truly pays/invests special attention to the authentic scope of the problem at hand.  Anything he claims his current Tesla's will be able to perform true self driving is just an investor sales pitch which will never be truly delivered without user's lives being placed at risk.

[BrianHG]

It looks as if the latest autopilot isn't too fussy about whether the driver is present or not!

Tesla's Autopilot 'tricked' to operate without driver

The Autopilot feature in Tesla vehicles can be tricked into operating without a driver, an influential consumer magazine in the US has found.

Consumer Reports engineers looked into claims that Autopilot can operate without a driver present.

They tested the Model Y on a closed track and concluded the system could be "easily tricked".

It comes days after a fatal Tesla crash in Texas. Police believe no one was in the driver's seat.

https://www.bbc.co.uk/news/technology-56854417


If you don't sufficiently defend against something, you can be sure that some moron is going to do it. 

https://xkcd.com/1559/


https://xkcd.com/1897/

[Ed.Kloonk]

^   So true.   

[SilverSolder]

Personally I don't want either one, I want a simple car with a proper manual gearbox so I focus 100% on *driving* the car when I'm behind the wheel.

That's certainly a pure pleasure when you are slowly moving in congestion.

Slowing moving in congestion is a drag no matter what. I've been driving a manual for over 20 years, every now and then I have to drive something with a slushbox, it makes no difference in heavy traffic but any other situation the manual is hugely superior.

Old school manuals are slowly disappearing in favour of paddle shifted "manumatics" and the like...

All part of justifying why the average price of a new car has passed $38K...    there has to be some tech in there to justify it! 

[Ed.Kloonk]

Old school manuals are slowly disappearing in favour of paddle shifted "manumatics" and the like...

All part of justifying why the average price of a new car has passed $38K...    there has to be some tech in there to justify it! 

When I was in England, manuals were -everywhere-. I asked around and the reason I was given was that the manual goes back decades. Nobody bought the automatic because you'd spend that extra money on other features. Or save the money, of course.

I'm wondering if is still the case and if the non-essential high-tech in cars there might be overlooked.

[wraper]

Tesla's by my count has less than 1/100th the required processing power to run full self driving as reliably as a human using just cameras.

Also, their method/scope of training also way too narrow focused just so they can get something barely functional with such small narrow minded neuro-net processor.

Full self driving will soon one day come, but, I think with Elon's narrow scope of the true development and processing requirements, he will end up falling to a newer generation/company who will bring the real thing to fruition unless he changes his business practices and truly pays/invests special attention to the authentic scope of the problem at hand.  Anything he claims his current Tesla's will be able to perform true self driving is just an investor sales pitch which will never be truly delivered without user's lives being placed at risk.

Are you an expert in AI?
All that is required to make a regular car drive without driver present on a seat.

[SilverSolder]

Old school manuals are slowly disappearing in favour of paddle shifted "manumatics" and the like...

All part of justifying why the average price of a new car has passed $38K...    there has to be some tech in there to justify it! 

When I was in England, manuals were -everywhere-. I asked around and the reason I was given was that the manual goes back decades. Nobody bought the automatic because you'd spend that extra money on other features. Or save the money, of course.

I'm wondering if is still the case and if the non-essential high-tech in cars there might be overlooked.

Yep, most European countries like their manual transmissions.  They work better with the often small, fuel sipping engines than slushboxes do. 

The US has always preferred automatics with their bigger cars, bigger engines, and bigger roads, and less "work" to drive.

What do people prefer in Australia?

[BrianHG]

Tesla's by my count has less than 1/100th the required processing power to run full self driving as reliably as a human using just cameras.

Also, their method/scope of training also way too narrow focused just so they can get something barely functional with such small narrow minded neuro-net processor.

Full self driving will soon one day come, but, I think with Elon's narrow scope of the true development and processing requirements, he will end up falling to a newer generation/company who will bring the real thing to fruition unless he changes his business practices and truly pays/invests special attention to the authentic scope of the problem at hand.  Anything he claims his current Tesla's will be able to perform true self driving is just an investor sales pitch which will never be truly delivered without user's lives being placed at risk.

Are you an expert in AI?
All that is required to make a regular car drive without driver present on a seat.

Not an expert, but I know enough to see that you need more that just the ability to recognize the road, people, obstructions, lanes, sign recognition, driving rules, map reading.  Having all the basic functions is not enough tied to cameras and scene recognition isn't enough.  They are missing the multiple abstract point of views which such a small processor can handle.  Like, what would happen in a situation where you need to merge in traffic, but the other drivers on the road are being pricks because they know they can break some driving rules and get away with it while they know Tesla's driving AI will not break the rules, never giving it a chance to merge, or since the AI wont aggressively push itself into an intersection at the end of a traffic light cycle to make a left turn where is some circumstance no one would allow it to get a chance at a left turn.

Also, all the firing of the managers for Elon's self driving AI is also suspicious that they already know certain limits have been reached with the existing hardware in his cars.  We have been promised for years ago by Elon that it would already be here, and yet a human driver still needs to be present and in control of the wheel.

Like I said, true autonomous driving is coming, but it's beginning to look like it wont come from the existing Teslas.

[wraper]

They are missing the multiple abstract point of views which such a small processor can handle.

It's not small by any means and there are 2 of them. https://en.wikichip.org/wiki/tesla_(car_company)/fsd_chip

to run full self driving as reliably as a human using just cameras.

It's a pretty low standard considering that most of the humans make a lot of mistakes, are inattentive, or drive outright erratic.

Also, all the firing of the managers for Elon's self driving AI is also suspicious

The same was done with Starlink due to insufficient development speed. Fired managers were hired by Amazon for project Kuiper, and it does not seem they are moving anywhere fast enough. By the time they will launch anything, they will be not needed since Starlink will take over all of the market.

[BrianHG]

They are missing the multiple abstract point of views which such a small processor can handle.

It's not small by any means and there are 2 of them. https://en.wikichip.org/wiki/tesla_(car_company)/fsd_chip

to run full self driving as reliably as a human using just cameras.

It's a pretty low standard considering that most of the humans make a lot of mistakes, are inattentive, or drive outright erratic.

  Make that at least 250x if not 500x more processing power power required.
You are interpreting my comments at the wrong level and like Elon, missing the true requirements of a system required to out-perform a human when said human isn't making any mistakes.

[wraper]

  Make that at least 250x if not 500x more processing power power required.

Frankly you are just pulling numbers out of your ass.

You are interpreting my comments at the wrong level and like Elon, missing the true requirements of a system required to out-perform a human when said human isn't making any mistakes.

It needs to be better than a real average human. Not better than an idolized flawless human who does not exist.

[BrianHG]

  Make that at least 250x if not 500x more processing power power required.

Frankly you are just pulling numbers out of your ass.

You are interpreting my comments at the wrong level and like Elon, missing the true requirements of a system required to out-perform a human when said human isn't making any mistakes.

It needs to be better than a real average human. Not better than an idolized flawless human who does not exist.

Again, I didn't mean there is an idealized perfect human.  I meant when comparing a driving situation where a human hasn't made a mistake.

Yes, my numbers are designed to nudge your brain a little to try to expand your view.  But I do not expect that tiny amount of GPU and NPUs with that tiny ram to be able to assess is a pedestrian on the sidewalk is in a fight, or drunk, or of a child appears to be ignoring his surroundings who looks as if they are about to turn onto the road the same way a human with attention to the situation would know to be extra cautious.

[Gyro]

They are missing the multiple abstract point of views which such a small processor can handle.

It's not small by any means and there are 2 of them. https://en.wikichip.org/wiki/tesla_(car_company)/fsd_chip

Just the right number for a driver - back seat driver argument! 

[wraper]

Again, I didn't mean there is an idealized perfect human. I meant when comparing a driving situation where a human hasn't made a mistake.

It's basically a definition of an idolized flawless human. Cherry picking some situations when human might respond better, does not mean that human will be a better driver overall. https://en.wikipedia.org/wiki/Cherry_picking

[BrianHG]

Again, I didn't mean there is an idealized perfect human. I meant when comparing a driving situation where a human hasn't made a mistake.

It's basically a definition of an idolized flawless human. Cherry picking some situations when human might respond better, does not mean that human will be a better driver overall. https://en.wikipedia.org/wiki/Cherry_picking

Continue reading the rest of my post.

As for cherry picking situations, these so-called situation are what the FSD must pass to be public-ally acceptable in a dynamic world with pesky humans in it.  And like I said, if Elon wont address this, you better believe that there are others who are working on ways of doing so.  They will be the future and they will the ones to succeed at a level 5 FSD.

Just sticking you head in the ground ignoring something which can be solved leaves the door open to those who will do the work and make something better.  Especially if company 'B' comes out and proves their AI can perform some situational awareness of the intent of other humans bot driving cars as well as pedestrians.

[SilverSolder]

I'm looking forward to a driverless car with a drop-down menu in its display:

[Marco]

Since we already have a xkcd comic, here's one from SMBC.

Tesla has always relied on extremely risky strategies to sell their car. They got lucky that all the firetrucks and police cars they hit never killed a cop or emergency responder, they got lucky no big lawsuits related to FSD promises got off the ground ... but their luck will run out. All that's staving off lawsuits for false advertising and breach of contract is that they keep doubling down, instead of throwing in the towel and admitting the timeframe is shot and they have no idea when or if it will ever work, but I can't see that strategy working indefinitely.

https://www.smbc-comics.com/comic/fsd

[wraper]

I'm looking forward to a driverless car with a drop-down menu in its display:

[RJSV]

   Thought the video, on neural networks, would help give an introduction, at least helping me understand the scale of the hardware involved.
- - Rick-Jack

[Marco]

Tesla build in a hack to try to detect flashing lights to help with all the false negatives for crashes against stationary vehicles. Hasn't helped yet.

I don't see why people expect so much from FSD when autopilot keeps getting into silly crashes.

[coppice]

When I was in England, manuals were -everywhere-. I asked around and the reason I was given was that the manual goes back decades. Nobody bought the automatic because you'd spend that extra money on other features. Or save the money, of course.

I'm wondering if is still the case and if the non-essential high-tech in cars there might be overlooked.

In England it used to be big car automatic, small car manual, and in England most cars are small. Things have really changed in the last 5 years. Automatics now slightly outsell manuals, and their market share is increasing rapidly. With electrification, whether partial (i.e. hybrids) or full, automatics will clearly take over. Even driving schools are finally considering taking automatics seriously as learner cars.

[Ed.Kloonk]

When I was in England, manuals were -everywhere-. I asked around and the reason I was given was that the manual goes back decades. Nobody bought the automatic because you'd spend that extra money on other features. Or save the money, of course.

I'm wondering if is still the case and if the non-essential high-tech in cars there might be overlooked.

In England it used to be big car automatic, small car manual, and in England most cars are small. Things have really changed in the last 5 years. Automatics now slightly outsell manuals, and their market share is increasing rapidly. With electrification, whether partial (i.e. hybrids) or full, automatics will clearly take over. Even driving schools are finally considering taking automatics seriously as learner cars.

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

[james_s]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

I wish they did that here. I think more people would choose a manual if they'd ever had the chance to experience driving one. As it stands most people never get the opportunity since almost all cars have a slushbox.

[madires]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

We have that already for a long time. Since last year people with a driver's license for automatics can take 10 driving lessons (each 45 minutes) and a brief driving test (15 minutes) to 'upgrade' to manual.

[nctnico]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

I wish they did that here. I think more people would choose a manual if they'd ever had the chance to experience driving one. As it stands most people never get the opportunity since almost all cars have a slushbox.

Having driven mostly manual cars I'd say having an automatic gearbox would be better. In the end shifting gears manually is just a tedious job.

What you likely mean is that driver's training in the US should be taken more seriously and the bar to pass the theoretical and practical exams should be set higher so the quality of driving skills improves.

[coppice]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

In the UK, and many other places, if you pass your test in an automatic, you cannot drive a manual car. They are two completely separate classes of licence. People visiting the UK, who try to rent a car, often have hassle, because the rental companies have very few automatics in their fleet.

[SilverSolder]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

In the UK, and many other places, if you pass your test in an automatic, you cannot drive a manual car. They are two completely separate classes of licence. People visiting the UK, who try to rent a car, often have hassle, because the rental companies have very few automatics in their fleet.

Seems a bit restrictive?  - after all, many can (and do) learn to drive a manual after taking their test...

[Ed.Kloonk]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

In the UK, and many other places, if you pass your test in an automatic, you cannot drive a manual car. They are two completely separate classes of licence. People visiting the UK, who try to rent a car, often have hassle, because the rental companies have very few automatics in their fleet.

Seems a bit restrictive?  - after all, many can (and do) learn to drive a manual after taking their test...

In that case you have to update the licence conditions. Here, you attain qualifications if you can ride a motorcycle or a operate a heavy vehicle or you can have a conditional licence where you require glasses.

But yes, the drive a manual condition is very discriminatory. We all know that it shouldn't be an issue if the driver is male.

[coppice]

Where I am they now place a distinction on your licence if you didn't learn (or pass the exam) in an manual. In other words, you cannot drive a manual unless qualified.

In the UK, and many other places, if you pass your test in an automatic, you cannot drive a manual car. They are two completely separate classes of licence. People visiting the UK, who try to rent a car, often have hassle, because the rental companies have very few automatics in their fleet.

Seems a bit restrictive?  - after all, many can (and do) learn to drive a manual after taking their test...

Then they have to put "L" plates on the car while learning to drive a manual, and you have to take the driving test again in a manual car. Its just like upgrading to any "higher" kind of licence, like one of the various truck driving licences.

There is a weird thing in the UK licence conditions. You can drive a delivery van with a car licence, but not a large truck. That seems reasonable. However, you can also drive a full sized double decker bus with a car licence, as long as nobody on board pays. If anyone pays anything for their ride you need a "public service vehicle" licence, where you get tested driving actual buses.

[james_s]

Having driven mostly manual cars I'd say having an automatic gearbox would be better. In the end shifting gears manually is just a tedious job.

What you likely mean is that driver's training in the US should be taken more seriously and the bar to pass the theoretical and practical exams should be set higher so the quality of driving skills improves.

No, I mean exactly what I said, learning to shift a manual gearbox should be required. Almost all cars here are automatic, I hate them, absolutely can't stand the feel, I like shifting gears, it feels like *driving*, with an automatic you just kind of aim. I hate that mushy, slushy disconnected feel and the car trying to guess what I want. A manual is especially beneficial in the snow as it offers much more control, and in mountainous areas such as where I live where it offers far superior engine braking when going down a mountain pass. When I got the car I currently have I could only find one with an automatic, so I yanked it out and replaced it with the manual gearbox and associated parts from the car that got wrecked. I'll give up a manual gearbox when I eventually get an EV.

[SiliconWizard]

Over here, automatic cars are the exception rather than the rule. I wonder what made them this popular in the US? What's the history behind it?

[PeteH]

...
I'll give up a manual gearbox when I eventually get an EV.

Coming from a manual transmission to an EV - the experience is very similar. Engine braking vs. Regen. Driving an EV, you feel just as coupled to the car if not more (like you're always in first gear) - depending on the make/model I guess.

[tom66]

There is a weird thing in the UK licence conditions. You can drive a delivery van with a car licence, but not a large truck. That seems reasonable. However, you can also drive a full sized double decker bus with a car licence, as long as nobody on board pays. If anyone pays anything for their ride you need a "public service vehicle" licence, where you get tested driving actual buses.

This led to what could be described as an amusing exemption for bus lanes:  some people literally bought buses to drive in those lanes.  You could find something like a small minibus and drive it around on a car licence, using the bus to skip traffic.

Councils weren't slow to update signage to "local" buses only, i.e. those registered with the local council.

[james_s]

Over here, automatic cars are the exception rather than the rule. I wonder what made them this popular in the US? What's the history behind it?

Wide open highways across vast distances is one aspect, lazy people, a period of immense affluence and technological development following WWII when all manner of fancy automatic gadgets to ostensibly make life easier were all the rage. Then there is the historically much lower fuel prices here, until the last 20 years or so an automatic transmission came with a quite significant penalty on fuel economy.

Probably 98% of the cars on the road in the US have automatics. A few years ago my uncle bought a brand new sporty BMW and he had to special order it with a manual, they did not have a single manual equipped car in stock on the lot.

[bdunham7]

Over here, automatic cars are the exception rather than the rule. I wonder what made them this popular in the US? What's the history behind it?

Large heavy cars with large displacement engines were common 5 decades ago and manual transmissions were impractical and not offered in those models.  My first car had a 7.5 liter engine.  People just got used to them, I guess.  Then when cars got downsized, manuals did make a comeback for a while, but automatics were almost always at least an available option if not the predominate (or only) configuration.

[tom66]

Large heavy cars with large displacement engines were common 5 decades ago and manual transmissions were impractical and not offered in those models.  My first car had a 7.5 liter engine.  People just got used to them, I guess.  Then when cars got downsized, manuals did make a comeback for a while, but automatics were almost always at least an available option if not the predominate (or only) configuration.

I don't buy the idea that a large and powerful engine makes a manual transmission impractical.

If anything it makes a traditional automatic problematic - you need a large torque converter and more cooling.  Whereas a manual transmission can probably get by without cooling the clutch and with just a bit more metal in the gears. 

In fact, some of the most powerful vehicles were offered with manuals only if you look back in history, e.g. old Lamborghini and Ferrari models.

The use of automatics was purely down to customer convenience & low cost of fuels in the USA.  Truly an interesting difference in how policies can shape an entire market.  Higher taxes drove Europe to smaller, more efficient cars.  It's not a bad outcome!  The USA will get there eventually as fuel passes $8 a gallon.

[james_s]

Large heavy cars with large displacement engines were common 5 decades ago and manual transmissions were impractical and not offered in those models.  My first car had a 7.5 liter engine.  People just got used to them, I guess.  Then when cars got downsized, manuals did make a comeback for a while, but automatics were almost always at least an available option if not the predominate (or only) configuration.

There's no technological reason manuals were not offered in many of those. The biggest trucks have always been manual, most are even today. Plenty of big V8 powered muscle cars were available with manual, race cars have always been almost exclusively manual.

[CatalinaWOW]

Practicality of transmission was definitely not the reason for automatic adoption in the US.  Americans liked them, and the penalty in fuel consumption wasn't enough to drive folks away.  I have never seen real numbers, but I would guess that in the auto transmissions of the era the penalty was under 20%, perhaps way under.  And that is comparing to an optimally operated manual transmission, something the average driver seldom if ever achieves.  By the seventies whatever difference there was started growing smaller as lockup torque converters began being adopted, and currently the advantage goes the other way because the computer and a large number of gears can operate the drive train more optimally than all but the best drivers.  Cost of acquisition is another factor in this decision, and initially favored manual transmissions.  But as more and more people chose automatics the economies of volume overcame the intrinsically lower cost of manufacture of the manual transmission.  Finally, maintenance costs tend to favor automatics in my opinion.  Solely because the average driver, and especially the driver who believes he is a king of speed absolutely murders clutches.  The manual has no intrinsic defense against this, while automatics do somewhat protect themselves.  The manual transmission itself is far cheaper to maintain.

While I learned on a manual, and have driven a wide range of manual gearboxes, including those without synchromesh I definitely prefer an automatic.  I drive to get somewhere, and having less to do is a plus for me.  I will stipulate that there are those who enjoy shoving levers around.  But then there are those who rue the loss of manual spark advance and manual chokes.   I enjoy the ability to start on a hill without skillful feathering of the clutch (particularly in a rental or other unfamiliar car).  I like not having one hand tied up with a shift lever.  It doesn't surprise me that many other people feel the same way.

[james_s]

I have never seen real numbers, but I would guess that in the auto transmissions of the era the penalty was under 20%, perhaps way under.  And that is comparing to an optimally operated manual transmission, something the average driver seldom if ever achieves.  By the seventies whatever difference there was started growing smaller as lockup torque converters began being adopted, and currently the advantage goes the other way because the computer and a large number of gears can operate the drive train more optimally than all but the best drivers. 

When I swapped a manual into my mom's 1986 Volvo 240 about 25 years ago the mileage increased from 24mpg to 29mpg so it was pretty significant. My current 1990 740 Turbo went from about 20mpg to 26mpg when I swapped in that and that's without putting any great effort into maximizing fuel economy. The difference was substantial, and the car feels so much more responsive. I don't think the fuel economy gap closed until the mid 2000's and even then a manual will typically do a little better if one knows what they're doing.

Modern manuals pretty much all have automatic hill hold, I could take it or leave it, shifting gears and starting on hills and whatnot is second nature for me, it's like walking, I don't even think about it.

[CatalinaWOW]

I have never seen real numbers, but I would guess that in the auto transmissions of the era the penalty was under 20%, perhaps way under.  And that is comparing to an optimally operated manual transmission, something the average driver seldom if ever achieves.  By the seventies whatever difference there was started growing smaller as lockup torque converters began being adopted, and currently the advantage goes the other way because the computer and a large number of gears can operate the drive train more optimally than all but the best drivers. 

When I swapped a manual into my mom's 1986 Volvo 240 about 25 years ago the mileage increased from 24mpg to 29mpg so it was pretty significant. My current 1990 740 Turbo went from about 20mpg to 26mpg when I swapped in that and that's without putting any great effort into maximizing fuel economy. The difference was substantial, and the car feels so much more responsive. I don't think the fuel economy gap closed until the mid 2000's and even then a manual will typically do a little better if one knows what they're doing.

Modern manuals pretty much all have automatic hill hold, I could take it or leave it, shifting gears and starting on hills and whatnot is second nature for me, it's like walking, I don't even think about it.

So by your numbers one car was 17% better, the other 23%.  Sounds like my 20% guess wasn't too far off.

I am happy that you have no difficulty on hills.  Is this true in every new car you get into?  When I get into a new manual vehicle it usually takes me a day or two to get used to the different clutch and engine torque characteristics.  Did you ever drive a Mazda rotary?  I owned one for a a year or two and driving characteristics were totally different than any other car I have owned.

I am also happy to hear that you are one of the superior drivers who can still best a current automatic.  I suspect though that the gas mileage for the overall population is better with current automatics.  Even though well over three quarters of the population is sure they are better than average drivers.

[SilverSolder]

The last generation of big American cars were not that bad on fuel...  I have an elderly 6 passenger Buick from 2000, with a V6, it does 35mpg on the highway at a steady 60 - 65.   It has an automatic with a lock-up converter, and very tall gearing...   It actually gets better mileage on the highway than my Escape hybrid!   - obviously once you are in town, things are very different...

[RJSV]

   Loved my Mitsubishi - Jeep hatch back manual.  The little car was light enough so clutch action on uphill wasn't hard, and big hatch door allowed for some oversize loading.  Approx 28 mpg for that 1991 model.
Also seen as Dodge Colt versions.

[SilverSolder]

Fuel economy actually hasn't got much better since the 90's - 00's for average cars.   Obviously there are some stellar exceptions, e.g. the Prius et al., that really upped the game on that front significantly.

[madires]

Neither does the SUV mania in Europe help to decrease fuel consumption.

[james_s]

So by your numbers one car was 17% better, the other 23%.  Sounds like my 20% guess wasn't too far off.

I am happy that you have no difficulty on hills.  Is this true in every new car you get into?  When I get into a new manual vehicle it usually takes me a day or two to get used to the different clutch and engine torque characteristics.  Did you ever drive a Mazda rotary?  I owned one for a a year or two and driving characteristics were totally different than any other car I have owned.

It varies, usually it takes a day or so before I'm really comfortable in a different car, but it's not often that I drive something unfamiliar with a manual, there just isn't much to choose from. Occasionally I drive my mother's Volvo 850 and sometimes I borrow a diesel flatbed truck from my friend's business, both of those are manual but the're both familiar to me now. I drove the same car for 17 years and when it got totaled I got another similar one and put the transmission from the wrecked car in it so it's been quite a long time since I've had to get used to a new car that I was driving regularly.

I never have driven a rotary but I'd love to try. My parents had a RX5 but it got rear ended and totaled before I was born. I'd like to try driving a semi or a bus with a 2 stroke Detroit too, looks like fun. I like driving all sorts of random stuff, probably the most unusual was a Yugo GVL, it was not a fantastic experience by any means.

[james_s]

Coming from a manual transmission to an EV - the experience is very similar. Engine braking vs. Regen. Driving an EV, you feel just as coupled to the car if not more (like you're always in first gear) - depending on the make/model I guess.

Yes I drove my dad's Tesla around for a couple weeks after he passed away and I quite liked the way it felt. It had that same sort of instantly responding "connected" feel where the speed of the powerplant is directly tied to the speed of the vehicle. Essentially the same as a manual except no gearbox is needed due to the very wide operating range of the electric motor. It felt really nice, even though I hate the lack of physical switches and knobs on the dash I would seriously consider buying one if I drove enough for it to matter anymore.

The absolute worst experience I've had is a CVT, those things are awful and just feel weird. Many if not most car makers program them with fake shift points now to mimic the feel of a conventional automatic which is IMO the worst of both worlds, the unreliability CVTs are notorious for without the economy advantage of infinite ratios.

[tom66]

It's worth trying something based on an e-CVT architecture.  Examples include Toyota/Lexus Hybrids and the original Volt.

These have the smoothness of an EV but have a conventional petrol engine too.  They are ridiculously smooth because the pairing of the motor-generator set acts like an electronically variable torque converter with nearly infinite ratio modulation, and very rapid control of this ratio.  A planetary gearset and clutch effectively connects the engine directly through a single ratio once over about 40 mph, so they are pretty efficient too for highway driving.

http://prius.ecrostech.com/original/Understanding/PowerSplitDevice.htm

I do really like the regen braking on my car though (Golf PHEV with a regular 6-speed DSG automatic),  it becomes a game to regen exactly enough to come to a stop at the red light or so on.  The brake pedal gets touched only when truly necessary.  I've been giving my partner driving lessons in it, and she's hooked and wants something electric as her first car.  Thinking an e-Up or e-Golf or maybe a newer Leaf/Zoe. 

[nctnico]

Fuel economy actually hasn't got much better since the 90's - 00's for average cars.   Obviously there are some stellar exceptions, e.g. the Prius et al., that really upped the game on that front significantly.

Yes and no. Cars with small, turbocharged engines and hybrids have a significantly lower fuel consumption compared to similar sized cars so progress has been made. It is just that the gas guzzling options are also still for sale. My next car is going to be a hybrid (for various reasons). Most likely a Toyota. For me fuel costs are by far the largest costs of a car so fuel economy is important.

[james_s]

It's worth trying something based on an e-CVT architecture.  Examples include Toyota/Lexus Hybrids and the original Volt.

These have the smoothness of an EV but have a conventional petrol engine too.  They are ridiculously smooth because the pairing of the motor-generator set acts like an electronically variable torque converter with nearly infinite ratio modulation, and very rapid control of this ratio.  A planetary gearset and clutch effectively connects the engine directly through a single ratio once over about 40 mph, so they are pretty efficient too for highway driving.

Hybrids were an interesting stopgap that made sense at the time, but they are quickly becoming obsolete except for a few edge cases where people really need more than 300 miles or so without a break for charging. If I were still commuting to an office I'd buy an EV, I've driven them enough and know enough people who drive them that I'm sold, quite simply they work, well, and carrying around a gasoline engine and all the mess that goes with that is totally needless for anyone that has a garage, driveway, or charging available at the office, a perk that is becoming more and more common. With the way fuel costs have been increasing it makes sense to dispense with fuel entirely. It cost me around $90 now for 300 miles on my gasoline car, and around $10 for 300 miles on a Tesla Y. It would be an absolute no brainer if I drove more than the 2,000 miles or so a year I do these days. Good friend of mine just bought a 2013 Tesla S, the original battery in that is going strong, so that pretty much alleviates the concerns I once had about longevity.

[SilverSolder]

It's worth trying something based on an e-CVT architecture.  Examples include Toyota/Lexus Hybrids and the original Volt.

These have the smoothness of an EV but have a conventional petrol engine too.  They are ridiculously smooth because the pairing of the motor-generator set acts like an electronically variable torque converter with nearly infinite ratio modulation, and very rapid control of this ratio.  A planetary gearset and clutch effectively connects the engine directly through a single ratio once over about 40 mph, so they are pretty efficient too for highway driving.

Hybrids were an interesting stopgap that made sense at the time, but they are quickly becoming obsolete except for a few edge cases where people really need more than 300 miles or so without a break for charging. If I were still commuting to an office I'd buy an EV, I've driven them enough and know enough people who drive them that I'm sold, quite simply they work, well, and carrying around a gasoline engine and all the mess that goes with that is totally needless for anyone that has a garage, driveway, or charging available at the office, a perk that is becoming more and more common. With the way fuel costs have been increasing it makes sense to dispense with fuel entirely. It cost me around $90 now for 300 miles on my gasoline car, and around $10 for 300 miles on a Tesla Y. It would be an absolute no brainer if I drove more than the 2,000 miles or so a year I do these days. Good friend of mine just bought a 2013 Tesla S, the original battery in that is going strong, so that pretty much alleviates the concerns I once had about longevity.

I think you just nailed the use case for gasoline:   Low miles driven.   

The overall costs are what matter.   E.g. I recently bought an older used "primitive" gasoline car for $5000...  it is in beautiful shape, no rust, low miles, comfortable, quiet.   Sure, it drinks copious quantities of gasoline costing $5 per gallon...   but on the other hand, there are no monthly car payments,  and very little depreciation. 

Electric cars just haven't been around long enough (in large enough quantities) to be able to do something similar with one of those...   but the day will come, of course.

[CatalinaWOW]

Just to pound SilverSolder's point home, assuming his gas guzzler gets 10 mpg, and using james_s 2000 miles per year you would be paying $1000/year in fuel costs.  It would take 25 years to break even on a $30,000 EV, even assuming electricity is free.

[nctnico]

Having driven mostly manual cars I'd say having an automatic gearbox would be better. In the end shifting gears manually is just a tedious job.

What you likely mean is that driver's training in the US should be taken more seriously and the bar to pass the theoretical and practical exams should be set higher so the quality of driving skills improves.

No, I mean exactly what I said, learning to shift a manual gearbox should be required. Almost all cars here are automatic, I hate them, absolutely can't stand the feel, I like shifting gears, it feels like *driving*, with an automatic you just kind of aim. I hate that mushy, slushy disconnected feel and the car trying to guess what I want.

Sorry but I still don't see how others besides you would benefit from needing to learn how to drive a manual car. It is like saying everyone should love apple pie for no good reason.

[tom66]

It's a useful skill if you need to drive a van or borrow someone's car, which might happen to be manual, but in this day and age automatics shift just as well if not better than any manual car.  I definitely prefer driving auto.  But if I had a 20yr old car I doubt I'd like it so much.  I still remember my mother's Opel/Vauxhall Astra with its 3-speed automatic, bloody awful thing.

[james_s]

I think you just nailed the use case for gasoline:   Low miles driven.   

The overall costs are what matter.   E.g. I recently bought an older used "primitive" gasoline car for $5000...  it is in beautiful shape, no rust, low miles, comfortable, quiet.   Sure, it drinks copious quantities of gasoline costing $5 per gallon...   but on the other hand, there are no monthly car payments,  and very little depreciation. 

Electric cars just haven't been around long enough (in large enough quantities) to be able to do something similar with one of those...   but the day will come, of course.

Electric cars could be found cheaply for a while too, my dad sold a Nissan Leaf for $5,000 when he got his Tesla Y, he offered it to me for free but at the time my household had three cars with two people and I was already hardly driving anywhere. You have a point though in that one is unlikely to find something with a >100 mile range in that price class so if they need range then gas powered is probably the way to go. On the other hand there are people like a friend of mine who commutes ~60 miles each way 5 days a week, she bought a Tesla and it costs peanuts to drive and with the standard 240V charging cord she charges it in the driveway. At today's gas prices down there in California it would cost a fortune to drive even a hybrid. They are not THAT much better on fuel than a conventional ICE, on the highway they are nearly the same, where they excel is stop & go city traffic.

I love my Volvo turbo wagon, it's the perfect blend of sports car, pickup truck and minivan, more cargo space than most midsized SUVs while having nearly as much ground clearance and still being low enough that I can load large items on a clamp on roof rack without needing a stool. There is simply nothing on the market that can replace it in all applications so I stick with it. If it were reasonably feasible to swap in a Tesla powertrain I would be all over that though.

[james_s]

Sorry but I still don't see how others besides you would benefit from needing to learn how to drive a manual car. It is like saying everyone should love apple pie for no good reason.

Because if they do it then they have had exposure to it, and then more people would choose it and there would be more manual cars available. It's a good skill to have, it's not like it's hard to do, if my technologically inept elderly mother can drive a manual with ease then so can anyone else. It's not saying everyone should love apple pie, it's saying everyone should try apple pie, if they don't love it they can eat something else. Likewise in school we all have to learn some things that we might not ever use again, because the point of school is to get a well rounded education and get a taste of everything. They eliminated shop class from most schools years ago and now they wonder why there is a shortage of tradespeople.

People who drive manual transmissions are more skilled, more aware, more engaged and attentive drivers. With a manual you cannot mistake the gas for the brake and plow through the wall of a building as happens too often. When you are busy driving with both hands you are forced to pay attention and put down the damn smartphone and focus on driving. I believe that when one is driving a car, they should be 100% focused on *driving* the car. All these gadgets that are supposed to make driving safer mostly just enable people to focus on anything but controlling the several thousand pound machine they are rolling around in.

[RJSV]

    What's the trend, then, starting of at '25 years payback', today?
   I mean, will that 'COME DOWN',... To what ..
   '$ 18 years, to payback...'.  In, say 2033, ?

This is Circus Grifter Math.  How about replace words like 'soon to improve' with, uh, numbers...(sorry), ...uh, numbers, (sorry, sorry).
   What exotic metal shortages forecast ? OK then, let's all Tesla-up!  I'm getting mine used, I hear..
At bargain price, just wait...hopefully not 25 years.

   I only play a fool, in a couple of movies.

[SilverSolder]

[...]
 They eliminated shop class from most schools years ago and now they wonder why there is a shortage of tradespeople.
[...]

Despite going in the academic direction, I took all the shop classes back in high school.  That's where I met some of the best teachers and made some of my best friends.  -  dropping all the manual skills was/is a huge mistake, there is only so much you can learn from Youtube! 

[SilverSolder]

    What's the trend, then, starting of at '25 years payback', today?
   I mean, will that 'COME DOWN',... To what ..
   '$ 18 years, to payback...'.  In, say 2033, ?

This is Circus Grifter Math.  How about replace words like 'soon to improve' with, uh, numbers...(sorry), ...uh, numbers, (sorry, sorry).
   What exotic metal shortages forecast ? OK then, let's all Tesla-up!  I'm getting mine used, I hear..
At bargain price, just wait...hopefully not 25 years.

   I only play a fool, in a couple of movies.

The first factor is that the investment vehicles (cars!) are designed for a life of 100,000 miles or 10 years, whichever comes first.  That's built into the specification of every component on the vehicle (except wear parts like brake linings, 12V battery, etc.) .   Pretty much all factory and extended warranties stop at that point, if not before.  That's the realm of "new" or "newish" cars, I hope we can agree.

Once a car has survived beyond its design life,  the economics favor reliable and robust designs that don't rust, with good parts availability both from the manufacturer and secondary sources.  This kind of car can be very economical to own, as in less than $50 per month depreciation  (e.g. buy a car for $5000 and keep it for 10 years).

Another economical way to go is to buy a newish car and keep it for even longer...  e.g. buy a car for $12,000 and keep it for 15 or 20 years!

Here is where electric cars may not work so well, if they are going to be needing expensive batteries every so often.  Still unknown!

[tom66]

I think there will end up being a pretty healthy battery rebuild and replacement market as EVs become more commonplace.  What will be important is for battery cells and modules to be available from manufacturers for a reasonable price.

[nctnico]

I think there will end up being a pretty healthy battery rebuild and replacement market as EVs become more commonplace.  What will be important is for battery cells and modules to be available from manufacturers for a reasonable price.

My advice when buying an electric car (whether it is hybrid, battery only or fuel cell) is to research if the battery is made up of modules which can be changed individually. That way a repair of a battery is not that expensive to begin with and you may not need the dealer to do this job. At this moment I'm seeing prices in the 800 euro ballpark for a refurbished battery pack for common Toyota hybrids.

[tom66]

Also compatibility between vehicles.

People have figured out you can put the 40kWh battery from a newer Leaf into an older 24kWh one and it more or less works - the only bug is the range-o-meter can't show all the digits!

[wraper]

I think there will end up being a pretty healthy battery rebuild and replacement market as EVs become more commonplace.  What will be important is for battery cells and modules to be available from manufacturers for a reasonable price.

My advice when buying an electric car (whether it is hybrid, battery only or fuel cell) is to research if the battery is made up of modules which can be changed individually. That way a repair of a battery is not that expensive to begin with and you may not need the dealer to do this job. At this moment I'm seeing prices in the 800 euro ballpark for a refurbished battery pack for common Toyota hybrids.

And if you find one, it's basically guaranteed to be garbage due to lack of thermal management. So you get a battery which can be serviced but do not last long.

[nctnico]

I think there will end up being a pretty healthy battery rebuild and replacement market as EVs become more commonplace.  What will be important is for battery cells and modules to be available from manufacturers for a reasonable price.

My advice when buying an electric car (whether it is hybrid, battery only or fuel cell) is to research if the battery is made up of modules which can be changed individually. That way a repair of a battery is not that expensive to begin with and you may not need the dealer to do this job. At this moment I'm seeing prices in the 800 euro ballpark for a refurbished battery pack for common Toyota hybrids.

And if you find one, it's basically guaranteed to be garbage due to lack of thermal management. So you get a battery which can be serviced but do not last long.

No, not at all. Look at Volkswagen's ID.3 battery for a modern day example. It is modular with water cooling. This battery pack has been designed to be serviceable at a module level to keep repair costs low.

[wraper]

I think there will end up being a pretty healthy battery rebuild and replacement market as EVs become more commonplace.  What will be important is for battery cells and modules to be available from manufacturers for a reasonable price.

My advice when buying an electric car (whether it is hybrid, battery only or fuel cell) is to research if the battery is made up of modules which can be changed individually. That way a repair of a battery is not that expensive to begin with and you may not need the dealer to do this job. At this moment I'm seeing prices in the 800 euro ballpark for a refurbished battery pack for common Toyota hybrids.

And if you find one, it's basically guaranteed to be garbage due to lack of thermal management. So you get a battery which can be serviced but do not last long.

No, not at all. Look at Volkswagen's ID.3 battery for a modern day example. It is modular with water cooling. This battery pack has been designed to be serviceable at a module level to keep repair costs low.

Yes it consists of modules. But where did you get an idea it's serviceable?

[nctnico]

Check Munro's ID.3 battery teardown. And you can buy the battery modules from companies that refurbish them.

[wraper]

Check Munro's ID.3 battery teardown. And you can buy the battery modules from companies that refurbish them.

Yes you should be able to buy modules salvaged from scrap batteries. But it does not mean it will work. IMHO as minimum it will require firmware hacking, depending on how they are made, it's likely they cannot be reprogrammed. Also what about balancing? I'm not so sure that another module with different wear will work good enough. In theory you can replace them, in practice it's a big question if and when it will become possible.

[RJSV]

   I ran some numbers, for a 3-door Dodge Colt 1991 model:
   A moderate distance commute, figured 34 miles each way to work.  Car was purchased at $6500 used, 1993 dollars, and lasted 16 years.  However, that's a single person commute car, not a huge cargo, and comfortable for 2 adults. 
   I brought that forward, today's gas at $ 8 per gallon, San Jose. CA. To get a more modern estimate.
Comes close to $25 per day, gas, or $3900 per year.
Replace that with a 'used Volt' hybrid, at $25 k.
   I'm not clear, how to bring the 1993 purchase price up to today, ...perhaps $10,000...I don't know.
But that gets to a difference of $15 k.
So using those numbers, I would pay extra $15 k, to save $4000 per year, very roughly.  Something like 4 years, might start to break even, assuming that I'm still vigorously working / traveling.
Ah, but free electricity.  No grid capacity problems. No special metals, to mine...
etc etc
   I think people would buy Tesla to get modern features like GPS and in-car media.  That part is another 30k, or more than double the 'Volt' hybrid scenario.
   I would have to ramp-up my daily use, in a more typically busy job.  What percentage, of world-wide potential users, can start plugging in (no pun), and get those same 'San Jose' region numbers?
   A lot seems to depend on doing MORE driving, to gain some efficiency in numbers.
   I do see a lot of 'Rebate' offers, from various govt. agencies, that skews the numbers.

[tom66]

I think a lot of people that would have bought a BMW or a Mercedes bought a Tesla instead - so that really skews the numbers - $60k MSRP car electric vs petrol.  For many the electric is a no brainer, if you are going to spend that money.  Hence the Germans now are releasing cars in that range to compete with the Model S and top spec Model 3.

For average people there are things like the ID.3 (which is not coming to USA) which is priced like a Golf, and the Leaf, and Renault Zoe (I don't think Renault sell in the US either) or myriad stupidly big electric SUVs.

Electric cars are (generally) good but it does annoy me that so many are SUVs.  We need less SUVs, not more... yes, better that they're electric, but even better if they are small efficient hatchbacks.  But for that to change you'd need to tax on weight/volume, or have just general social change over the concept of driving 2.5 tonnes around to go to Wal-Mart just because once in a while you also go to Home Depot and then still fail to fit everything in without having it dangerously hanging out of a window.

[SilverSolder]

   I ran some numbers, for a 3-door Dodge Colt 1991 model:
   A moderate distance commute, figured 34 miles each way to work.  Car was purchased at $6500 used, 1993 dollars, and lasted 16 years.  However, that's a single person commute car, not a huge cargo, and comfortable for 2 adults. 
   I brought that forward, today's gas at $ 8 per gallon, San Jose. CA. To get a more modern estimate.
Comes close to $25 per day, gas, or $3900 per year.
Replace that with a 'used Volt' hybrid, at $25 k.
   I'm not clear, how to bring the 1993 purchase price up to today, ...perhaps $10,000...I don't know.
But that gets to a difference of $15 k.
So using those numbers, I would pay extra $15 k, to save $4000 per year, very roughly.  Something like 4 years, might start to break even, assuming that I'm still vigorously working / traveling.
Ah, but free electricity.  No grid capacity problems. No special metals, to mine...
etc etc
   I think people would buy Tesla to get modern features like GPS and in-car media.  That part is another 30k, or more than double the 'Volt' hybrid scenario.
   I would have to ramp-up my daily use, in a more typically busy job.  What percentage, of world-wide potential users, can start plugging in (no pun), and get those same 'San Jose' region numbers?
   A lot seems to depend on doing MORE driving, to gain some efficiency in numbers.
   I do see a lot of 'Rebate' offers, from various govt. agencies, that skews the numbers.

The numbers for electric look best if you drive a lot - your example is 16K miles per year, assuming 48 working weeks (4 weeks vacation with no driving).

Consider the case if you drive the national average 12K per year, now the break-even is maybe 5-6 years.

Consider the case if you drive half of that,  or if you own more than one vehicle so neither one gets driven far...   gasoline is still competitive.

[nctnico]

It also greatly depends on where you can charge a BEV. If you have to rely on public charging, then an efficient hybrid is cheaper and more convenient. The current crop of BEVs are nice as a small action radius roundabouts.

[RJSV]

WOW, thanks SilverSolder, that example I had given would now actually extend, (at least pre-covid), which means 'Silicon Valley's phenomina of pulling in workers from afar.  My case, 34 miles in to San Jose to work, is 1980's.  Now, same job, almost same roads (choked), and we're talking TRACY, CA. to San Jose, for a coveted Apple job.  I'd be surprised, if that didn't double my miles per year! (To 32,000).
   And VACATION estimate, Whaaaaa, woewoe wait; no 4 weeks, more like 2 1/2 weeks per year.
There are lots of (smart) folks stuck on roads, at 5:15 am start of day.
   But, aren't there other 'mecca' type geography, where a job-rich area, maybe Minnesota I don't know, that have those mega-commute ?  Somebody else going to respond here, saying:
"...You should see I-5555 at 5 am in Omaha...bloody hell..."
   Yeah, I liked, your concept, of folks having the bigger family vehicle, used a lot less during the week. BUT, there could be regulations / restrictions looming, for anything not 'windmill' related. Not that I dislike clean approaches to power generation.
Having a windmill 'thrown' at you can be messy.
   But, serious, I'm not thinking that commuting long, Tracy to Cupertino (San Jose), is statistically a large thing, world-wide.(?) Maybe 2 % ?
Thanks.

[SilverSolder]

WOW, thanks SilverSolder, that example I had given would now actually extend, (at least pre-covid), which means 'Silicon Valley's phenomina of pulling in workers from afar.  My case, 34 miles in to San Jose to work, is 1980's.  Now, same job, almost same roads (choked), and we're talking TRACY, CA. to San Jose, for a coveted Apple job.  I'd be surprised, if that didn't double my miles per year! (To 32,000).
   And VACATION estimate, Whaaaaa, woewoe wait; no 4 weeks, more like 2 1/2 weeks per year.
There are lots of (smart) folks stuck on roads, at 5:15 am start of day.
   But, aren't there other 'mecca' type geography, where a job-rich area, maybe Minnesota I don't know, that have those mega-commute ?  Somebody else going to respond here, saying:
"...You should see I-5555 at 5 am in Omaha...bloody hell..."
   Yeah, I liked, your concept, of folks having the bigger family vehicle, used a lot less during the week. BUT, there could be regulations / restrictions looming, for anything not 'windmill' related. Not that I dislike clean approaches to power generation.
Having a windmill 'thrown' at you can be messy.
   But, serious, I'm not thinking that commuting long, Tracy to Cupertino (San Jose), is statistically a large thing, world-wide.(?) Maybe 2 % ?
Thanks.

Yeah, the average number of miles driven is something like 14,000 per year in the USA...   Everybody's needs are different, the numbers are individual for each person.  For example, if you commute 30 miles to the next town on open highways, you can get close to 40mpg in a standard gasoline powered Honda Civic...

On the other hand, if you are stuck in dense traffic for an hour and a half each way, a standard internal combustion engine is going to get extremely poor fuel economy...   whereas a hybrid (e.g. a Prius) can still get 40mpg in that scenario -  and a plug-in hybrid might be able to do most of the journey on pure electric, then falling back to 40mpg hybrid mode...

If you don't drive long distances then @nctnico's suggestion of an electric runabout is a good idea, and can be an overall pretty economical affair, e.g. a Nissan Leaf.

[CatalinaWOW]

The Nissan Leaf is quite practical.  Green mavens will point out that it meets the vast majority of America's requirements for distance on a charge.

But there must be something very wrong with the Leaf.  It has been on the market for 10 years and is still eligible for the US Federal Tax credit for manufacturers who have sold less than 700,000 vehicles in the US.  In the meantime Tesla, GM and Ford have already crossed that threshold with their much less practical vehicles.  And in GM and Ford's case a much later start.

[tom66]

I'm a big EV fan, but I wouldn't buy a Leaf.

There are a few reasons:

* Nissan build quality is just seemingly worse year after year.  They are becoming a bit like Peugeot did in the mid 2000's.  They need to seriously invest in making higher quality products for the price they charge.  I know friends with older Nissan vehicles (mid 2010 age) that are rusting and falling apart. 

* The battery longevity is really poor.  This is true for 24kWh, especially true for 30kWh and somewhat true for 40kWh models.  A lack of any cooling system, combined with a subpar chemistry means most cars will be down 30% of their capacity within 7 years.  That's just not acceptable, especially given the car has quite limited range and charging speed to begin with.  The 62kWh cars have been out for a relatively short time, and do at least have some cooling improvements, but prior history makes them a dubious buy for now.

* RapidGate and Nissan's response to it.  To protect their 40kWh batteries, Nissan massively limit the charge rate once you have done more than one charging session in a typical drive.  This is because the battery gets too hot to allow further fast charging, and there is no cooling system to remove that heat.  So a car that can only charge at 45kW, which is already not at all competitive, will only manage 30kW at the next stop.

* Price!  Not competitive at all.  A 62kWh Leaf - equivalent in many ways to the 58kWh ID.3 or somewhat to the 50kWh Peugeot e-208 (a little smaller size & battery)  is £35,000.  The ID.3 is £5,000 cheaper (was £7k before the car shortage nonsense) and the e-208 is almost £10,000 cheaper.  If you spec the Leaf up fully, it starts to get as expensive as a Polestar 2 or Model 3.  Insane pricing.

* It's still Chademo.  What are Nissan thinking? 

[SilverSolder]

A used Leaf for sub $5K may be an OK local runabout?

[nctnico]

* It's still Chademo.  What are Nissan thinking?

Isn't there something like a conversion cable? Recently I read a story about owners of an older BEV who nearly got stranded because all the public chargers got upgraded and are a now incompatibly with their vehicle. As a result they where not able to charge their BEV along the highway as they where used to. I have not investigated at all but I would be highly, highly surprised if this problem can't be solved by a simple conversion cable.

[tom66]

CCS to Chademo is not easy because Chademo requires the charging side to supply 12VDC to provide various signals - in theory to activate the contactors in the battery pack (though this might just be a logic-level input on some vehicles.)

You would therefore need a small battery in the CCS adapter, which could charge from the 400VDC when available, so that it could provide the 12VDC.  CCS does not provide any DC power besides the HVDC, and it provides only a few mA on control signals before it has authorised/powered up.

The opposite way around is 'easier' (and Tesla did make one - not sure if they still do).  The 12VDC there is actually very useful, they use a simple boost converter to push the DC terminals up to 400VDC which helps one of the isolation tests complete.  Not sure why that's not needed for CCS, but the Tesla Chademo adapter certainly isn't/wasn't a simple bit of kit.

Really - we should all just be using CCS - it's a bit like Beta vs VHS now - they each had their distinct advantages and disadvantages but it's clear CCS has won.   IMO the big advantage is a combined AC/DC charging port makes packaging much easier, and CCS has active support for 350kW whereas Chademo is still stuck at 200kW (and most stations are only 50kW - a few rare ones support 100kW but with the Leaf being capped at 55kW and being pretty much the only customer *ahem* except for a brand-new Lexus SUV for some reason, there's no demand for any more really.)