Long-Term Fundamentals of Tesla Motors (TSLA)

[wdolson]

I expect burning of fossil fuels will be banned. Spewing toxins into air we all breath is not a right.

As alternatives become available, ICE will largely be banned in more and more areas. I read the French ban on ICE vehicles in Paris that will be going into effect in a couple of years has an exception for restored classic cars.

There is a huge difference between even as many as 100,000 ICE restored classic weekend cars in the world and the 1 billion ICE on the road today. For one thing very few of them will be driven except for special occasions because people will realize EVs are so much cheaper, but enthusiasts will keep them running and restored as hobby cars. They also won't be drivable long distances because the fuel for them will be hard to find.

The total emissions from these cars will be down in the noise compared to other emissions. Look back 110-115 years or so when the ICE to horse ratio was close to the EV to ICE ratio today. Horses were seen as the big pollution problems. The manure collected off the streets of big cities was becoming a staggering problem. The term "tail pipe emissions" originated as a euphemism for horse waste.

My father was born in 1920 and he clearly remembers cars being pitched as the pollution free alternative to horses. It wasn't until we got large number of ICE that we realized that the gasses emitted from car tailpipes wasn't good either.

 

[EVNow]

There is a huge difference between even as many as 100,000 ICE restored classic weekend cars in the world and the 1 billion ICE on the road today.

Ofcourse - but we don't allow DDT or allow people to dump mercury etc. Its possible burning gas will simply not be tolerated - or at the other end because of climate change we get into a mad max lawless situation where nothing is ever enforced.

 

[adiggs]

I expect burning of fossil fuels will be banned. Spewing toxins into air we all breath is not a right.

It'll be a lot easier to ban the burning of fossil fuels when the technology is working in a lab, and looks like it's going to make it into volume production, that will replace the burning of fossil fuels. So we're seeing limitations being implemented on personal autos - I think those don't start getting passed into law without Tesla and EV's in the state they're in today, as it's not hard to imagine a world without fossil fuel being burned in 10 years.

Nobody today is talking about banning fossil fuel burning to power our airplanes. Or our ships. Or even for commercial trucks. Shipping is making a big move to bunker fuel (or require onboard scrubbers to clean the exhaust). Because we don't have line of sight to a non-fossil fuel solution for ocean going shipping (for instance).


And really - if fossil fuels are out of personal autos and commercial trucking (the writing will be on the wall in 10 years - it'll be obvious, not only to us here, but also to random person-on-the-street) - that might not yet be enough of a reduction in fossil fuel emissions for everything we need, it WILL be the biggest step down the road.


I don't expect to see a law in any country, even in Norway, that blanket bans fossil fuel burning. Not in 20 years and probably not ever. In the short term, the replacement technology for all industries that burn fossil fuels doesn't exist. Longer term, the volume of fossil fuels being burned in the final couple of industries that still need them, plus the hobbyists keeping antiques running, won't matter.

 

[wdolson]

It'll be a lot easier to ban the burning of fossil fuels when the technology is working in a lab, and looks like it's going to make it into volume production, that will replace the burning of fossil fuels. So we're seeing limitations being implemented on personal autos - I think those don't start getting passed into law without Tesla and EV's in the state they're in today, as it's not hard to imagine a world without fossil fuel being burned in 10 years.

I think it will be more than 10 years. With production of cars and lights trucks at about 100 million a year and a fleet on the road of around 1 billion, switching all car production to EVs tomorrow would barely build enough cars to meet that goal. And there are many poorly developed areas in poorly developed countries where electricity is not that common, even fossil fuels are not that easy to come by, but there are street vendors selling it by the bottle.

But I do think we'll reach a point where people will think anyone daily driving an ICE is like someone wanting to use one of those old brick cell phones from the 1980s. Once that point is reached ICE will become rare. The poorest will probably be the last to electrify. The infrastructure to support EVs will be installed in poor neighborhoods last and the poor who need cars will first be buying cast off relatively new ICE, then finally used EVs. The poor will be able to buy iCE cheap because nobody will want them.

And really - if fossil fuels are out of personal autos and commercial trucking (the writing will be on the wall in 10 years - it'll be obvious, not only to us here, but also to random person-on-the-street) - that might not yet be enough of a reduction in fossil fuel emissions for everything we need, it WILL be the biggest step down the road.

Within 10 years I expect the public to generally be convinced EVs are the way to go. Being able to produce enough EVs (producing enough batteries) and being able to afford EVs will be the two barriers to adoption. Some governments might help the process by offering rebates to lower income people who buy EVs, but not all countries will do that.

I don't expect to see a law in any country, even in Norway, that blanket bans fossil fuel burning. Not in 20 years and probably not ever. In the short term, the replacement technology for all industries that burn fossil fuels doesn't exist. Longer term, the volume of fossil fuels being burned in the final couple of industries that still need them, plus the hobbyists keeping antiques running, won't matter.

There is little point in passing a law against something few want anymore. There are no laws that I know of against bloodletting, but nobody does it anymore because we now know it causes more harm than good. Leeches have been used in some recent techniques, but the reasons are very different.

By the time we reach minimal fossil fuel use, the total emissions from fossil fuels will probably be less than it was in the early years of the industrial revolution.

 

[EVNow]

I think it will be more than 10 years.

If self driving ever becomes real, it can transform the industry faster. They will be easier to regulate (no ICE robotaxis !) and the poor, the young and the elderly may not want to own cars at all. This sinks new ICE car sales and causes a glut n the used car market.

 

[mspohr]

And there are many poorly developed areas in poorly developed countries where electricity is not that common, even fossil fuels are not that easy to come by, but there are street vendors selling it by the bottle.

I think we need to rethink the situation in developing countries.
Small PV systems are easy to deploy and are supplanting kerosene lighting across developing countries. Kerosene is expensive and the economics are compelling.
I could see the same situation with EV replacing ICE.

 

[Mader Levap]

I expect burning of fossil fuels will be banned. Spewing toxins into air we all breath is not a right.

Nope. It will be redtaped to hell and back, though. At least for car owners.

And blanket ban on everything related to fossil fuel burn won't happen any time soon, if ever.

 

[wdolson]

If self driving ever becomes real, it can transform the industry faster. They will be easier to regulate (no ICE robotaxis !) and the poor, the young and the elderly may not want to own cars at all. This sinks new ICE car sales and causes a glut n the used car market.

For urban areas that's probably true, when it happens. I can see that in most of King County, but 2/3 of Clark County will not have much in the way of autonomous taxi service for a very long time.

We're longer from a self driving future than a lot of people think. First the software needs to be worked out. That's getting there, but it's far from proven. Then there will be multiple self driving systems out there. The auto makers might need to negotiate among themselves and work out a standard. A standard will probably be necessary for the next step which is to get regulators to approve these systems for use on all public roads. Then once all those hurdles are achieved, car makers need to retool and crank up production of these cars.

Then there is the fickle public and whether they want to do all their travel in autonomous taxis. With any new technology there are always unanticipated things that come up and this is a major shift. We could have hoards of Luddites put out of business by automated vehicles doing all they can to sabotage them.

But assuming the tech works, it's approved, and the public wants to use it, the total number of cars out there will decline. But that's going to take time to achieve. Removing mirrors from cars and replacing them with cameras has been feasible for more than 10 years and it's still only legal in a few countries. Self driving is higher reward, but much higher risk for both public safety and the disruption to the job market.

I think we need to rethink the situation in developing countries.
Small PV systems are easy to deploy and are supplanting kerosene lighting across developing countries. Kerosene is expensive and the economics are compelling.
I could see the same situation with EV replacing ICE.

Though to add EVs to the mix for developing countries requires larger EV systems. Most of the places that are developing are hot, sunny climates so PV systems will work well, but they still need to be produced and installed and that's going to take time.

 

[neroden]

So, the long-term fundamentals of Tesla.

I have only ONE concern now, and that's the appalling customer communications mess. If they get that straightened out -- and it's not rocket science, but it does require hiring people who have expertise and know how to do it, and not just treating it as if it's "easy" -- then everything else looks great to me.

Everything else seems to have been largely straightened out. All the other serious problems remaining (like USB music) are fundamentally due to the customer communications problems, IMO.

(Yes, they'll totally miss the timeline on "full self driving", probably by over a decade, but it doesn't matter because they're far ahead of everyone else.)

 

[EVNow]

(Yes, they'll totally miss the timeline on "full self driving", probably by over a decade, but it doesn't matter because they're far ahead of everyone else.)

There FSD will also be "geo-fenced" and work in places they prioritize (major metropolitan markets). I think it will take 3 to 5 years to get this working. Will take much longer for it to work "everywhere".

 

[EVNow]

For urban areas that's probably true, when it happens. I can see that in most of King County, but 2/3 of Clark County will not have much in the way of autonomous taxi service for a very long time.

It will have - but probably take longer to pickup.

Either way - imagine if robotaxis are available in major metros, that can reduce ICE sales by 20%. Which will drive all ICE companies to bankruptcy.

We're longer from a self driving future than a lot of people think. First the software needs to be worked out. That's getting there, but it's far from proven. Then there will be multiple self driving systems out there. The auto makers might need to negotiate among themselves and work out a standard. A standard will probably be necessary for the next step which is to get regulators to approve these systems for use on all public roads. Then once all those hurdles are achieved, car makers need to retool and crank up production of these cars.

I know Musk keeps saying this - but regulators don't care. They will just change the rules to let robotaxis drive the roads but companies take full liability (like they did in Florida).

 

[mspohr]

Tesla s approach to FSD doesn't rely on maps or local data so if it works somewhere, it will work everywhere.

 

[wdolson]

It will have - but probably take longer to pickup.

Either way - imagine if robotaxis are available in major metros, that can reduce ICE sales by 20%. Which will drive all ICE companies to bankruptcy.

Most main stream manufacturers will be making more EVs soon. The Europeans have two prods pushing them forward. Their domestic governments are demanding it coupled with competition from Tesla. The Japanese, South Koreans, and Americans are at least a step behind, but EVs will be battery supply limited when the demand shifts.

I expect Fiat-Chrysler to fail, and some other car companies might not make it, but I expect at least half the old car companies will survive the shift.

I know Musk keeps saying this - but regulators don't care. They will just change the rules to let robotaxis drive the roads but companies take full liability (like they did in Florida).

I'm not sure that's going to happen. Florida tends to be more cavalier about people's safety than many states. And all it will take is a few lawsuits with teeth when people get rundown by robo-taxis because of software bugs to make both the manufacturers and government regulators more cautious. If the responsibility is on manufacturers and/or operators of robo-taxis, the insurance will likely be sky high.

There will be bugs that kill people. That's certain no matter how much testing is done. It can be minimized, but it will happen.

Tesla s approach to FSD doesn't rely on maps or local data so if it works somewhere, it will work everywhere.

Elon has said there will need to be more precise GPS and more accurate road data to make FSD without a driver a reality. The data available today just isn't accurate enough for a computer to manage on its own. That mapping effort is going to take a while and it will be focused first on major highways (like GM's system), then high density cities, and finally get to rural areas.

 

[Doggydogworld]

Elon has said there will need to be more precise GPS and more accurate road data to make FSD without a driver a reality. The data available today just isn't accurate enough for a computer to manage on its own. That mapping effort is going to take a while and it will be focused first on major highways (like GM's system), then high density cities, and finally get to rural areas.

When did he say this? In the s/h meeting yesterday? I wasn't paying 100% attention. Well, actually about 5% attention, ha.

 

[wdolson]

When did he say this? In the s/h meeting yesterday? I wasn't paying 100% attention. Well, actually about 5% attention, ha.

It was about a year ago. I forget where, it might have been Twitter.

 

[mspohr]

The Tesla Autonomy Day presentation from 22 April 2019 would be the best resource for the latest info. Tesla autonomy seems completely focused on AI. I don't think there was any mention of collecting better maps and local data but I don't have the time to watch all four hours again.

Here's a brief TOC

Kent Hinson1 month ago (edited)
1:59:19 Teaching the machine with labels.
2:01:39 Teaching the machine with variety of data.
2:04:24 the world is strange. Using real data is better then simulation.
2:05:18 The three most important things for teaching neural networks.
2:06:07 Teaching object detection.
2:08:31 Neural net process. "Data Engine"
2:14:39 Path prediction around corners. "Paths it can't even see"
2:19:19 3D reconstruction using only video.
2:20:45 Neural net vision just as good as radar.

 

[Doggydogworld]

The Tesla Autonomy Day presentation from 22 April 2019 would be the best resource for the latest info. Tesla autonomy seems completely focused on AI. I don't think there was any mention of collecting better maps and local data but I don't have the time to watch all four hours again.

Here's a brief TOC

Kent Hinson1 month ago (edited)
1:59:19 Teaching the machine with labels.
2:01:39 Teaching the machine with variety of data.
2:04:24 the world is strange. Using real data is better then simulation.
2:05:18 The three most important things for teaching neural networks.
2:06:07 Teaching object detection.
2:08:31 Neural net process. "Data Engine"
2:14:39 Path prediction around corners. "Paths it can't even see"
2:19:19 3D reconstruction using only video.
2:20:45 Neural net vision just as good as radar.

Thanks. I watched it twice before, enough for one lifetime

Musk has called maps "brittle", though he does mention feature maps to help identify stop signs and such. I'd be stunned if he said they needed more precise GPS and better road data for FSD.

 

[wdolson]

Maybe Elon changed his mind about GPS. It wouldn't be the first time he said something then ignored it later. The one most people remember is the announcement the Model Y would have falcon wing doors.

 

[neroden]

I'm not sure that's going to happen. Florida tends to be more cavalier about people's safety than many states.

Highest pedestrian death rate in the nation, one of the higher car passenger death rates, one of the higher gun death rates

 

[dqd88]

I think it will be more than 10 years. With production of cars and lights trucks at about 100 million a year and a fleet on the road of around 1 billion, switching all car production to EVs tomorrow would barely build enough cars to meet that goal. And there are many poorly developed areas in poorly developed countries where electricity is not that common, even fossil fuels are not that easy to come by, but there are street vendors selling it by the bottle.

But I do think we'll reach a point where people will think anyone daily driving an ICE is like someone wanting to use one of those old brick cell phones from the 1980s. Once that point is reached ICE will become rare. The poorest will probably be the last to electrify. The infrastructure to support EVs will be installed in poor neighborhoods last and the poor who need cars will first be buying cast off relatively new ICE, then finally used EVs. The poor will be able to buy iCE cheap because nobody will want them.



Within 10 years I expect the public to generally be convinced EVs are the way to go. Being able to produce enough EVs (producing enough batteries) and being able to afford EVs will be the two barriers to adoption. Some governments might help the process by offering rebates to lower income people who buy EVs, but not all countries will do that.



There is little point in passing a law against something few want anymore. There are no laws that I know of against bloodletting, but nobody does it anymore because we now know it causes more harm than good. Leeches have been used in some recent techniques, but the reasons are very different.

By the time we reach minimal fossil fuel use, the total emissions from fossil fuels will probably be less than it was in the early years of the industrial revolution.

Govts can and should mandate EVs today to the extent that battery production allows. All other cars should be hybrid. That is feasible and would help reduce emissions.