TSLA Market Action: 2018 Investor Roundtable

[FS_FRA]

FWIW Frankfurt up ever so slightly (+0.72%) this morning, albeit on practically no volume.

 

[Fact Checking]

I don't think I replied to these FSD skeptical arguments yet, but I think FSD is important to Tesla's valuation and thus to its short term stock price (see my FSD prediction at the end of the comment):

I'm on record as saying full self-driving won't happen for at *least* five years more, maybe decades -- it's just much harder than people think it is.

And in support you listed a number of 'hard' scenarios that FSD AutoPilot won't be able to handle, one of them is blind curves/corners:

Honestly I'm not sure I've ever encountered a truly blind curve.​

Right. I have, there are three in my town.

You seem to agree that Tesla is pretty advanced in the 'vision' part of full self-driving, which creates a 3D scene of objects and velocity vectors. Your argument is that 'driving' in that 3D scene is the hard part and that Tesla 'hasn't even begun' that process yet.

I hope that's a fair summary of your arguments - let me know if it isn't!

I disagree with your arguments:

• Here's a video, at timestamp 1:34, what AutoPilot V9 sees in a more or less 'blind corner'.
• As you can see AutoPilot refuses to recognize a lane/road that's not visible enough yet.
• I believe the 'driving' logic will simply detect that scenario in your town and declare the road a dead end, stop the car and alert the driver - or if there's no driver alert a remote driver or route around it using another approach.
• Full self-driving doesn't mean the car has to be able to drive through every scenario a human would risk. Full self-driving is likely going to be a very, very paranoid and careful driver who is easy to scare into defensive driving and outright refusal to drive.

I.e. I believe the focus of the 'driving' part will simply be physical safety, which consists of the following elements:

• Can the car see the road. If not, it won't drive there.
• Are there any objects (cars, pedestrians, obstacles) on the road that prevent the car from driving there.
• If all criteria are met, drive.

FSD driving is the typical 'perfect is the enemy of good' problem. FSD vision, perception and reaction times have to be better than humans, but driving doesn't need to be perfect or even more efficient than human driving, it needs to employ safe driving - and refusing to drive in certain scenarios is an entirely accepted outcome.

You also seem to not take regional differences into account: road systems in the U.S. are much simpler, larger and generally safer (in terms of layout, not necessarily in terms of drivers) than those in Europe. If Tesla is able to tap FSD for the U.S. market only that's a huge chunk of revenue already.

Finally, you also don't seem to take customer differences into account: FSD will be most disruptive to the transportation/logistics business. But trucks are already limited in where they can drive - and it's quite probably that an 18-wheeler won't typically go on routes with blind corners, right? They'll stick to highways and easy to access logistics centers.

Just because FSD will initially have trouble mastering narrow Irish, Welsh or Scottish roads and towns without disengaging doesn't mean the valuation of Tesla's FSD business has to be zero.

It is simply not useful to make an "average driver" car. Psychologically, people won't accept it: we have documentation that people will demand that an autonomous car be *substantially better* than an average driver; for some reason people are way more unhappy if hurt by a mediocre-competence robot than by a mediocre-competence human. Musk knows this, and has quoted a need to be 10 times better and 100 times better at various times. You will not get this by replicating average. It may reduce crashes if deployed as "enhanced driver assistance" (people will accept it if there's a driver to blame) but it will never be allowed as "full self driving".

I believe here you are conflating 'safe' driving with 'efficient' driving, which are two different attributes.

FSD needs to be an order of magnitude safer than human drivers - and it's probably already at that stage or even better for highway driving, and AutoPilot is probably already a two orders of magnitude safer driver in bad weather (due to the radar that sees through rain and fog) and at night (due to 8 cameras that are sensitive into the infrared), and of course due to being 100% alert all the time and never tiring.

FSD doesn't have to be efficient at driving to be useful, at least initially. It's quite likely that human drivers will be able to master tricky driving scenarios a lot better than FSD software for many years to come. But the weird driving scenarios are not what matter: what matters are the 1 billion vehicle miles driven in the U.S. every year, and whether FSD can handle a meaningful percentage of those routes. It will be self-correcting after that: customers will self-select and those will adapt FSD faster who have higher utility for it.

TL;DR: As a contrast to your prediction 'of at least 5 years but possibly decades', I do think Tesla is going to release the first usable versions of FSD next year already: traffic light recognition, road sign recognition and maybe driving around corners that have working traffic lights, helped by GPS and maps.

It will still be a 'paranoid' FSD that falls back to the human driver frequently, but it will be able to drive autonomously in a surprisingly large percentage of the real routes that people take, and this percentage will increase with every new release in a largely organic, iterative process. I also expect there to be a real San Francisco to New York FSD road trip attempt by Tesla next year, with a very low number of disengagements.

 

[Carl Raymond]

Is it not becoming clear that Tesla has made an extremely significant tech breakthrough with this technology? A breakthrough that is probably in itself worth many billions when applied?

I feel this is yet another largely unnoticed moat TSLA has, and is not valued in whatsoever.

Yes, I think they are ahead. In tech a lead does not last long. Others can soon emulate the board and the unified two frame input.

What they can't copy is the number of hardware equipped cars on the road. Nor the volume of learning data Tesla has available. Nor the fact that Tesla's cars are electric with low low running cost - no ICE can compete in a no paid driver scenario.

And I'm sure we can all rattle off Tesla's other big competitive advantages. Giga, superchargers, no dealers, made to order, OTA, battery reputation, buy online without haggle.

 

[shlokavica22]

Apologies if it was already posted- Troy's take on the Lemur is quite fascinating - Tesla starts production of the Model 3 Mid Range - Teslike.com

The important bits:

• Based on my latest calculation, I estimate that the Model 3 MR battery pack has 3,648 cells compared to 4,416 for LR and 2,976 for SR.
  
• In terms of gross margins, after the latest price adjustments, the Model 3 MR has only 1.8% worse gross margins than the LR.

In other words, if the battery production is currently the limitation factor, MR can allow to make ~20% more cars with margin 13.2% (instead of 15% that the LR has).

 

[Alphacrux]

I don't think I replied to these FSD skeptical arguments yet, but I think FSD is important to Tesla's valuation and thus to its short term stock price (see my FSD prediction at the end of the comment):



And in support you listed a number of 'hard' scenarios that FSD AutoPilot won't be able to handle, one of them is blind curves/corners:



You seem to agree that Tesla is pretty advanced in the 'vision' part of full self-driving, which creates a 3D scene of objects and velocity vectors. Your argument is that 'driving' in that 3D scene is the hard part and that Tesla 'hasn't even begun' that process yet.

I hope that's a fair summary of your arguments - let me know if it isn't!

I disagree with your arguments:

• Here's a video, at timestamp 1:34, what AutoPilot V9 sees in a more or less 'blind corner'.
• As you can see AutoPilot refuses to recognize a lane/road that's not visible enough yet.
• I believe the 'driving' logic will simply detect that scenario in your town and declare the road a dead end, stop the car and alert the driver - or if there's no driver alert a remote driver or route around it using another approach.
• Full self-driving doesn't mean the car has to be able to drive through every scenario a human would risk. Full self-driving is likely going to be a very, very paranoid and careful driver who is easy to scare into defensive driving and outright refusal to drive.

I.e. I believe the focus of the 'driving' part will simply be physical safety, which consists of the following elements:

• Can the car see the road. If not, it won't drive there.
• Are there any objects (cars, pedestrians, obstacles) on the road that prevent the car from driving there.
• If all criteria are met, drive.

FSD driving is the typical 'perfect is the enemy of good' problem. FSD vision, perception and reaction times have to be better than humans, but driving doesn't need to be perfect or even more efficient than human driving, it needs to employ safe driving - and refusing to drive in certain scenarios is an entirely accepted outcome.

You also seem to not take regional differences into account: road systems in the U.S. are much simpler, larger and generally safer (in terms of layout, not necessarily in terms of drivers) than those in Europe. If Tesla is able to tap FSD for the U.S. market only that's a huge chunk of revenue already.

Finally, you also don't seem to take customer differences into account: FSD will be most disruptive to the transportation/logistics business. But trucks are already limited in where they can drive - and it's quite probably that an 18-wheeler won't typically go on routes with blind corners, right? They'll stick to highways and easy to access logistics centers.

Just because FSD will initially have trouble mastering narrow Irish, Welsh or Scottish roads and towns without disengaging doesn't mean the valuation of Tesla's FSD business has to be zero.



I believe here you are conflating 'safe' driving with 'efficient' driving, which are two different attributes.

FSD needs to be an order of magnitude safer than human drivers - and it's probably already at that stage or even better for highway driving, and AutoPilot is probably already a two orders of magnitude safer driver in bad weather (due to the radar that sees through rain and fog) and at night (due to 8 cameras that are sensitive into the infrared), and of course due to being 100% alert all the time and never tiring.

FSD doesn't have to be efficient at driving to be useful, at least initially. It's quite likely that human drivers will be able to master tricky driving scenarios a lot better than FSD software for many years to come. But the weird driving scenarios are not what matter: what matters are the 1 billion vehicle miles driven in the U.S. every year, and whether FSD can handle a meaningful percentage of those routes. It will be self-correcting after that: customers will self-select and those will adapt FSD faster who have higher utility for it.

TL;DR: As a contrast to your prediction 'of at least 5 years but possibly decades', I do think Tesla is going to release the first usable versions of FSD next year already: traffic light recognition, road sign recognition and maybe driving around corners that have working traffic lights, helped by GPS and maps.

It will still be a 'paranoid' FSD that falls back to the human driver frequently, but it will be able to drive autonomously in a surprisingly large percentage of the real routes that people take, and this percentage will increase with every new release in a largely organic, iterative process. I also expect there to be a real San Francisco to New York FSD road trip attempt by Tesla next year, with a very low number of disengagements.

I’ve been curious about what happens on the Tesla network when deteriorating weather conditions force the fsd car to simply pull over and drop driving. Would tesla require a licensed and capable driver to take over? This would seem to eliminate using fsd for the elderly who can’t drive, for example. I suppose this is acceptable at first, fsd has to start somewhere. But then this wouldn’t be true level 5, only level 4.

In the spirit of the title of the forum, I see we are up about 4 points pre-market. This seems slightly excessive for the amount that Dow futures are up? Any other news?

 

[Tslynk67]

Gave you a funny because here in Germany speed differences often are *much* greater. But I (and scientists looking at efficiency of traffic flow and safety) agree that it is better to keep them low - and much more relaxed if when passing, you don´t have to expect someone coming from behind at 50mph faster than you.

Yeah, you have to keep your wits about you driving on the German autobahns. Some old giffers going 120, other nutters 250 - needs sharp reactions at times, and 10x more so at night.

I find a speed 140-160 gives best results, but the ability to accelerate up to 200 in an instant, just to annoy the expensive ICE cars looming in the rear-view mirror, is too tempting not to deploy.

 

[Fact Checking]

Yes, I think they are ahead. In tech a lead does not last long. Others can soon emulate the board and the unified two frame input.

What they can't copy is the number of hardware equipped cars on the road. Nor the volume of learning data Tesla has available. Nor the fact that Tesla's cars are electric with low low running cost - no ICE can compete in a no paid driver scenario.

I agree, in fact I think Tesla's "FSD and ride-sharing moat" has even more elements to it:

• Tesla's AI Chip: Competitors cannot easily copy the AI chip that is already 10 times more power-efficient than one of the top providers of such solutions, NVidia. Designing a new chip takes at least 2-3 years - but the cycle is often longer. So for years they won't be able to emulate Tesla's V10 neural network layers in real-time on a real vehicle.
  
• Tesla's labeled training data: competitors cannot easily copy the labeled data (images) that Tesla's AutoPilot division collected and created in the last 5 years or so. This is a huge database of probably hundreds of thousands (maybe millions) of images, carefully categorized. This is the 'secret sauce' that trains the vision network to recognize objects and weird scenarios.
• Tesla's training hardware: It's unclear whether Tesla's AI hardware includes back-propagation support: if it does then their own hardware will also allow much more power-efficient (and thus cost-efficient) training of their networks. As the size of neural networks increases the cost of training increases exponentially, so there's a true "wall" that makes training performance a real issue. Karpathy is very focused on two issues: performance and the process that creates and maintains the labeled data, and performance is a key issue both on the vehicle platform and on the back-office platform that trains the networks.
• Tesla's lack of LIDAR: Most of Tesla's competitors (Waymo, Ueber, Lyft, etc.) took the unnecessary detour of using LIDAR data. This pollutes their neural networks and their development teams: there's a lot of inertia and vested interest in keeping LIDAR - but in the end it's a dead-end and a significant cost. This might take years for them to sort out internally.
• Tesla's "shadow mode" testing: Most competitors rely on human test-drivers and hundreds of cars to train their self-driving solutions. Tesla can deploy their new networks on their fleet and test the new network in 'shadow mode', and compare simulated AutoPilot reactions with how the human driver (or the active version of AutoPilot) takes. This speeds up training, testing and safe deployment enormously.

And, to make the list complete, here's the ones you already listed:

• Tesla's installed base/fleet: A fleet of over 300,000 HW2 or later vehicles already includes all the cameras, radar and sonar sensors for practical FSD. This not only gives Tesla an easy customer base to sell an FSD solution to, but it is also a feedback loop to improve the networks.
• Tesla uses their market leading EV as a ride-sharing/renting fleet: A fleet of EV vehicles that are the most power efficient mass manufactured passenger EVs on the planet, with a component count a fraction of that of a comparable ICE vehicle (~40k parts versus ~10k parts in a Model 3) has both much lower "fuel" costs and lower maintenance costs. So for ride-sharing and automated vehicle loans Tesla has a superior price point that ICE-fleet competitors such as Ueber or Lyft won't be able to match.

 

[JBRR]

Panasonic second-quarter profit down 15 percent as investment in Tesla batteries weigh
https://www.panasonic.com/global/corporate/ir/pdf/2018_2q/2q_financial_results_e.pdf

Looks like Panasonic incurred additional opex because they had to ramp battery production at Gigafactory 1 faster than initially anticipated.
Guiding to reach 35GWh capacity by end of Fiscal Year 2019. (March 31st 2019 is end of FY19 if I'm reading their financial calendar correctly. Consolidated Financial Results and Supplemental Financial Data for Fiscal 2018, the year ended March 31, 2018 | Headquarters News | Panasonic Newsroom Global)

 

[Fact Checking]

In other words, if the battery production is currently the limitation factor, MR can allow to make ~20% more cars with margin 13.2% (instead of 15% that the LR has).

Note that the LR already had margins of 20.5% in Q3, which included 11k vehicles made in Q2.

So LR margins will further improve in Q4, getting close to the 25% target, and if Troy's calculation is correct then MR could likely have margins over 20% in an average configuration, and over 15% even at the base price.

(Note that Model 3 cash margins are even better, about 5% higher than the GAAP margin. So increasing automotive revenue will have a disproportionately positive effect on cash flow as well.)

 

[Fact Checking]

Looks like Panasonic incurred additional opex because they had to ramp battery production at Gigafactory 1 faster than initially anticipated.

BTW., this should be a reply to those people who asked "why didn't Panasonic ramp up faster??". The reason is that it's very expensive to expand cell manufacturing, and Panasonic has to front these investment costs. (Tesla leases it from Panasonic and finally ends up owning the lines (edit: this part is speculation, maybe they don't end up owning the lines - see the followup post further below), but that takes years of lease payments.)

So Panasonic had to be conservative in their ramp up: they only committed to expanding cell capacity one Tesla demonstrated a Model 3 week with 5,000 vehicles produced.

It's all good now: Panasonic will be sharing the cash Tesla makes on all these cells in the quarters and years to come.

 

[Fact Checking]

I’ve been curious about what happens on the Tesla network when deteriorating weather conditions force the fsd car to simply pull over and drop driving. Would tesla require a licensed and capable driver to take over?

My expectation would be:

• Initially FSD will require a licensed driver to be physically present, probably with hands on the wheel required.
  
• Eventually, once the legal/regulatory environment permits it, there would be "remote drivers" which would basically be a service you could hire from Tesla: the remote driver takes over if there's some trouble, and there's also a network of FSD-service cars that physically drives out to any permanently stuck FSD car. This would be similar to roadside assistance services, a bit higher cost because FSD troubles are expected to be much more frequent initially than mechanical failure. Pricing might initially be also 'per incident' or 'per minutes driven', not flat.
  
• I'd also expect fleets to be mindful of weather predictions: if there's a tornado warning they would take cover or avoid risky cells. If there's heavy snow cover or new snowfall forcasted they'd warn owners, or if there's sleet or icy conditions expected, etc.
  
• There might be some 'no FSD drive' geo-fenced areas FSD cars cannot enter without a driver present: areas with poor cellular service or known-tricky sections of road. Most urban environments (where ~80% of the vehicle miles are generated) would be fine.

I.e. I think it will gradually improve, with professional 'remote drivers' seamlessly helping along FSD cars. Customers driving in the car might not even be aware of who is currently driving the car (although the car might chose to inform them).

As the FSD software gets better, the 'remote driver' service would cost less and less, and in the end it might be a flat pricing structure - and maybe even included in the price of the FSD option itself.

If you rent out your car to Tesla's ride-sharing network then the remote driving service would be part of the ~30% fee Tesla takes.

 

[X Fan]

Tha

BTW., this should be a reply to those people who asked "why didn't Panasonic ramp up faster??". The reason is that it's very expensive to expand cell manufacturing, and Panasonic has to front these investment costs. (Tesla leases it from Panasonic and finally ends up owning the lines, but that takes years of lease payments.)

So Panasonic had to be conservative in their ramp up: they only committed to expanding cell capacity one Tesla demonstrated a Model 3 week with 5,000 vehicles produced.

It's all good now: Panasonic will be sharing the cash Tesla makes on all these cells in the quarters and years to come.

Thanks for the update. Can you please advise the lease terms between Panasonic and Tesla?

 

[jeewee3000]

TL;DR: As a contrast to your prediction 'of at least 5 years but possibly decades', I do think Tesla is going to release the first usable versions of FSD next year already: traffic light recognition, road sign recognition and maybe driving around corners that have working traffic lights, helped by GPS and maps.

I agree with many of your arguments, but FSD is definitely more than one year away.

Life gets in the way of theory crafting.

Your argument basically adds up to: "current Tesla tech + Tesla's NN chip early next year + shadow mode on growing amount of Tesla vehicles = Neural Net can figure FSD out exponentially fast."

As @Lycanthrope pointed out, the current Autopilot and maps are best in the U.S., and you can expect that to remain that way for some time. If the coast-to-coast demo (U.S.) releases by let's say summer 2019, the U.S. might get a U.S.-only beta of Navigate on Autopilot from start to finish by end of 2019.

In fact that wouldn't be FSD as it's not applicable in Europe, Asia or anywhere else. Road systems are indeed way more complex in medieval towns that have true BLIND corners that are just wide enough for one small vehicle to pass through at the apex of the bend (the rest of the road is two lanes). If the system shuts down in those edge cases it is not FSD. The system should learn to stop, sound the horn, listen for horns from the other side, watch for headlight flare from the other side, and then take the turn in a decisive manner as to not stay in the danger zone for too long.

TL;DR:
- Will Tesla release beta systems with more and more autonomy by the end of next year? Most likely, yes.
- Is Tesla ahead in the FSD-race? I believe they are, yes.
- Is FSD happening within five years, let alone one? Not so likely.

Bonus: What's in a name. I basically disagree with your definition of FSD.

 

[Fact Checking]

Tha


Thanks for the update. Can you please advise the lease terms between Panasonic and Tesla?

This was pointed out by @brian45011 in this comment, quoting the 10-Q:

"Panasonic has partnered with us on Gigafactory 1 with investments in the production equipment that it uses to manufacture and supply us with battery cells. Under our arrangement with Panasonic, we plan to purchase the full output from their production equipment at negotiated prices. As these terms convey to us the right to use, as defined in ASC 840, Leases, their production equipment, we consider them to be leased assets when production commences. This results in us recording the cost of their production equipment within property, plant and equipment, net, on the consolidated balance sheets with a corresponding liability recorded to financing obligations. For all suppliers and partners for which we plan to purchase the full output from their production equipment located at Gigafactory 1, we have applied similar accounting. As of June 30, 2018 and December 31, 2017, we had cumulatively capitalized costs of $955.4 million and $473.3 million, respectively, on the consolidated balance sheets in relation to the production equipment under our Panasonic arrangement. We had cumulatively capitalized total costs for the Gigafactory 1 of $4.14 billion and $3.15 billion as of June 30, 2018 and December 31, 2017, respectively."​

So if this is is capitalized by Tesla and expensed as a capital lease, not as an operational lease, then that would suggest a lease-to-own construct.

I might be wrong about that though. @brian45011's comment also includes links to two contracts that define the Gigafactory 1 arrangement.

 

[bhtooefr]

I haven't heard of tesla making vans. Where did this rumor start?

Elon Musk himself: Elon Musk says Tesla likely to build an all-electric cargo van on its… (link to an Electrek article on archive.is (Fred doesn't get the clicks))

I'd think it's way low on the priority list, way below Y, semi, and pick up truck. I guess Amazon does use those Sprinter vans...and some local chain grocery stores use those little commercial vans so there is some demand. I worked for a new car dealer who carried those commercial van models and we hardly sold any of them, ever. I don't think I even witnessed one sold over a 4 month period. The model Y will be a game changer though. Crossovers were easy sales.

I see tons of Ford Transits (and Transit Connects, but that's something smaller), and I used to see a fair amount of Mercedes Sprinters (although the Transit being cheaper displaced them). Note that the vans mostly get sold through fleet sales, so things may look very different for how they're sold.

And, it'll likely be a simple variant of the pickup, so it could probably be released alongside it.

 

[Fact Checking]

Your argument basically adds up to: "current Tesla tech + Tesla's NN chip early next year + shadow mode on growing amount of Tesla vehicles = Neural Net can figure FSD out exponentially fast."

No, there's a crucial difference in how I frame the argument:

• Neural networks will be limited to 'perception', i.e. vision and spatial awareness, and maybe NNs will have supporting role in alerting the FSD-driving system about common sources of danger.
• The FSD driving software is not going to use neural networks to drive: Tesla would be crazy to do that. Neural networks are non-deterministic to a large degree, it's not possible to review/validate/audit them in the usual sense of the word, etc.
• Instead I believe Tesla's FSD driving software uses regular procedural logic and deterministic programming - similar to how SpaceX's self-landing rockets are using real-time software to land their rockets safely. These are basically control loops with a very advanced notion of the spatial position of the vehicle compared to the 3D scene, including probable future motion of other vehicles and persons. This is non-trivial but mature code that has gone through several revisions already, which is more than 5 years old, and which is iteratively enhanced to support FSD routes. This is the software that drives the car in AutoPilot already.

I.e. I don't think Tesla is going to 'train' their networks to 'drive'. Neural networks are for vision and perhaps hazard precognition, but not for driving.

I.e. the 'driving' portion of FSD is already mostly done: it's their current on-ramp-off-ramp "Navigate on AutoPilot" software in V9, extended with logic related to traffic signs and traffic lights, etc.

Bonus: What's in a name. I basically disagree with your definition of FSD.

Well, I define 'FSD' in the way Tesla has always framed this problem: the ability of the car to go from A to B without the driver having to take action. In that sense on-ramp-off-ramp Navigate on AutoPilot will already be de-facto FSD version 0, once the turn-signal confirmation can be turned off.

I.e. it's a statistical feature: its utility depends on the number of disengagements per given route. When FSD is introduced in such a fashion it's obvious that it can be introduced gradually, iteratively, on a regional basis, etc. This is also what matters most to the economics of FSD.

I.e. I understand FSD to be "level 4" autonomy or a close approximate, which is self-driving with a number of important qualifiers which can include 'human presence'.

If you define the 'full self driving' as 'fully equivalent to and much better than a human driver' then we won't reach such level of self-driving for decades: probably not until the AI in the car achieves legal personhood and can drive wherever she desires to.

But in terms of Tesla's valuation (which this 'Market Action' thread is rumored to have some connection to) what matters is full self-driving economics.

 

[FS_FRA]

Elon Musk himself: Elon Musk says Tesla likely to build an all-electric cargo van on its… (link to an Electrek article on archive.is (Fred doesn't get the clicks))



I see tons of Ford Transits (and Transit Connects, but that's something smaller), and I used to see a fair amount of Mercedes Sprinters (although the Transit being cheaper displaced them). Note that the vans mostly get sold through fleet sales, so things may look very different for how they're sold.

And, it'll likely be a simple variant of the pickup, so it could probably be released alongside it.

In Europe and Asia the small vans (Ford Transit, Mercedes Sprinter, Mercedes Vito, Renault Trafic, Toyota HiAce, Hyundai, VW Crafter, etc) are hugely popular for last mile and inner-city delivery (not to mention inter-city passenger carrying people carriers). For sure there is a significant market out there for Tesla to address - eventually.

I do however think that Tesla should not try to address every niche market out there (see the ridiculous model range from Mercedes, VW, and BMW as example).

 

[vigleik]

• S&P 500 inclusion requires four profitable quarters in a row by some sources, five by others, i.e. it would happen around this time next year or early 2020. Current inclusion criteria used by the S&P 500 committee seem to be somewhat opaque, but I don't think there's a rule to have a profitable year, is there?

Sigh. We’ve been over this literally dozens of times on this thread. S&P 500 inclusion does not require four profitable quarters in a row. The sum of the last four quarters, as well as the last quarter, “should” be GAAP profitable. (And there is some disagreement about what “should” means.)

By now it’s highly probable that the conditions will be satisfied after Q1 2019 results, so May of next year or so.

Some websites are misinformed, so to verify you better go straight to the source.

 

[Anstandswauwau]

Eventually, once the legal/regulatory environment permits it, there would be "remote drivers" which would basically be a service you could hire from Tesla: the remote driver takes over if there's some trouble, …

I'd argue that relying on remote drivers to achieve AlmostFSD would be an easy, yet expensive shortcut similar to what LIDAR is right now: It's an unnecessary in-between solution you'd rather want to skip entirely if you can. Why expensive? Well, I don't think this would scale economically at all – unless maybe, you'd figure out a very sensible way to crowdsolve it.

… and there's also a network of FSD-service cars that physically drives out to any permanently stuck FSD car. This would be similar to roadside assistance services, a bit higher cost because FSD troubles are expected to be much more frequent initially than mechanical failure. Pricing might initially be also 'per incident' or 'per minutes driven', not flat.

That approach would even scale worse. Plus, that's like financially punishing your customer for every bug they discover. You really don't want to do that.

No, the proper way to do it is to bring the driver-intervention-rate just all the down and to fallback to the driver if necessary. Stop the car if needed (by the way, you'd even have to do this with the remote driver-approach). Once the intervention-rate is low enough, every FSD disengagement won't be seen as an annoyance but rather as a share-worthy black swan event.

 

[shlokavica22]

Jaguar Land Rover says sales plunged 13.2% to 129,887 vehicles for the three months to 30 September.

The UK’s largest car maker reported revenues of £5.6bn and a pre-tax loss of £90m on the back of lower sales.