Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[Dan Detweiler]

Completely off topic here, but it is Saturday.

Wow! Absolutely LOVE the new Hold feature on 36.1! Just feather the accelerator as you approach a stop to gauge your deceleration and then just let up to come to a complete stop. BRILLIANT! Took me about 5 minutes to get used to. The brake pedal is now nothing but an emergency stop pedal!

As for the increase in power, the car does seem to pull harder when launched. Could be my imagination though. We'll see what real world comparisons say.

Unfortunately, the new construction cone recognition appears to be just for hardware 3 cars. I have 2.5 and saw nothing as I passed several construction barrels.

If people only knew what they were missing by not owning a car that improves with age. Stock would be about $600 if they did!

Dan

 

[⚡️ELECTROMAN⚡️]

Its got legs!!!
Tesla Batteries Investigated for Possible Defects

https://www.washingtonpost.com/tech...officials-probe-alleged-tesla-battery-defects

NHTSA, Investigating Tesla Fire Reports, Demands Data on Battery Software Changes

This seems really concerning. I’m thinking these news article will adversely affect the stock price next week. Am I just being paranoid?

 

[kbM3]

Completely off topic here, but it is Saturday.

Wow! Absolutely LOVE the new Hold feature on 36.1! Just feather the accelerator as you approach a stop to gauge your deceleration and then just let up to come to a complete stop. BRILLIANT! Took me about 5 minutes to get used to. The brake pedal is now nothing but an emergency stop pedal!

As for the increase in power, the car does seem to pull harder when launched. Could be my imagination though. We'll see what real world comparisons say.

Unfortunately, the new construction cone recognition appears to be just for hardware 3 cars. I have 2.5 and saw nothing as I passed several construction barrels.

If people only knew what they were missing by not owning a car that improves with age. Stock would be about $600 if they did!

Dan

Absolutely not off topic. Very relevant to an investment and the general Tesla thread. In addition to your favorable opinion many others are chiming in. A lot of comments on Reddit are saying that auto pilot is greatly improved with this release, in addition to echoing your opinion on power and regen. Just seems like more and more good news lately.

 

[LJS22]

Completely off topic here, but it is Saturday.

Wow! Absolutely LOVE the new Hold feature on 36.1! Just feather the accelerator as you approach a stop to gauge your deceleration and then just let up to come to a complete stop. BRILLIANT! Took me about 5 minutes to get used to. The brake pedal is now nothing but an emergency stop pedal!

As for the increase in power, the car does seem to pull harder when launched. Could be my imagination though. We'll see what real world comparisons say.

Unfortunately, the new construction cone recognition appears to be just for hardware 3 cars. I have 2.5 and saw nothing as I passed several construction barrels.

If people only knew what they were missing by not owning a car that improves with age. Stock would be about $600 if they did!

Dan

That could be a good sign if software specific code is being written for HW3. Elon talked about staying on HW2.5 until they could take advantage of the new HW. Something along the lines of being on 2.5 would be better due to the software being written for it.

 

[ZeApelido]

Welp the FUD has Peak Stupid

The Insane Shortsightedness of Buying a Tesla

 

[WarpedOne]

I'm not giving them a click..

 

[Mike Ambler]

The only place Tesla can easily build factories is China. China simply won't allow Tesla to dominate their EV market.

Guys,
We want to build 4 more gigafactories...export only!
BTW we will be buying 90% from your local suppliers.
WDYR?
Elon

 

[KarenRei]

This seems really concerning. I’m thinking these news article will adversely affect the stock price next week. Am I just being paranoid?

So strange reading that thread, how thrilled they are with themselves for trying to hurt the company's public image.

I've long ago lost any sympathy I ever had for them. I sincerely hope they never see a penny from Tesla, and get stuck with a bunch of legal and court costs.

 

[The Accountant]

A Tesla article in a Florida publication.
Here's a summary:
- Author does not like Telsa because he does not like Elon
- Author drives a Tesla and falls in love with it.

From the article:
"I have to admit I was not a fan. That Elon Musk seems like a bit of a whack job, what with his spaceship company and his plan to colonize Mars....."

"That was the moment I fell in love. There I was, driving down the interstate at 70 miles an hour, not using my hands or feet, and it was exhilarating. I was in the future."

Mr. Chatterbox Test Drives a Tesla and Sees the Future

 

[The Accountant]

This seems really concerning. I’m thinking these news article will adversely affect the stock price next week. Am I just being paranoid?

Tesla will deliver over 105,000 cars in Q4, showing a bottom-line profit (again!), increased margins (again!) and tons of free cash flow (again!) on the heels of the Pick-Up reveal and in the midst of G3 China deliveries in Q1 with Model Y arriving possibly by late Q2.
What's concerning? batteries something something...what?

 

[Remus]

Absolutely none of that is really correct. At this moment in time, Tesla has the best battery technology available. But in just a few years that can easily change. There is no reason to believe Tesla can ramp up car production while major companies 20 times their size can't ramp up battery production.

The major auto makers will spend HUGE amounts of money on designing the best batteries because they CAN. They have the money and will spend it because they are realizing they can make good EVs or they can become the 21st century Kodak.

You also need to realize that a small difference in battery technology won't make such a large difference in sales. Cars are sold through marketing. Small differences in technology are easily mitigated.

You mentioned Kodak yet your entire post is contradictory with history that tells us time and time again they can't.

Toyota and Honda made killings in the U.S. by building more reliable cars. Their ads would be useless if they didn't build their reputation. Small differences in engineering results in big difference in longevity of the cars, which is the fundamental reason they command a price premium now. The big three used to be much bigger than Toyota, not now.

Historically companies 20 times larger almost never successful in changing their business. You mentioned Kodak. So the people in Nokia Ericsson Motorola are idiot who never heard of Kodak?

Heard of Sony and their legendary Walkman? What's their defense against the iPod?
When Steve Jobs returned to Apple, what's Sony's deployable capital comparing to Apple's? Probably more than 20times! Apple was on the death bed yet Sony was the top brand in consumer electronics.

Have you heard of RCA? Go read about it.

Stop dreaming about those 20 times larger companies can wake up one day. They weren't sleeping at all. They are very good at what they do but incapable of changes. History has shown time and time again they can't.

 

[StealthP3D]

This seems really concerning. I’m thinking these news article will adversely affect the stock price next week. Am I just being paranoid?

It might temporarily impact the share price but this is the kind of "nothing-burger" the market will quickly shrug off. Tesla has had hundreds of them over the years and, in the bigger picture, they are inconsequential. In the immediate picture, it certainly can't help but it's only one small impact of many potential influences. The shorts certainly hope it will induce people to sell but I don't know anyone who would be afraid of something so inconsequential.

 

[ReflexFunds]

This being an investing thread, is that scenario - FSD technology available and widely provided regulatory approval in '20 or '21 an important component of your Tesla valuation / investment thesis?

More generally, I'm wondering who else that posts in the thread has this as a serious or even critical component of their valuation (as in - Tesla's valuation is only going to go up bigly if it's true, and it's about right or overvalued if it's not true).


For me, with what I've personally experienced of Tesla's driver assist (omg - I love it), what I know about the world of data science / machine learning (it's my job, but I don't work in or have experience with vision systems), and what I think about how hard it's going to be to get cars driving themselves around unattended (technically, and regulatory approval), if my investment thesis / valuation was dependent on a 20/21 availability, then I'd exit my Tesla position approximately immediately.

The risk of failure of that outcome is too high - not commensurate with the reward.

My personal investment thesis doesn't have driverless vehicles running around to collect the next fare for 10 years, and results in a 10+ year buy and hold investment window (where 10+ is probably shorthand for "until I die"). If actual driverless cars happens sooner than 10 years, then that's a lot of gravy

If Tesla had no Autopilot or Robotaxi business plan, I think Tesla would still be a great investment and I would still be invested.
If Tesla fails to ever deliver on Robotaxi's I think it's advanced driver assist Autopilot features will still be an extremely valuable product which will save a huge amount of lives.
That said, Robotaxis make up the majority of my Tesla current valuation (which is far higher than the current market cap), despite what looks like almost zero value attributed to it in the current market cap. People often make the mistake of thinking the most probable outcome is the most important outcome or the only outcome worth focussing on. This is a big mistake, the most important scenario is the one with the largest expectation value (whether that is positive or negative). Future value created (or destroyed) * probability of getting there. My valuation model consists of hundreds of probability weighted scenarios for future cash flows.

I also work with machine learning, and Tesla's strategy looks to me like by far the most likely to deliver on a real commercial Robotaxi business at scale. Tesla is betting that self driving is a really difficult problem to solve and needs 10s of billions of miles of real world driving experience, carefully filtered and processed, to reach an accuracy 5-10x better than humans (to begin with). Everyone else is betting that self driving is a relatively easy problem to solve and can be achieved instead with very expensive hardware, simulation (relying mostly on the imagination of software developers) and just 10s of millions of miles of real world driving experience.

At this stage it still looks relatively unlikely, yet possible, that Tesla delivers a feature complete unsupervised FSD 5-10x safer than humans within the next two years. They have made huge progress on the march of 9s on Navigate on Autopilot just with HW2. The real test will be just how much of a step change improvement they can achieve when they start making use of HW3 and the larger neural nets. In terms of regulation, I think Tesla will receive approval in at least some jurisdictions fairly soon after getting unsupervised safety to 5-10x. Approving Robotaxis is essentially the trolley problem - do you divert the trolley and kill one person instead of leaving it on course to kill 5-10. In surveys 90% of people would make the switch in the trolley problem. Out of the hundreds of different potential jurisdictions that could approve Robotaxis, it seems highly unlikely to me that every single one would chose to let the 5-10 people die, running their regulator based on emotion rather than statistics, or else are all owned by the trucking/taxi/auto lobby. At the point where Tesla knows its software is safe enough, it is worth them spending a huge amount of money on any necessary lobbying/standard corporate bribery to get approval over the line.

While the probability of delivering Robotaxis in the near future is still low, if Tesla does deliver on Robotaxis, it is game over. Tesla has won. They are almost certainly the most valuable company in the world and preventing runaway global warming in time will suddenly look very achievable.

 

[jerry33]

INC ARTICLE:
Let's get this clear from the start. I want a Tesla and have seriously considering buying one. I've also stated publicly and repeatedly in this column that Tesla will completely disrupt the automobile industry, and that the established car makers are incapable, culturally, of making a comparable electric car. And I generally write very positively about Elon Musk, although I don't always agree 100 percent with him.

Despite all that, however, I have decided NOT to buy a Tesla, not because I don't think it's a great product but because I believe that what makes it great today will turn it into a headache tomorrow. Allow me to explain.

This entire article belies the experience of actual Tesla owners. Just a stupid hit piece preying on those who are insecure with technology.

 

[j6Lpi429@3j]

it boggles my mind that people think waymos mapping could ever be the way to go. How does mapping cope with the unexpected, like a fridge falling off the back of a delivery truck? or a tree blown over by a storm last night? it CANNOT.
Waymo is typical wework.magic leap style venture-capital scam nonsense. A great way to wow the odd technologically illiterate VC and get them to fund you to the next level. A TERRIBLE and unworkable technology for achieving what they are pretending they are working towards.

 

[Maitri982]

Sure. But hacks are not trivial either and they were constrained with compute and number of cameras. But anyway, would you claim that what they showed earlier this year was also a hack? According to Elon’s and Karpathy’s tweets it seems that they struggle with parking lots(Summon) and nonstandard traffic lights. Well even with HW2 neural networks they seem to do parking lots reasonable now and we have been seeing those blue traffic lights in so many presentations now that I assume they have made some progress on them by now.

I have heard so many people being skeptical of FSD. I get it, what they have release to customers have been overpromising and underdelivering. But the HW2/HW3 regulatory approval delay is outside of their control and we are looking at totally different compute, inputs and outputs and datasets. And likely different control also. Imo we should be very careful with judging HW3 by what we have seen of HW2.

I don't know if it was a hack... It did follow single route. But given that this time they didn't just show video that was likely assembled after many attempts and this one had passengers in it, it was certainly more impressive. Without looking at code I cannot say whether there was some hack to make that specific route work better or not.... And that's the problem with things like that. We need to see it in many indeterminate situations to know how good it is.

I am huge Tesla fan and want them to have successes at every level, but I am also Tesla owner and experienced in machine learning and I know what they are up against.

I think it's likely that for many years we will have something that is like 99 - 99.9% of the time good. That would be incredible and would save lives and help them sell many cars... But you will still have to pay attention... So it's not true FSD yet. I am hoping even that very good level of functionality comes in a couple of years we it will make driving much more relaxing like NOA does in some situations.

 

[mongo]

it boggles my mind that people think waymos mapping could ever be the way to go. How does mapping cope with the unexpected, like a fridge falling off the back of a delivery truck? or a tree blown over by a storm last night? it CANNOT.
Waymo is typical wework.magic leap style venture-capital scam nonsense. A great way to wow the odd technologically illiterate VC and get them to fund you to the next level. A TERRIBLE and unworkable technology for achieving what they are pretending they are working towards.

I don't agree with the mapping approach (scaling and brittle issues), but lidar is good at detecting obstacles so the only issue in your scenarios is making path planning software to navigate around the fridge or branch.

To @ReflexFunds point. It is not just miles collected, it is the ability to test and validate with the huge and growing fleet with OTA improvements. How can anyone else release a general case driving system when that version is the only one a cars gets?

 

[KarenRei]

I don't agree with the mapping approach (scaling and brittle issues), but lidar is good at detecting obstacles so the only issue in your scenarios is making path planning software to navigate around the fridge or branch.

To @ReflexFunds point. It is not just miles collected, it is the ability to test and validate with the huge and growing fleet with OTA improvements. How can anyone else release a general case driving system when that version is the only one a cars gets?

It's more like... no need for visual recognition of edge cases; we can just trust maps! Let's take this bridge, for starters! No need to learn to recognize road deck damage.

Map says the bridge is here, too, I can see the ground, and my radar even sees the reflection off the rebar!

This map shows no spike strip here!

What military checkpoint? I'm going to just keep driving through without stopping.

Temperature's getting a bit warm on this route; I'll just turn up the AC.

DANGER! OBSTACLE IN THE ROAD! SLAM ON THE BRAKES!!!

 

[woodisgood]

Welp the FUD has Peak Stupid

The Insane Shortsightedness of Buying a Tesla

For those that did not click, article basically says, "I like Tesla, but eeeeeeeek, the future is scurrrrrry!"

 

[EVNow]

it boggles my mind that people think waymos mapping could ever be the way to go. How does mapping cope with the unexpected, like a fridge falling off the back of a delivery truck? or a tree blown over by a storm last night? it CANNOT.
Waymo is typical wework.magic leap style venture-capital scam nonsense. A great way to wow the odd technologically illiterate VC and get them to fund you to the next level. A TERRIBLE and unworkable technology for achieving what they are pretending they are working towards.

That is not how Waymo works.

They use Lidar to create HD maps and locate the car within the 3D space when driving with high confidence. That doesn't mean they don't use vision. They use vision for all kinds of things Tesla NN is yet to learn. For eg. they can recognize and follow the directions of the person giving hand directions.

The basic fight is now about who can get there faster ...
- Waymo expanding their L4 system to top 100/200 cities in the world
- Tesla getting from L2 system everywhere to L4 system everywhere

(edit :

NHTSA : What NHTSA has given guidance in terms of FSD.
Waymo + : Features Waymo lists as targets, in addition to NHTSA features.

Tracking FSD Feature Complete
)