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

[MP3Mike]

Here is a nice video discussing expectations of autonomous vehicles:

 

[DaveT]

Others have given a lot of possibilities … but I'd think about it like this.

They needed to give a number to do the calculations around how much money TN can make. So, I'd look at 38k as a very rough estimate of what they need to do - larger battery, more miles, ruggedized interior, more communication hardware etc. If they are more than rough estimates, then, EM is really serious about getting FSD working next year.

Perhaps $38k is what they think the Model 3 long range model customized as robotaxi will cost at the end of next year.

 

[Krugerrand]

I wonder what shadow mode learns from my driving. I use the whole lane and more, depending on traffic. If there is a pot hole or even rough pavement, I avoid it while still staying in my lane. If there is a bicycle, I give it wide berth if there is no oncoming traffic. One thing that I don't like so much about NOA and TAAP is its unremitting commitment to the center of the lane.

Not if there’s a rabbit in the road.

That right there is funny on more than one level, ‘Thumper’.

 

[Carl Raymond]

Elon also mentioned special tire design and customized to be highly efficient. Tires could be 2x more expensive, but last 3x miles for example.
He did say Model 3 based.
Maybe rear seat screens would be nice.

This one is interesting. Efficient tyres wear less, but that also means less grip. The safest tyre design is very soft, but it would wear out in an unaffordable short time. There’s a spectrum spanning from softest to longest lasting.

It’s not unreasonable to assume that a car that (almost) never causes accidents can move along this spectrum a notch. Especially as the percentage of other cars on the road which almost never cause accidents starts to increase.

 

[EVNow]

You can't solve vision 100%. At least not with neural nets.
Lidar can detect things your vision NN misses, or misidentifies. It's just an extra layer. It costs too much for consumer cars and you can't sell 50-100k performance sedans festooned with bulbous protrusions. But the cost is negligible for a robotaxi that can generate 30-50k/year of revenue. And nobody cares if their robotaxi has obviously visible safety equipment. In fact, most would prefer it.

Personally I don't have Lidar and I can drive fairly well. So, NN can solve vision - and the aspects of driving NN can't solve, neither can Lidar.

Still, let us assume you are right. Here is the other issue. To solve driving you need lots of data. Billions of miles of driving. If you have an ugly expensive Lidar, you can't use normal cars driven by millions of people to gather data - so expensive Lidar is hindering solving driving.

Purely in terms of robotaxi costs, you are right. With a $50k Tesla robotaxi, I get an NPV of $257k. With a Lidar $100k robotaxi, that would be $215k. Not much of a difference, unless there is a lot of competition which forces you to reduce the price substantially below $1/mile. Let us say, the competion forces 0.30/mile. Tesla robotaxi would have a NPV of $14k (20% IRR) but a lidar taxi goes to -28k NPV. Ouch.

So, if both Tesla and Lidars solve FSD then Tesla can drive down price and send Uber etc into their permanent state of endless losses.

ps : Because of my glasses, my night vision (with all the glare in the perennial rain we have here) is really bad. I hate driving at night when it rains - really looking forward to FSD

pps : This guy, BTW, works for Google.

Tejas Kulkarni on Twitter

depth from unconstrained video (unknown camera parameters). the results look really impressive! (link: https://arxiv.org/pdf/1904.04998.pdf) arxiv.org/pdf/1904.04998…​

 

[mongo]

Elon has backpedaled a bit, and has recently given 30 years as the lifetime of the solar roof as opposed to the life of the house.

Glass lasts near forever (if hail proof), but the solar output is only guaranteed for 30 years. So they need to validate the heck out of the cells and connections.

Tesla Solar Roof

That's why we offer the best warranty in the industry - the lifetime of your house, or infinity, whichever comes first.

Power warranty 30 years
Weatherization warranty 30 years

I'm guess weatherization relates to aesthetics/ color fade.

 

[dc_h]

Steve Jobs had to get fired and spend some time in the so called wilderness to wisen up and become a better leader.

Doesn’t anyone think the bad Uber and Lyft IPO’s aren’t in part to the Tesla loud on the horizon? Maybe he should have waited longer for the reveal, but that could have put an extra 20 billion in Uber’s pocket.

 

[EVNow]

Perhaps $38k is what they think the Model 3 long range model customized as robotaxi will cost at the end of next year.

Yes - if I have to guess, it is an afternoon (or evening) worth of work by one of their car costing guys. They have a list of may be 10 things they want changed for robotaxi and he looked up the price differential. Million mile battery is something they probably have spent more time on and have a decent idea.

 

[EVNow]

Sorry, I don't get it. Missing a sack?

"take it with a sack of salt" instead of "take it with a grain of salt".

 

[Duffer]

Has it though? My company is being acquired in an all stock deal that was announced about the same time. It won't close until late summer.

I could easily be mistaken. However, I thought Tesla was on its 3 extension offer to Maxwell because not enough shares were tendered? I forgot who, but there was a post that listed which banks/hedge funds that controlled Maxwell shares that were not being tendered thereby keeping the deal from going through.

 

[EVNow]

That's only $500/car of savings.

Thanks for this - I was trying to calculate per car depreciation to get a better idea of margins. This helps.

BTW, anyone has any links to EM talking about margins in the cap raise call ? It is obviously crucial to figuring out profitability going forward. 20% makes it very difficult, if it goes up by a few pp, it gets easier.

 

[EVNow]

3. The battery investor event.
I think it is not in Elon's interest to give hard estimates on MXWL tech making it into real cars.
If he says 3 years, I may cancel the Y order and wait 1 more year. Anybody here thinks he should be transparent and make some promises?

Musk should be transparent and not give any estimates.

 

[ruralectric]

"take it with a sack of salt" instead of "take it with a grain of salt".

Got it. Was kicking myself that I didn't instead say, "Missing sack? Nope. Still there. Last time I checked."

Cheers.

 

[jbcarioca]

I believe some did outsource some or all of the engines. You can tell which ones because they're the ones that went bankrupt.

Really it is good to be accurate, whether or not we like the truth.
Every major auto manufacturer has outsourced some engines. Some much more than others, but every one has had at least some badge-engineered models. This would be a very long post we’re i to attempt a comprehensive list. Both engines and transmissions are the subjects of JV or outsourced engines.
Examples include: the Mercedes M274, built by Nissan and also used in an Infiniti Q50. The BMW/PSA joint 4 cylinder engine. In diesels joint ventures and pure outsourcing have been very common.

Manufacturers almost never admit their outsourcing engine/transmission deals. The sometimes even deny the facts. Honestly, engines have become much less distinctive as technology content and strict emission/efficiency demands have risen. Today, at a minimum, all manufacturers depend on outside sources for critical engine components even if they actually build them themselves. More often these days, shared development is the norm and pure outsourcing has become common.

It is precisely that reality that hampers BEV development, because no traditional manufacturer actually has sufficient in-house engineering and production talent. None, repeat, none have enough vertical integration to actually control their new product development. In that they resemble major airframe builders, in that Boeing and Airbus are really more systems integrators than they are integrated manufacturers.

Tesla has been very vertically integrated in large part because major suppliers would not deal with Tesla at all, or on favorable terms. In the end necessity has resulted in competitive advantage.

 

[Carl Raymond]

Doesn’t anyone think the bad Uber and Lyft IPO’s aren’t in part to the Tesla loud on the horizon? Maybe he should have waited longer for the reveal, but that could have put an extra 20 billion in Uber’s pocket.

Yes. You are not alone.

Tesla, TSLA & the Investment World: the 2019 Investors' Roundtable

 

[uselesslogin]

I could easily be mistaken. However, I thought Tesla was on its 3 extension offer to Maxwell because not enough shares were tendered? I forgot who, but there was a post that listed which banks/hedge funds that controlled Maxwell shares that were not being tendered thereby keeping the deal from going through.

Yeah then someone else said that the big holders never tender until the last minute so the extensions were actually for SEC approvals and things like that. I don't know for sure either though.

 

[Electroman]

Would you be comfortable knowing that key parts of the car like screen or battery did not have enough stress testing to determine whether they'd last?

My first reaction was to assume you are a troll to spread the same rumor that the article's author has done.

And that person is known Tesla hater who once started the 'Tesla death watch' count down website, and has consistently been wrong on all his highly negative opinions. If three different members write 4 posts each with the phrases like,

"Tesla failed to do testing"
"Tesla has the habit of using defective parts to save pennies"

then it becomes a fact for the ignorant. Which is the stated aim of these authors. You will soon notice that, this article will be quoted extensively by many in Sleazy Alpha, and then in Business Insider and finally it will show up in WSJ - and by then it becomes an undisputable fact.

What makes you think Tesla did not do adequate testing? Just because a component, which is absolutely new to automotive industry had more than acceptable failures, that doesn't mean Tesla at that time willingly and knowingly used a defective part. Trollish phrases like these would be welcome in Yahoo groups and Seeking Lies.

 

[anthonyj]

Enhanced Summon coming next week!

Elon Musk on Twitter

 

[M3Rider]

My first reaction was to assume you are a troll to spread the same rumor that the article's author has done.

And that person is known Tesla hater who once started the 'Tesla death watch' count down website, and has consistently been wrong on all his highly negative opinions. If three different members write 4 posts each with the phrases like,

"Tesla failed to do testing"
"Tesla has the habit of using defective parts to save pennies"

then it becomes a fact for the ignorant. Which is the stated aim of these authors. You will notice that, this article will be quoted extensively by many in Sleazy Alpha, and then in Business Insider and finally it will show up in WSJ - and by then it becomes an undisputable fact.

What makes you think Tesla did not do adequate testing? Just because a component, which is absolutely new to automotive industry had more than acceptable failures, that doesn't mean Tesla at that time willingly and knowingly used a defective part. Trollish phrases like these would be welcome in Yahoo groups and Seeking Lies.

I haven't been watching TSLA long enough to know all pre-history. Heard about death watch, but never looked up who started it.

So, given these facts I apologize that I linked that article. Nevertheless, I hope that Tesla takes all the time they need to confirm how MXWL tech affects the cells long term and don't rush to roll it out.

 

[MP3Mike]

Enhanced Summon coming next week!

Elon Musk on Twitter

That tweet is more than a month old...