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

[wipster]

382 in Germany (really!). See you in the morning!

 

[mrdoubleb]

Looks like GF1 is supplying Austin for the Semi, at least initially...

Austin will definitely make cells for Cybertruck, probably Model 3+Y, and eventually the Semi..

IMO this is simply getting Semi production rolling as quickly as possible.... strong hunch this will be the first product produced by Austi, al signs seem to indicate that.

What is interesting is that we were hearing about a Semi pilot line being built just few months ago and I still think the target is to start production this year - no way Austin will be ready by then. So if those rumors were true, the new GF1 line could be producing cells for the Semi pilot production as well. Knowing Tesla's history of continuous innovation and development, it fits that they would want to create a V 0.5 production line first, produce a few dozen (or a hundred?) Semis to take care of their own logistics needs and build the Austin lines with that experience. Plus it would be a heck of a marketing move to start switching Tesla logistics to their own trucks.

 

[insaneoctane]

I'm surprised by how many folks don't seem to get the significance of the 1M battery. Several previous opinions offered with good points. I'll add a couple more...

• Disruption. Tesla keeps changing the game. They are playing 4D chess while the other mfgs play checkers. Tesla is always improving technology. 300K miles was good enough last decade (even if Tesla was the only company making batteries last even that long). This allows V2G options, lower warranty cost, premium brand premiums, enhanced resale, longer vehicle in service etc. Mentioned up thread, people (even if subsequent owners) don't keep cars past 10 to 15 years because they become unreliable and expensive... But what if they didn't?
• Helps convert MORE former ICE buyers. 1M mile battery and motor ought to knock down any longevity concerns. Elon's loves knocking down excuses.
• Even a battery lottery loser wins. Battery degradation and longevity follows a statistical distribution. Tesla warranty is designed to cover only fairly extreme outliers. Someone who pack goes from 325 miles to 210 miles in 3 years isn't currently covered by warranty. That is losing the battery lottery. A 1M mile battery should offer far better statistical result.

 

[saniflash]

Can someone explain why this works? Or point to some links. I don’t get why this reduces the margin.

simplified: if you're holding shares (or wrote puts) on margin, the worst that can happen is that the stock goes to 0; if you buy puts, you buy the right to sell the shares at a price >0; thus, your exposure (and the broker's risk) is reduced.

 

[MC3OZ]

What is interesting is that we were hearing about a Semi pilot line being built just few months ago and I still think the target is to start production this year - no way Austin will be ready by then. So if those rumors were true, the new GF1 line could be producing cells for the Semi pilot production as well. Knowing Tesla's history of continuous innovation and development, it fits that they would want to create a V 0.5 production line first, produce a few dozen (or a hundred?) Semis to take care of their own logistics needs and build the Austin lines with that experience. Plus it would be a heck of a marketing move to start switching Tesla logistics to their own trucks.

I'm wondering what the actual process for constructing the Semi will be...

In particular what the cab will be made from, and how it will be made..

I think they need to do GA at Austin for any serious volume, they will not have the workforce at GF1.

I'm half expecting GF1 to ship a frame with motors wheels and a battery back, Austin to get the cab from somewhere and put it all together.
it is likely the cab doesn't need to provide much structural strength and can be light weight..

The start of Semi production is one of the more interesting mysteries IMO...

I will not be surprised, to see Semi GA in a tent...how cold does it get in Austin?

The answer is winter in Austin is slightly warmer than San Francisco, but much the same.

 

[Buckminster]

Will the cold hard reality of the wildfires & smoke hit home for people, & make them think more positively about Investing in green companies, this week - in particular TSLA?

Not referring to the folks that are in the midst of fleeing from this terrible calamity - as they have bigger things to worry about - but the folks that are impacted peripherally by the smoke, and will be making investment decisions this week.

It occurred to me that it may now be easier to ask Engineers to move to Austin post the fires.

 

[Buckminster]

Premarket opened at 384.

 

[UncaNed]

"Wood and ARK are nice but they don’t include Tesla Energy."

Yes they do if they include TSLA, no??

 

[Rexjamo]

• Even a battery lottery loser wins. Battery degradation and longevity follows a statistical distribution. Tesla warranty is designed to cover only fairly extreme outliers. Someone who pack goes from 325 miles to 210 miles in 3 years isn't currently covered by warranty. That is losing the battery lottery. A 1M mile battery should offer far better statistical result.

That’s more than a 30% loss. I thought more than a 30% loss and you’d get a replacement battery?

 

[hobbes]

Interesting information: https://twitter.com/carsonight/status/1305332657622978560?s=20

Indeed very interesting! Read the whole twitter thread, many more insightful discussion in there.

 

[hobbes]

My wife won't let me sell our 2006 Lexus R400h, even though she has an X and I drive an S, because the Lexus is still working fine and she has this imaginary situation where we find ourselves suddenly without warning needing to take a long road trip where we can't afford the time to stop and charge.

I guess rental cars are not the thing in the US ?!

 

[hobbes]

doubtful on GF1 supplying Austin factory without a vast expansion in capacity which doesn't seem to be on the cards given commentary by Panasonic. Going forwards it seems more likely all new Tesla factories will have their own Tesla in house battery cell production on site.

Read the whole twitter thread - carsonight expects GF1 seeding GF Austin before cells are produced locally, just like it did for Shanghai.

 

[emerson06]

Increased EV incentives in Sweden: 70k SEK ($8000) if you buy a new EV. Up from $6800.

 

[Artful Dodger]

Considering that Tesla recently:

• licenced to sell electricty in western Europe (EPEX spot power exchange)
• sent out a survey in Germany about switching Energy Suppliers:
  • supply contract linked to owning of a rooftop solar system
  • Tesla Powerwall home storage device with Tesla software (Autobidder?)
    
  • package to include:
    • home energy storage
    • solar panels
    • Tesla Wall Connector EV charger
    • access to a public EV charging network
    • electricity derived from renewable energy sources

This weekend we learned that certain v3 SuperChargers in Germany are allowing 3rd Party EVs (those with CCS2 connectors) to charge at those Tesla stations.

I think Tesla is about to offer an new energy bundle to customers in Germany.

This is HUGE. The TAM is just Western Europe!

Cheers!

 

[emerson06]

Increased EV incentives in Sweden: 70k SEK ($8000) if you buy a new EV. Up from $6800.

UPDATE: Also an increase in the tax for high-emission vehicles.

 

[Paul_SF]

https://twitter.com/elonmusk/status/1305435240576425984?s=21

 

[Prunesquallor]

....
I will not be surprised, to see Semi GA in a tent...how cold does it get in Austin?

The answer is winter in Austin is slightly warmer than San Francisco, but much the same.

It's SUMMER that would be the problem. No tent.

 

[CorneliusXX]

It seems self-evident that battery life matters, especially for resale buyers. One of the major concerns about EVs is range loss. Wouldn't a million mile battery help with this tremendously? It certainly matters for any kind of automated driving service, ride-sharing drivers, etc. For a car that is still a bit expensive, a battery with such tremendous life helps justify costs and further sets Tesla apart from the competition (as if there is any real competition, but humor me).

A million-mile battery would be a nice feather in Tesla's cap among the rest, and it's among a constellation of metrics that any buyer might consider, but especially so in the resale market.

The question is "how much does it matter?". Certainly a longer life is better than a shorter one, and for fleet buyers that are less emotional about vehicle purchases then it's likely to matter more.

But I'd assume there's an upper bound to how much the extra battery longevity can increase the second hand value of a vehicle as they're still anchored to new vehicle prices. If you had the choice of buying a 5yo vehicle with 200k miles on the clock (800k miles of range left), vs a new vehicle that has upgraded electronics, new interior, probably better handling and efficiency given Tesla's continuous innovation, is lighter or has more range due to battery improvements - how close could you get to new vehicle prices before your decision jumped from buying used vs new? I doubt you'd pay 80% of the new vehicle price for the 5yo vehicle even if it has 80% of its life left - but everyone is different.

The real value increase for the increased battery longevity is for the robotaxi fleet where as long as the vehicle is clean and works it can be sent out on the road to make money.

 

[CorneliusXX]

Sandy Munro Compares Tesla's Motors & Invertors With The Competition

I hope Sandy can get his hands on a Lucid - It would be very interesting to see a drivetrain comparison between the two.

 

[CorneliusXX]

I called it =)

https://twitter.com/karpathy/status/1305302243449516032
https://twitter.com/karpathy/status/1305306119372300290

Andrej Karpathy
@karpathy
Transformers
. Specifically, organizing information processing into multiplicative message passing in graphs; generalizing, simplifying, unifying, improving neural nets across domains. For a while there I was growing bit jaded with slowing progress on neural net architectures
2:28 AM · Sep 14, 2020·Twitter Web App

Replying to
@karpathy
feels like a lot is kicked up in dust, and the closest we've come to a full refactor of your typical neural net. stop me if I'm being overly dramatic

Any chance you could give a dumbed down summary of the changes?