1x Tsla Discussion Threads

[hpcg] 11/5/2024 13:12

cfb2 - why would any analyst have model 2 numbers for 2025? As far as we are aware it has been sidelined, and even without that it would take 1-2 years from the unveiling to production starting. There was maybe some kind of prototype seen at the Berlin works earlier this year, but it looks more like a hacked about 3/Y. The model 2 is inevitable for Europe at some point, but the company also has a refresh program to get going, while the truck will never see tarmac outside of North America.#

These matter because a) free cash flow is now negative, b) stock compensation expense either needs to increase dilution or becomes ineffective as a pay incentive. One outstanding benefit of a public listing is to be able to pay employees partially in stock, and the mega caps have made full use of that in their rise.

[simon gordon] 11/5/2024 12:46

Inside EVs - 9/5/24:

I Went To China And Drove A Dozen Electric Cars. Western Automakers Are Cooked

A trip to the Beijing Auto Show reveals just how advanced China's EVs are. So what are the so-called "foreign" automakers doing about it?

[hosede] 11/5/2024 10:14

I really don't think Tesla is the way forward to save the planet (actually save us)

[blusteradjuster] 11/5/2024 09:59

It’s an unexpected twist that some NIMBY/degrowther greens are indirectly allied with big hydrocarbon.

Technology is and always was the solution.

[hpcg] 10/5/2024 23:37

Protesters trying to storm the Tesla factory in Germany is not a good news story. I know it has been ongoing, but things seem to be heating up.

[cfb2] 10/5/2024 22:48

hpcg: Some analysts are starting to model in the robotaxis but most aren't. If you aren't then you're only modelling in the cars and power storage business and it's difficult to justify the price, especially if you don't believe their model 2 will exist by the start of 2025.

simon: It made me chuckle too but also because the regulators are asleep at the wheel.

I have no idea when AGI will happen.

What I do know is the current AI is good enough to disrupt numerous industries. It'll start at the monotonous and mundane and work it's way up with pseudo intelligence. It doesn't need AGI for it to be disruptive.

[dominiccummings] 10/5/2024 21:55

"DC, the software to land a rocket vertically at a precise point is pretty easy."

Odd that only one company seems to have solved the problem. If easy, you would have thought many would do it?

[simon gordon] 10/5/2024 18:06

H,

Sure the big money is focused on what you say, the smaller money seems transfixed on FSD.

-----

cfb2,

I have no idea if what you say is correct. I don't read the papers like you. Because the AI hype (alarmism) is so immense I try to read the sceptics.

This tweet made me laugh:

I met someone helping the British government with AI regulation. When I asked what they were going to regulate, he said he wasn't sure yet, and this seemed the most intelligent thing I've heard anyone say about AI regulation so far.

-Paul Graham, 1st May 2024.

[hpcg] 10/5/2024 17:58

The conversation here is pretty irrelevant to Tesla. Money mangers are not much focused on AI, how well FSD works (in terms of their opinion), how well or not the cars are built. Theories around false accounting are also irrelevant because those are so few and far between. Their decisions are made on publicly available data around sales volumes, margin, and data analytics of owners' opinions, potential buyers opinions and so forth. Opinion counts on build quality, FSD, value for money, repurchases, the competition and such like. No one throwing hundreds of millions around does so on their own opinion. Some will use charts, others are trackers and closet tracker. Anyone using a chart is selling.

Almost no one can predict the future of autonomous vehicles, neural networks, AI and so forth. So no one is using that kind of speculation to trade the shares.

[simon gordon] 10/5/2024 16:43

Bluster,

Did you watch the Computerphile video? Seemed pretty science-based.

From what I see in the AI scene quite a few thinkers are exasperated by the hype / bull. That poster is one of them.

[blusteradjuster] 10/5/2024 16:34

All technology would be 'magic' to some, depending on when they lived.

That's not to say AGI is coming anytime soon - but that summary is peurile, Simon.

[cfb2] 10/5/2024 14:07

Pierre: I agree with your assessment that automated cars are at least an order of magnitude more complex than landing a spaceship and the problem is much less constrained. What I am tracking is the number of times drivers intervene when the FSD software is running and that is rapidly decreasing from one release of software to the next.

Tesla are currently working towards level 4 automation. Level 5 automation is allowing the car to drive in any road conditions. The FSD software will not engage if it believes the road or weather conditions are too perilous.

You may be interested in the following video comparing Mercedes' approved automated driving over the same stretch of road driven by Tesla FSD. There are 40+ interventions by the Mercedes compared to 0 by Tesla.

[1pencil] 10/5/2024 14:05

Its not according to Scotty though, he is just quoting someone else.

[hosede] 10/5/2024 14:00

cfb
I guess he'll be able to have long chats with The Donald then :-)

[cfb2] 10/5/2024 13:39

I watched just over 4 mins of Scotty Kilmer's video saying Elon is going to jail:


It's quite difficult to watch because the guy clearly rants and rambles a lot and the editing of the video is disjointed because it has been cut every ten seconds to remove it.

According to Scotty, Tesla are making up the figures and their cars are not being made profitably. This would mean that their accounts are fraudulent. Which would also mean the free cash and earning must have come from an outside source to spend on factories, multi-billion data centres and salaries...Russians? Aliens? He doesn't say.

Sandy Munro has torn down various Tesla vehicles and Munro Associates make their money by selling reports to OEMs to show how Tesla makes their cars at that price. Scotty provides no evidence to his accusations.

If you read the comments below the video it seems few of his subscribers agree with it either.

I'm all for listening to others but not when their brains are drooling out of their ears.

[1pencil] 10/5/2024 11:02

Scotty Kilmer on YouTube is worth a listen, followed this guy for years as often tinkering with classic cars - he has an uncanny knack of of calling market trends ahead of time.

Even his subscribers wonder how he gets away with some of the things he says, last video case in point.

[pierre oreilly] 10/5/2024 00:02

DC, the software to land a rocket vertically at a precise point is pretty easy. Everything is known. The only small difficulty would be getting the hardware to react correctly to the software.

The software to drive a car autonomously is so complex (orders of magnitude more than landing a rocket) that imv, it will never in the foreseeable future, be generally implemented without causing accidents. Maybe less than humans, but human accidents are expected, softwear accidents will be corporate negligence.

I know there are trials of them under very specific circumstances (like no rain or other bad weather etc, slow max speeds etc etc) but imv they'll always be at research level and not generally implemented. And musks broken assurances backs that up.

[blusteradjuster] 09/5/2024 12:30

Istr a storm-in-a-teacup (full of water?) about US tourists visiting Europe complaining of the lack of available drinking water such that they were left feeling thirsty.

One comment/cause-hypothesis that stuck with me was the suggestion that all the walking around (as we do, in European cities) was taxing their unfit bodies causing sweat and dehydration..

[hpcg] 09/5/2024 12:06

CATL - now there's a business I'd long like to have been able to invest in.

cfb2 - when I moved back from Houston to London the best part of 20 years ago I never got around to buying a car. It isn't necessary in London, and I know far more people that don't own cars than do. Not owning one has helped me grow my portfolio. I probably will buy one in the future at some point, but I am somewhat fearful that I might walk less as a consequence, which is what happens to a lot of people.

[blusteradjuster] 09/5/2024 09:52

Seba's basic premise is x% improvement (lower cost/greater performance/longer life) per year for n years.

Exponential mathematics.

Different tech, different value of x => similar profile.


There's an uncanny consistency/reliability to it - even though it is the compound effect of various minor or major improvements.


Anyway, let's see how CATL's latest claim bears out.




New commercial vehicles’ battery launched by CATL with Yutong, with “lifespan of up to 15 years and 1.5 million km”

[zho] 09/5/2024 09:35

Tony Seba's talks are v interesting, although he did predict back in 2016 or 2017 that we were just a couple of years away from price parity between EVs and ICEs.

Also, that we should expect EVs to last for 200,000-250,000 miles (due to far fewer moving parts) and the life and costs of batteries appear to make half this a far more plausible limit.