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

[Rarity]

The presumption that compute resources are scarce is in no way supported. Abundant resources are just as likely.

Particularly in light of Tesla's statements such as those regarding the potential to sell time on Dojo at some point.

Optimus and FSD most significant constraints are more likely to be found in the wetware trying to determine how to accomplish the task most efficiently.

The future is unknown, but I think it's clear that that there is a huge imbalance right now between supply and demand on compute. It has been this way since the introduction of Chat GPT-3.5 earlier this year. The scrum has been intense. Tesla's CEO, its prior CFO, NVIDIA's financials, and Oracle's CEO are all telling us this plain as day.

You can be sure that Tesla paid a king's ransom for its 10,000 GPU H100 cluster that it just brought on line. Tesla readily paid the price, publicly thanked NVIDIA for their kindness in selling it to them, and undoubtedly wished for more. This despite the fact that Tesla's a penny-pinching miser relatively speaking.

It is said that Microsoft is investing $50 billion per annum on compute infrastructure. I wouldn't doubt it, although I have not confirmed that number. We soon may need to be thinking in those terms.

 

[lascavarian]

This could be interesting - Sandy Munro is going to be on AutoLine live tomorrow.

There will be a couple other guests. One from Reuters.

 

[Carl Raymond]

Better than expected jobs data made investors nervous about interest rates, apparently. Nervous Nellies are to blame.

So when a legacy auto company bankrupts releasing its labour pool, the market surges, right?

 

[GhostSkater]

Some field report on the EV state here in Brazil

Was in São Paulo for two days, the amount of EVs I saw dwarfed my previous visit, lots of BYD “taxi”, some other electric delivery contraptions and a reasonable number of private owners vehicle

But the BYD taxis was a lot, to the point that not even looking for them I saw everywhere

Obviously I took the opportunity of doing a test drive in both the BYD Dolphin and the GWM Ora, both for the equivalent of $30k, pretty nice, but expensive for what it is, but the big fail and what tells me Tesla has nothing to fear, the software in both was absolutely garbage, confusing mess of menus, bad screen and the biggest turn of, there is no built in route planing with charging, none, zip, nada. You are on your own using third party apps to figure it out, no estimate SoC on arrival also even without considering chargers

Both would be ok for urban and short highway trip, but that combined with a low range, doesn’t make it a compelling purchase, which is a shame, because I absolutely loved the Ora and might get one in the future if Tesla takes too long to offer something here

Both can’t do a 250 km trip without charging, the GT version of the Ora can but it’s considerably more expensive, and that only in optimal conditions, bring a little cold, wind and rain and range drops a lot

Sadly I couldn’t find any Teslas, I knew there were some used ones for sale nearby, but they were in the same stores that sell Ferraris, Lamborghinis, Rolls Roxie and similar, and my lack of shame only goes so far

Adding to all that, BYD store was packed, GWM only me there

 

[Thekiwi]

I'm seeing this FUD article all over the internet today, and all the Tesla haters are piling on about how this proves Tesla cars are piles of junk.

Think I'll go for a long drive in my awesome Model Y to clear my head about it!

Just read the article. Unfortunately it seems to lean more “true” than “FUD”.

Clear documented evidence that tesla indeed knew they were shipping cars with sub-optimal suspension components with a high failure rate leading to accidents, and internal communications confirming that they had finally fixed the issue with a redesigned part in 2022. They were fixing the part for free in China, but in the US service staff told to blame the drivers behavior for the part failing. NHTSA also confirms they are investigating similar issue with power steering.

Wouldn't surprise me in the least if this leads to an actual physical recall of all pre-2022 cars. (One wonders how service centers will be able to deal with that.)

Also large class action lawsuit seems inevitable.

Would be happy to hear from others as to why this wont happen (but please no “because Reuters is dumb” responses please.)

 

[Usain]

All Joe said is that he saw 40 cyber trucks during his flight.

We don’t know how many of those were made before today, and we also don't know how many have been loaded onto carriers and taken away today.

Todays production could be 0. Or it could be 100. There is no way to know without continued coverage or at least being able to identify individual units to see if there are units sitting there for longer than a day or not.

If I was a well resourced hedge fund I would have interns sitting on the side of the road 24/7 monitoring transporters leaving Tesla factories in Austin, Fremont, Berlin (and lathrop for good measure). Or at least invest in some solar powered webcams.

That's true. We don't know if that means 40 per day. But it's the best data point we've got and it's definitely the most we've seen so far.

My point was that even at the low number of 40 per day, it's a higher rate than Lucid and many other EV makers. The ramp is only officially 3 weeks old.

 

[Usain]

Just read the article. Unfortunately it seems to lean more “true” than “FUD”.

Clear documented evidence that tesla indeed knew they were shipping cars with sub-optimal suspension components with a high failure rate leading to accidents, and internal communications confirming that they had finally fixed the issue with a redesigned part in 2022. They were fixing the part for free in China, but in the US service staff told to blame the drivers behavior for the part failing. NHTSA also confirms they are investigating simialr issue with power steering.

Wouldn't surprise me in the least if this leads to an actual physical recall of all pre-2022 cars. (One wonders how service centers will be able to deal with that.)

Also large class action lawsuit seems inevitable.

Would be happy to hear from others as to why this wont happen (but please no “because Reuters is dumb” resposnes please.)

I read it as well and I agree with you. But I do wonder why NHTSA hasn't taken any action on the suspension issue. It makes me think we don't have the whole story.

 

[heltok]

I saw an interesting point in the Optimus discussions related to this. Although Bot training seems easier, there are way more degrees of freedom to control as outputs. Think of it this way: Optimus deciding where to move is roughly same complexity as FSD. But once it arrives and starts a task, the Optimus has to decide how to move the left arm shoulder joint for the task in front of it, then the left elbow, then ... how far to rotate the left wrist? When does the index finger get involved? How much pressure until it stops squeezing? Then answer all those questions for the right arm... then the neck twist... you get the idea.
In short, isn't the phase space of Optimus' existence an order or two of magnitude larger than that of a car running FSD? Does or doesn't that mean Bot training will suck a LOT more compute cycles?
Not an AI expert, but asking for one now!!!

Controlling a lot of outputs is not super hard. We have had many control algorithms for this. Reinforcement learning has no issue with this. Some cute examples:

The main difficulty is performing task that require longer term planning, have unclear reward functions, require fine manipulation and working in dynamic environments. Such as folding laundry or walking a dog. But even these we will soon solve with massive amount of data and iteratively improving the system.

 

[Artful Dodger]

Sandy Munro, a respected industry stalwart, was agap at what he saw (and published his 2nd half-hour video highlighting what he heard and saw on Nov 30)

Lol! dat face!

Sandy Munro STUNNED: How Cybertruck Is Made (INSANE) | SOL w. SMR (Dec 19, 2023)

Cheers to the Longs!

 

[MP3Mike]

Clear documented evidence that tesla indeed knew they were shipping cars with sub-optimal suspension components with a high failure rate leading to accidents, and internal communications confirming that they had finally fixed the issue with a redesigned part in 2022. They were fixing the part for free in China, but in the US service staff told to blame the drivers behavior for the part failing. NHTSA also confirms they are investigating similar issue with power steering.

Wouldn't surprise me in the least if this leads to an actual physical recall of all pre-2022 cars. (One wonders how service centers will be able to deal with that.)

I thought that Transport Canada, the Canadian equivalent of NHTSA, had already investigated this and determine that there was no recall necessary, that even when the part broke that the vehicle was still controllable. (I was pretty sure NHTSA was involved in that but I'm not sure.)

Even in China Tesla blamed the driver's hitting things like potholes causing the eventual failure of the parts, but they weren't going to fight China's equivalent of NHTSA, as they would likely end up forcing Tesla to do it anyhow. (Just like Tesla disagrees with NHTSA on the AP recall they just did, but they are doing it anyhow because NHTSA would likely force it otherwise.)

 

[Thekiwi]

I thought that Transport Canada, the Canadian equivalent of NHTSA, had already investigated this and determine that there was no recall necessary, that even when the part broke that the vehicle was still controllable. (I was pretty sure NHTSA was involved in that but I'm not sure.)

Even in China Tesla blamed the driver's hitting things like potholes causing the eventual failure of the parts, but they weren't going to fight China's equivalent of NHTSA, as they would likely end up forcing Tesla to do it anyhow. (Just like Tesla disagrees with NHTSA on the AP recall they just did, but they are doing it anyhow because NHTSA would likely force it otherwise.)

The Canadian outcome sounds promising as a good retort ot the Reuters article, but do you have a link with any information, because when I google it I only get notice of a 2023 recall in Canada for a faulty suspension part on Model 3 2018/2019 years, which is perhaps more evidence to the contrary.

 

[MP3Mike]

The Canadian outcome sounds promising as a good retort ot the Reuters article, but do you have a link with any information, because when I google it I only get notice of a 2023 recall in Canada for a faulty suspension part on Model 3 2018/2019 years, which is perhaps more evidence to the contrary.

The only thing I have handy is someone that complained to Transport Canada and relayed what they were told.

As you may know from my earlier posts on this thread, I did file a formal complaint regarding this same event, which you have described as a "clear manufacturing defect", with Transport Canada. TC, I understand, is similar to NHTSA in regard to automobile safety. Transport Canada (2020 - 0975) recently notified me that they concluded their investigation as follows:

"We have determined that a failure of the rear lower control arm on the Model S is not detrimental to the safe operation of the vehicle and will be deactivating our file" and added "No reports of loss of control. And testing verified this."

In the end it is a balance, the part is designed to break in a collision to help absorb, and redirect, crash forces. Maybe originally it was designed to break a little too early, but there is a trade-off between strength and safety.

 

[M|S|M MYP]

This can't be a coincidence, can it?

 

[capster]

Well, I’m getting mine clear coated and then wrapped. These things will rust otherwise. They’ll also get a patina that some people are going to complain about. And yeah, finger prints etc… will bother another set of people.

Hmm, I’ve been debating the Tesla urethane wrap. It’d be nice to see the CT in person in white and in the clear.

What clear coat do you recommend?

 

[mongo]

Also large class action lawsuit seems inevitable.

Would be happy to hear from others as to why this wont happen

What would you say to the arbitration clause in the vehicle order paperwork that requires writing Tesla to opt-out within 30 days?

 

[2daMoon]

Hmm, I’ve been debating the Tesla urethane wrap. It’d be nice to see the CT in person in white and in the clear.

What clear coat do you recommend?

Just spray the whole thing with bedliner.

Or maybe cover it with closed cell foam and have a Nerf Cybertruck.

 

[capster]

Ramp now is 40 CTs per day (probably)

https://twitter.com/x/status/1737506727854055588

Q1 is going to be pretty cool .

Hmm, so say 40 per day at 5 days a week is 200 week. Assuming that my Foundation Series is, on average, halfway through the 1000, my CT should be ready in, uh, about 2 weeks.

 

[Krugerrand]

Hmm, I’ve been debating the Tesla urethane wrap. It’d be nice to see the CT in person in white and in the clear.

What clear coat do you recommend?

Nyalic is the choice.

 

[wipster]

So did NVDA.

Yeah, that's weird... approximately the same percentage too.

 

[unk45]

Thanks for the analysis @unk45.

Regardless of how original the idea is, it will likely result in lower interest rates for Tesla customers, correct?

Not necessarily. Most credit union rates offered directly are lower than commercial banks offer.
CUSO originated credit union loans are typically at higher rates, although sometimes still below typical bank offering. Because Tesla has historically not subvened interest rates pricing advantage does need lower credit union pricing.

Because we do not know the terms between Tesla and the CUSO there are other ways to optimize consumer offers. One of those is a ’collateral substitution agreement’ similar to that common for securitization, which essential assure high credit quality. If that practice, or even a tenor, loan to value ratio limit or FICO score minimum might apply. Tesla paper has been extraordinarily high quality so explicit terms might be attractive anyway.

All this become quite technical. Despite that I think that people with excellent credit who are placed with a credit union probably will have excellent pricing. Those with blemished credit might nit be placable in this way, or have limitation in deal structure or higher prices,

Sorry for the ambiguity. Without knowing the contractual terms the major thing we know is that Tesla gains from now funding sources and Tesla buyers also have another option. That is a good thing, with better pricing likely for some buyers.