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

[Carl Raymond]

Does hard-to-build mean expensive-to-build?

A little anecdote…

Two years ago I built a house, from a custom designed prefab steel kit. The pieces were built to the mm. It was my first house. Prior projects were all single rooms.

By the latter stages I had learned a valuable lesson. The more time I spent sharpening my pencil, using the laser level, precision marking the floor, using ONE datum, preparing, the less time I spent building. Invest the extra half hour at each stage to save half a day of late stage fudging.

My interpretation of Elon’s email is that Cybertruck *will* be cheap to build in 2024. But it will be a little slower in the initial months ensuring every dimension critical component is within these new tighter specs.

edit: Elon has also recently tweeted that Optimus parts are hard to build. Same same.

 

[Electroman]

We will have to see. The rumor is that the top range will be 350 miles at launch, but rumors are just that. I'm not believing anything until it's confirmed, same with that pricing.

Yup. There will be a ton of disappointment at the range and cost when the first set of variations are unveiled.

 

[ByeByeJohnny]

View attachment 967683

I don't get Elons response here.

Why does he call an opinion on the software 'complaining'? Every Beta program I've been in has begged for opinions and observations. This guy suggests that something should be changed in the approach. That's input, not complaining. It could be more specific but I'm sure he was ready to be specific if given the chance. Or has Tesla said that they don't want input on this specific version?

 

[mongo]

I don't get Elons response here.

Why does he call an opinion on the software 'complaining'? Every Beta program I've been in has begged for opinions and observations. This guy suggests that something should be changed in the approach. That's input, not complaining. It could be more specific but I'm sure he was ready to be specific if given the chance. Or has Tesla said that they don't want input on this specific version?

Read the chain...
James got an early non-wide-public release after requesting it using internal contacts. Then made a post complaining it needed work. Well duh, it's an early release of a Beta software package...
"Unfortunately I have to say"???, no you don't. No you didn't.

@elonmusk No. Let me make something clear: James contacted me directly to be included in *early* beta, which is only ~1000 cars, mostly employees. Early beta explicitly has issues or it would be rolled out widely, so publicly criticizing something he had asked for is wrong.

Pope of Muskanity @RationalEtienne · Aug 24, 2022 Okay, it's not that you minded the criticism of that very early version that James asked for, but the public channel used to send the complaint. For the record, I know that James loves Tesla and that he was, through his criticism, trying to help as best he could. Love y'all.

Elon Musk @elonmusk · Aug 24, 2022 I do not doubt that. Criticism, public or private, is welcome after we go to wide beta. Early beta has many known issues. The reason we release it to a limited number of cars is to discover unknown issues.

https://twitter.com/elonmusk/status/1562507192342761473

 

[Tachyon]

New referral award announced
“ cybertruck delivery event invite”.


Refer friends and family and earn credits that you can redeem for new awards—including one year of Premium Connectivity or an invitation to attend our upcoming Cybertruck Delivery Event in Austin, Texas.

Cybertruck Delivery Event invite redemptions are first come, first served and are valid for you and one friend. Space is limited. If you miss your chance to claim, check back later—additional tickets will be released as the event approaches.”

 

[StarFoxisDown!]

“ cybertruck delivery event invite”.


Refer friends and family and earn credits that you can redeem for new awards—including one year of Premium Connectivity or an invitation to attend our upcoming Cybertruck Delivery Event in Austin, Texas.

Cybertruck Delivery Event invite redemptions are first come, first served and are valid for you and one friend. Space is limited. If you miss your chance to claim, check back later—additional tickets will be released as the event approaches.”

Gotta love how Elon pours cold water on the idea of a Cybertruck delivery event happening soon only for Tesla to make it rather clear that Cybertruck delivery event will in fact happen soon

 

[Thekiwi]

Gotta love how Elon pours cold water on the idea of a Cybertruck delivery event happening soon only for Tesla to make it rather clear that Cybertruck delivery event will in fact happen soon

Vaguely remember a referral program many years ago that involved people getting free roadster 2.0’s - some people still waiting on those?

 

[ItsNotAboutTheMoney]

Gotta love how Elon pours cold water on the idea of a Cybertruck delivery event happening soon only for Tesla to make it rather clear that Cybertruck delivery event will in fact happen soon

Not soon. Upcoming is not soon.

 

[insaneoctane]

Does hard-to-build mean expensive-to-build?

A little anecdote…

Two years ago I built a house, from a custom designed prefab steel kit. The pieces were built to the mm. It was my first house. Prior projects were all single rooms.

By the latter stages I had learned a valuable lesson. The more time I spent sharpening my pencil, using the laser level, precision marking the floor, using ONE datum, preparing, the less time I spent building. Invest the extra half hour at each stage to save half a day of late stage fudging.

My interpretation of Elon’s email is that Cybertruck *will* be cheap to build in 2024. But it will be a little slower in the initial months ensuring every dimension critical component is within these new tighter specs.

View attachment 967708

edit: Elon has also recently tweeted that Optimus parts are hard to build. Same same.

Yeah, I read it differently. Unlike the CT, Legos & soda cans very low cost.

 

[dhanson865]

First case I've ever seen of a full on software based unintended acceleration scandal. Surprising how many steps it takes to trigger and I'm sure someone somewhere will try and say Tesla has the same issue.

Nissan recalls nearly 80,000 Leaf EVs in Canada and US for unintended acceleration

Nissan is recalling a large number of their Leaf electric vehicles (EVs) over concerns of unintended acceleration. The recall covers nearly 80,000 units in both Canada and the US. According to the recall notices posted […]

driveteslacanada.ca

On 2018-2023 Nissan Leaf EVs with affected VCM software, there is a risk of sustained engine torque if a driver takes both of the following actions within 8 seconds after deactivating cruise control, Intelligent Cruise Control or ProPILOT Assist functions: (a) switches driving mode (i.e. from ‘D’ to ‘B’ position, or ‘ECO’ mode, or e-Pedal ‘ON’); and (b) then applies and releases the accelerator pedal.

 

[Chunky Jr.]

Argh. This doesn't sound like it will be anytime soon. If they are still polishing it then ok. Sure would be nice to get some sort of info though.

https://twitter.com/x/status/1694460811291578722

This is very puzzling to me because I thought Elon and Franz have said that the Cybertruck was designed from the ground up with manufacturing in mind.

 

[Singuy]

This is very puzzling to me because I thought Elon and Franz have said that the Cybertruck was designed from the ground up with manufacturing in mind.

With manufacturing in mind but not with OEM suppliers in mind. Based on what Elon have been saying how there are "many new technologies in this truck" and they even went and made their own inhouse windshield wiper tells me there are more inhouse components than usual.

 

[Pezpunk]

This is very puzzling to me because I thought Elon and Franz have said that the Cybertruck was designed from the ground up with manufacturing in mind.

I think it’s absolutely ludicrous that people bought the obvious lie that cybertruck was built with anything other than its (ugly) looks as the top priority. It’s going to be difficult and problematic to build. I said that at its announcement, and the next three years will prove me even more right than the last three years already have.

 

[MrMoo]

I think it’s absolutely ludicrous that people bought the obvious lie that cybertruck was built with anything other than its (ugly) looks as the top priority. It’s going to be difficult and problematic to build. I said that at its announcement, and the next three years will prove me even more right than the last three years already have.

Those are the "many new technologies" he was talking about. They had to jump through a lot of hoops to get that look.

There is a reason people don't make cars like this. I hope what he plans to do pays off because they've probably spent so much RND money and time on this thing that if the money and effort went elsewhere we would probably have roadster + Model 2.

I also have a feeling that the build quality and after market problems are going to be a massive headache.

 

[betstarship]

I think it’s absolutely ludicrous that people bought the obvious lie that cybertruck was built with anything other than its (ugly) looks as the top priority. It’s going to be difficult and problematic to build. I said that at its announcement, and the next three years will prove me even more right than the last three years already have.

You hate it so much that you love it. We all do. Let it spank you hard with its hard angles and ugly looks to an order and delivery, baby.

Do it.

 

[MC3OZ]

CT is new and innovative that is why it is initially difficult to build.
The reason other carmakers don't build cars like this is, innovation is largely foreign to them, especially risker major innovation.
Elon hasn't said it will be difficult to build forever,

It is initially harder to build because some parts and processes are hard and require tight tolerances. They don't mention the parts of the build that are easier, because they only get the wins on the easy parts, once they can do the hard parts at the required rate.

I agree CT is polarising, some seem to what to make every Tesla Cyber vehicle, others think one Cyber is one too many.
No one is forcing you to buy it.

I think it will average out as similar to the Model 3 ramp perhaps slightly better, there is a slight chance that it might be smooth sailing from day 1, because Tesla already has the ducks lined up. But for any innovative new product, problems are not unexpected.

It isn't a "bet the company proposition". Once Tesla is able to successfully build one Cyber vehicle ion high volumes, the next one would be a lot easier.

Those that went through "Production Hell" haven't signed up for a second dose of hell on a whim. If they are going through hell again, it is because it is worth it.

 

[Singuy]

So about that mindblowing post *rumor*

1. Making about 10 trucks a day
2. Deliveries in 3-4 weeks
3. During or right after the delivery event, SCs around the US will deliver 5-6 trucks each
4. First build are high end model(tho no details on range).
5. Cost in the mid 70s but final price not given
6. The 5 customers per SC are chosen already not based on reservation numbers but based on "rank" (I am assuming social influence?).

https://twitter.com/x/status/1694508411151667551

 

[heltok]

I think lots of people fail to understand the nVidia beat. They think "wow high profits" but fail to understand that nVidia is supplying a tool that companies buy because it makes them more money than it costs them, not a fun consumer product. Same with Dojo, some analysts like Gary Black think that Dojo will only be valuable if it becomes AWS which fail to understand the product. No, Tesla will not sell compute(at least for a long while) because they can use it better themselves!

https://twitter.com/x/status/1671647116538544128

https://twitter.com/x/status/1686513363495346178

When nVidia beats it's great for them. But it's also an indication that they are producing more H100s that companies like Tesla need. Which will lead to better products/development for all the companies that are H100 starved like Microsoft, Tesla, X.ai etc. That's where the big value of the product will be made.

 

[Singuy]

I think lots of people fail to understand the nVidia beat. They think "wow high profits" but fail to understand that nVidia is supplying a tool that companies buy because it makes them more money than it costs them, not a fun consumer product. Same with Dojo, some analysts like Gary Black think that Dojo will only be valuable if it becomes AWS which fail to understand the product. No, Tesla will not sell compute because they can use it better themselves! Like Elon said:

https://twitter.com/x/status/1686513363495346178

When nVidia beats it's great for them. But it's also an indication that they are producing more H100s that companies like Tesla needs. Which will lead to better products/development for all the companies that are H100 starved like Microsoft, Tesla, X.ai etc. That's where the big value of the product will be made.

They are creating more H100s in accordance to their wafer agreements with TSMC. However most of their beat comes from just price gouging when there are poor alternative as their margins jumped 20% QOQ.

 

[MrMoo]

They are creating more H100s in accordance to their wafer agreements with TSMC. However most of their beat comes from just price gouging when there are poor alternative as their margins jumped 20% QOQ.

I don't see why they would do that. They must have a demand problem. /s