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

[Musskiah]

Today V12.3 shifted far right onto the white line to avoid a dead raccoon in the road and immediately got back to the center of the lane. Roadkill detection detected!

In other news, I sure am seeing a ton of commentary on Reddit and YouTube comments and X about people buying FSD subscriptions today. I have a feeling this is going to be bigger than we all anticipate. Again the average income of a US Tesla owner is $150k, so for 10-20% of them to buy a subscription in the next quarter or two is quite possible. That will add a few $100M to the bottom line every quarter
Hmmmm....

How many Tesla owners are in the US right now? Like 1.5M was it?

 

[NicoV]

Cybervan is my dream vehicle. FSD, AWD, off-road capable, 300+ miles of range, room to live and sleep inside. For me that's been the dream for years. Model Y has been great, but I'm ready for the Cybervan!

I wanted a Cybertruck as a camper, but it didn’t turn out to be suitable for that purpose. I’ll take a Cybervan camper!

 

[MC3OZ]

The pessimistic version doesn't even make sense. The only way Tesla stops growing is by running out of cash or they abandon all projects. They are know for never abandoning large growth vector projects even when they are running out of cash so I really have no idea under what circumstances growth stops forever.

Tesla has a substantial cash buffer, we will see if the end of quarter results add to that pile, or reduce it a bit.

I am still surprised that Q1 2024 production numbers are as low as they seem to be, so we need to find out why that has happened, and if any draw down of existing inventory has happened, "Why" is always more interesting that "what".

We know Megapack, Semi and the Gen3 car are growth vectors that seem certain to happen and in the case of Semi and Megapack the expansion can be low capex. For the initial Gen3 production, we know it is being squeezed into a slightly expanded Austin footprint, and some of the capex has already been spent.

Then we have FSD and Optimus. A lot of the FSD capex has already been spent, but there may be some payments still to be made., For Optimus capex will partially be determined by scale, but smaller scale lower capex seems possible,

In relation EV inventory piling up at dealerships, there also seems to be plenty of ICE inventory piling up at least for some brands. The car industry can be cyclical, and no one has claimed the auto portion of the business is immune to car industry cycles. Megapack, Semi, FSD and Optimus are fairly immune to car industry cycles, and economic cycles in general.

This will be an interesting earnings call, we are about to find out how "why".

 

[Krugerrand]

Tesla has a substantial cash buffer, we will see if the end of quarter results add to that pile, or reduce it a bit.

I am still surprised that Q1 2024 production numbers are as low as they seem to be, so we need to find out why that has happened, and if any draw down of existing inventory has happened, "Why" is always more interesting that "what".

We know Megapack, Semi and the Gen3 car are growth vectors that seem certain to happen and in the case of Semi and Megapack the expansion can be low capex. For the initial Gen3 production, we know it is being squeezed into a slightly expanded Austin footprint, and some of the capex has already been spent.

Then we have FSD and Optimus. A lot of the FSD capex has already been spent, but there may be some payments still to be made., For Optimus capex will partially be determined by scale, but smaller scale lower capex seems possible,

In relation EV inventory piling up at dealerships, there also seems to be plenty of ICE inventory piling up at least for some brands. The car industry can be cyclical, and no one has claimed the auto portion of the business is immune to car industry cycles. Megapack, Semi, FSD and Optimus are fairly immune to car industry cycles, and economic cycles in general.

This will be an interesting earnings call, we are about to find out how "why".

Why production is lower? You can start with domestic terrorism. Then you can add in the Highland ramp at Fremont that’s been slow.

 

[MC3OZ]

Why production is lower? You can start with domestic terrorism. Then you can add in the Highland ramp at Fremont that’s been slow.

That is what I am expecting, most of "why" will be one off factors, or issues like batteries, parts and logistics, that can be improved,

Hopefully the situation in the Red Sea improves, and we don't get a repeat of domestic terrorism...

Factory ramps progress, and are usually done once.

 

[Musskiah]

Why production is lower? You can start with domestic terrorism. Then you can add in the Highland ramp at Fremont that’s been slow.

Don't forget both Lunar New Year in China in Feb and the drawdown of the number of workdays - both effecting GigaShanghai. I think this last piece is the only non-one-off that requires critiquing. Why decrease the number of workdays? Is it truly demand driven? Everything else is unconcerning.

 

[Stretch2727]

I am still surprised that Q1 2024 production numbers are as low as they seem to be, so we need to find out why that has happened, and if any draw down of existing inventory has happened, "Why" is always more interesting that "what".

Giga Berlin lost 2 weeks due to a part shortage then we had the electrical fire there which shutdown the plant for a week+.

With the Model 3 Highland launch in Fremont looks like they lost about 40K of production.

This is probably 60K units lost.

 

[MC3OZ]

Why decrease the number of workdays? Is it truly demand driven?

Lower Chinese demand triggers more need to ship cars out of China to other markets.

That requires markets to ship to, and shipping capacity.

If Shanghai was previously shipping to Canada and is no longer shipping to Canada, that is a fairly major change.

it takes time to open new markets...

BYD is building some of its own RORO ships, companies like Stellantis may be trying to increase exports out of China.

If factories in the EU and US close, and more cars are being shipped from China to other markets, the shipping logistics needs to keep up with the rate of change.

 

[insaneoctane]

So, if we assume Tesla is now past local maxima solution, and they are now marching towards the actual FSD solution (while arguable, evidence of late implies this certainly could be true)....any competition paying attention now knows exactly what to do (and they can skip the last 6 years of effort ($$$) that Tesla has performed). What's the cheapest a competitor could copy Tesla's approach? Importantly, what is the cost today? What is the estimated cost to do it in 1 year? 3 years? What's the monthly fee necessary for an acceptable ROI on that investment? I think that the copy cats will come and they will come soon. Likely in China first. They will Influence the allowable fee to charge (currently approximately $200/mo). Tesla is likely going to maintain a superior product with better safety statistics, but I'm curious how long folks think Tesla can charge $200+/mo? I suspect there is a range of opinions here and why I'm curious of the various thoughts.

 

[MC3OZ]

What's the cheapest a competitor could copy Tesla's approach?

I assume that they need to capture the data, some of that can be simulations, but we don't know how much real world data from actual cars driving around is required.

V11 also serves as a tool to benchmark and improve V12 against, any competitor would be doing that native, we don't know low long V12 was worked on in the background.

And it also seems to require a lot of compute hardware just to complete the training, assuming they have all of the data.

So it probably requires deep pockets and a well resourced project, probably a group of Chinese manufacturers collaborating with government support. I am not sure how willingly the US and EU will accept Chinese FSD operating on their roads, or data from those driving environments being processed in China.

I don't see an alliance of EU carmakers getting results very fast, especially since they haven't really started.

So competition will probably happen eventually, but it will not be easy to achieve,

 

[jschwefel]

As good as FSD 12.3 appears to be, how long will be before we see LSD, or Limited Self Driving. Tesla should be able to quickly and easily train a neural network to test drive the cars, self park, load and unload on car carriers etc.

 

[Ipcress]

Is anyone here aware of Xiaomi SU7? It got 50,000 orders in one day. The spec is comparable to Model 3 and the price is 10% lower. It might already impact M3 sales in China for the coming months.
Last time someone said don't worry about Xiaomi, which is just a phone manufacturer, and how could it make a good car?
I somehow agreed with that, but now I think that is a really careless underestimate.
Although Xiaomi is losing money on each car they sell, but their phone-car ecosystem is pretty good. There are millions of Xiaomi phone users in China, and they can all integrate their information and Apps into Xiaomi SU7 easily, and that is quite an attraction for them to consider SU7.

Many said Friday TSLA dropped because of SU7. I hope the impact will not extend.
Currently the advantage of TESLA M3 is the supercharge network only. Not sure about FSD in China.

I think the lack of quality iPhone and Android integration is without doubt keeping some people from buying a Tesla. No actual evidence, but it sure irritates the heck out of me!

 

[phoggy123]

I have been overall bearish on TSLA over the last 6-9 months. As far as the stock price, I expect it to bottom out over the next 3 months (unless the economy/market takes a real dive). My hope and belief is that we have seen margins bottom out this quarter. I think volume will be limited on the upside and we should hope that overall deliveries match 2023. But as I have said in prior posts, we are in the lull on moving metal but I think profit is going to surprise to the upside. Tesla not opening new factories will be a headwind in production but a tailwind in profit. I think Megabuck is going to surprise to the upside and I think we are very very close to FSD making a real impact in earnings.

I purchased FSD about 1.5 years ago on a 2022 Model Y. FSD Beta 11 was not serious, FSD 12 is very serious and magical at times. I was lamenting my full purchase of the software as I was contemplating trading the Y on an R1T...but not anymore. If the rate of improvement keeps increasing in 2024 this is going to be something everyone talks about and starts printing money. The hope is that it ends up moving metal as people see it with their own eyes.

Cybertruck is complicated, Its both the best and worst or Tesla. The best that they have the balls to build something significant, new, and without pause. The worst is that it has to be so stark and controversial. But Tesla wouldn't be Tesla without both of these. It's why I love the company and continue to hold. They do what they want without apology knowing it could (but probably won't) fail.....We need more companies willing to fail with products by taking chances. I will say that until the price comes down, it's going to be niche. Those that buy early are going to get wrecked in terms of depreciation (as early adopters tend to do) and they will eventually become as common as Jeep Wranglers IMO. I have decided that I need to wait until founders edition ends and the real price begins before I think about cashing in my reservation when it comes my time.

Overall I haven't yet decided what the long term for Tesla is, when FSD/robotaxi becomes reality I wonder if they will actually sell more than 2M vehicles to private parties. Why bother with the day to day with this market if you can license FSD to Benz, BMW, other high end cars? By this time Tesla will have probably 10M FSD capable cars on the road, they have already said S+X are nice but not in the long term vision. I believe they will corner more than 25% of the near city transportation market, people will probably keep a primary car because of kids and personal items, but the second car will become increasingly a bad financial decision when there are many robotaxi's in the wild. Why bother with owning to commute to work and pay for operating, housing, insuring, taxes, ties, etc. It's going to take a while for Americans to give up their rides, but taxi's all but disappeared in under 3 years to Ridesharing. I'll let the others better with forecasting run the numbers but that seems like an absurd revenue stream that to date nobody is "pricing in". I honestly have tried to not give Optimus any thought because then my mind wanders to 10K share price or some crazy numbers because it's basically unfathomable what the TAM is.

Now the only thing left is to decide if post production and deliveries, should the numbers disappoint do I hope for 140's or just say F it and plow cash in. In the long run it won't matter, but it's nice to get a better price. If FSD progress isn't linear and more logarithmic, then ATH's won't be that far off.

 

[Knightshade]

As good as FSD 12.3 appears to be, how long will be before we see LSD, or Limited Self Driving. Tesla should be able to quickly and easily train a neural network to test drive the cars, self park, load and unload on car carriers etc.

I'd suggest folks thinking "RT NEXT WEEK" spend some time reading the threads from users in the FSD forum- you'll see plenty of examples from folks, including the ones who agree 12.3 is a significant improvement, of it still doing outright and unsafe stuff. Myself included.

I think the free trial might well get some extra subscriptions, but the car is still, very very obviously once you've used it for a little while, not ready to do any driving without a human supervising as well as being ready to always fully take over.

As to car carriers, given the current speed of self parking I expect that would actually backlog the logistics lot compared to having a human do it right now... though Tesla claims self park will be getting faster in the future.

And given you still need humans to lock down the car on the carrier it's not like you'd even be saving labor cost.

 

[Singuy]

Another new FSD rollout, 12.3.3. Just got 12.3.2.1 today and it's much better than the original 12.3 release.

WTF is going on....

https://twitter.com/x/status/1774232797445587133

 

[Quesder]

I have been overall bearish on TSLA over the last 6-9 months. As far as the stock price, I expect it to bottom out over the next 3 months (unless the economy/market takes a real dive). My hope and belief is that we have seen margins bottom out this quarter. I think volume will be limited on the upside and we should hope that overall deliveries match 2023. But as I have said in prior posts, we are in the lull on moving metal but I think profit is going to surprise to the upside. Tesla not opening new factories will be a headwind in production but a tailwind in profit. I think Megabuck is going to surprise to the upside and I think we are very very close to FSD making a real impact in earnings.


I honestly have tried to not give Optimus any thought because then my mind wanders to 10K share price or some crazy numbers because it's basically unfathomable what the TAM is.

Now the only thing left is to decide if post production and deliveries, should the numbers disappoint do I hope for 140's or just say F it and plow cash in. In the long run it won't matter, but it's nice to get a better price. If FSD progress isn't linear and more logarithmic, then ATH's won't be that far off.

Thanks for supporting FSD and paid $10,000 (or $12,000?) early, while some people bought in 2016/17 with $6000 turned out to think it is a scam and filing lawsuits.

I really hope TSLA can go to $10k with Optimus, but at this point some are challenging: AI training requires Chips and GPU Cards; how many does TESLA acquire? Do they have more than MSFT + OpenAI, or at least the amount required for Optimus training?

At this point, my 4000 TSLA shares will give me $12000 booking loss if it goes down to $140s. Hopefully that will be the bottom.

 

[2daMoon]

Did you find a supported brake controller for your Model Y?

I have a 22 MYLR w factory towing. Tesla's low voltage system sends 15.5 v to the brake controller wires. Prodigy-Tekonsha told me their operating range doesn’t support 15.5. Tesla disconnected the Aux power to the 7 pin. Which ruled out the RF solution I have.

RedArc has a 24v controller that can handle the 15.5 v. but there was an issue with Tesla and RedArc wiring working out of the box,

I worked with the Rocklin CA GM and team for two years trying to get a supported solution. Was told ‘Corp didn’t expect people to tow.’ Still working on it.

I have a Nucamp TAB 320S Boondock, perfect rig if I can find a supported brake controller. In the meantime , it’s bricked.

Get a DC-DC converter that drops the input voltage into the desired range. Verify it is rated for what ever the current spec is for the brake controller.

 

[Quesder]

Tesla announced that the 6Mth car is produced. Could this be used to estimate the Q1 delivery number?

 

[Buckminster]

https://twitter.com/x/status/1774302857921245530

Nag possibly gone

https://twitter.com/x/status/1774302983892943176

 

[insaneoctane]

https://twitter.com/x/status/1774302857921245530

Nag possibly gone

https://twitter.com/x/status/1774302983892943176

Hmmmm....no longer beta. Now "supervised". Seems like a significant nomenclature shift.