Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

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

This site may earn commission on affiliate links.
they're collecting visual data about road markings, traffic signs, and are also sending CANbus data. How they take visual data from the cameras and convert it into 10kb per kilometer, I have no idea.
No visual data is being collected. Can't be done over CAN, it is very low bandwidth and very few bits are programmable.

Mobileye could be flagging and classifying road data like signs and encoding the type and location, but that is it.

I'd love to see Mobileye improve their data collection and demonstrate it, but when I worked with them it was like pulling teeth for them to innovate.

Edit: I'm done clogging this thread with this conversation. Take it to the engineering thread in the future.
 
It's because he has the P/E of TSLA at 38 which to me is laughable considering TSLA would have just increased earnings 100% and will increase earnings again over 50% (both number I think are way too low btw)

Sorry but you can't have P/E of a company be lower than it's actual earnings growth rate 🥴 Let's just say, I very much disagree with those share price numbers. I also disagree with the dilution amounts per year. 2020 and 2021 were heavily influenced by Elon's compensation plan, dilution going forward will likely be 2% or 0% if Tesla does just a minor stock buy back

Actually you CAN have a company growing 50%+ per year with a PE lower than it's growth rate. It just hasn't been common in recent years due to the market booming. PE is whatever the market deems it to be, there is no hard and set rule for PE Ratios.

I feel similarly to PKE though, we both think PE ratios will compress as we enter a recession and a strong bear market forms. In time TSLA will boom out of the slump as we exit a recession and things recover, BUT I think we are years away from that yet. For the next few years my gut feeling is we won't see a new ATH for TSLA until around 2024 or so, despite record quarters and earnings. The reason would be PE compression.

Of course if PKE and I are wrong then we simply get much wealthier much faster. Whoopsie!!! :p
 
Anyone want to buy a 4680 cell - only $800 😂!

Man, they expect 800-something cells in the pack, so they´ll make $640,000 if that works so they can buy another 10 Model Y with that, nice Ponzi scheme ;).

Sounds like video should drop Wednesday next week.

They didn't order the Model Y, they bought it off someone. I'm sure they paid > $100k for it to get it this quickly.

Then factor in how much they pay employees, overhead, etc.

Sure, they make money, but nothing like what you are making it out to be.
 
  • Like
Reactions: JRP3
re: Munro Live on YouTube, one wonders how they can be sure they will receive an Austin Y with 4680 and not 2170 ;)

Not yet noted here, later in the current video they also state they are heading to IDRA to view current GigaPress production (presumably the 9000) and hope to discover, among other things, who has presses on order (Sandy mentions (mispronounces?) Geeley has a known order for a press.

Because it's not a Long Range, and they are buying it from an existing owner. This is not an order they placed through Tesla.
 
To stay on topic: the China Shanghai situation was an internal political struggle between the Xi Jinping leading clique and Li Keqiang's adversary whose stronghold is in Shanghai.

Quite plausible, but Tesla‘s factory can still be caught in this political struggle cross fire.

In addition, the longer the Ukraine war goes, the more knock on effects against China will emerge. China imports lots of oil and food. While Chinese imported oil from Russia comes from Eastern Siberia, those facilities are going to start falling apart without the help of the major oil services firms which have all left Russia. In less than a year probably. Anyways, I sure am glad Berlin and Austin are coming online!
 
  • Informative
  • Like
Reactions: capster and MikkoJ
No visual data is being collected. Can't be done over CAN, it is very low bandwidth and very few bits are programmable.

Mobileye could be flagging and classifying road data like signs and encoding the type and location, but that is it.

I'd love to see Mobileye improve their data collection and demonstrate it, but when I worked with them it was like pulling teeth for them to innovate.
Does Mobileye's OEDR need significantly more work? If they already have object detection pretty nailed down, then building out crowdsourced maps is the next step and seems much more efficient than a visual system constantly rediscovering all road features each time it approaches.

I've driven cars that I'm guessing are using this Roadmap database, because the ADAS features somehow even pick up temporary road construction signs, new playground zones, and stuff that should not exist in any static map.
 
That sounds a tad too fast of an increase to me, from a realistic point of view. It would be extremely nice if true though!
It sounds a bit fast.......until you consider that Austin has been at an artificially capped production rate since pretty much end of Q1 thanks to the 4680 ramp. They might have the parts/supplies set up to go way faster than that. Supplies/parts that have just been sitting there, waiting for a ramp. Just because the 4680 ramp was going slower than expected doesn't mean that the suppliers for parts slowed down production since those supplier contracts were inked many months ago. So that's how I could see Austin doing a "burst" of 5,000 for a week and sustaining 2,000/week.

Actually you CAN have a company growing 50%+ per year with a PE lower than it's growth rate. It just hasn't been common in recent years due to the market booming. PE is whatever the market deems it to be, there is no hard and set rule for PE Ratios.

I feel similarly to PKE though, we both think PE ratios will compress as we enter a recession and a strong bear market forms. In time TSLA will boom out of the slump as we exit a recession and things recover, BUT I think we are years away from that yet. For the next few years my gut feeling is we won't see a new ATH for TSLA until around 2024 or so, despite record quarters and earnings. The reason would be PE compression.

Of course if PKE and I are wrong then we simply get much wealthier much faster. Whoopsie!!! :p
Ok.......then everything else in the market would need to drop by about 75% then. Because you can't have Google, Amazon, Microsoft, Nvidia, etc....having P/E's that are double their earnings growth rate while TSLA's P/E is lower than it's growth rate :rolleyes:
 
Does Mobileye's OEDR need significantly more work? If they already have object detection pretty nailed down, then building out crowdsourced maps is the next step and seems much more efficient than a visual system constantly rediscovering all road features each time it approaches.

I've driven cars that I'm guessing are using this Roadmap database, because the ADAS features somehow even pick up temporary road construction signs, new playground zones, and stuff that should not exist in any static map.

This is the Chevy Cruiz, Waymo, Zoox, etc. method. It's EXTREMELY geofenced, and the report from NHTSA on crash rates showed that it actually hasn't worked out so well. Crash rates substantially higher than a human driver, and much much higher than Tesla (which is lower than a human driver).
 
It sounds a bit fast.......until you consider that Austin has been at an artificially capped production rate since pretty much end of Q1 thanks to the 4680 ramp. They might have the parts/supplies set up to go way faster than that. Supplies/parts that have just been sitting there, waiting for a ramp. Just because the 4680 ramp was going slower than expected doesn't mean that the suppliers for parts slowed down production since those supplier contracts were inked many months ago. So that's how I could see Austin doing a "burst" of 5,000 for a week and sustaining 2,000/week.


Ok.......lso then everything else in the market would need to drop by about 75% then. Because you can't have Google, Amazon, Microsoft, Nvidia, etc....having P/E's that are double their earnings growth rate while TSLA's P/E is lower than it's growth rate :rolleyes:
The growth rate in this projection is 110% in 2023 compared to 2022 and 48,5% in 2024. By your definition PE should drop by over 50% in one year then...
 
  • Like
Reactions: Thekiwi
This is the Chevy Cruiz, Waymo, Zoox, etc. method. It's EXTREMELY geofenced, and the report from NHTSA on crash rates showed that it actually hasn't worked out so well. Crash rates substantially higher than a human driver, and much much higher than Tesla (which is lower than a human driver).
I'm pretty certain Mobileye's approach is different, these aren't hard HD maps that attempt detailing every little feature and are very rigid. These are crowdsourced maps that are constantly updated based on data coming from the vehicles running the systems. What I'm experiencing is in a random city in Canada, somehow the ADAS recognizes construction zones and other temporary road signage etc.

The Autopilot highway crash stats provided by Tesla are not comparable to something like Waymo, Tesla has had to do 0 reporting of crashes in autonomous vehicles because they're not running anything above Level 2. Let Tesla put true autonomous vehicles on the road with transparent reporting, even geofenced to a part of California or something, and then it'll be apples-to-apples.
 
  • Disagree
Reactions: BornToFly
Share price predictions are a dangerous game at the best of times, but with Tesla its a near impossibility.
In the short term, yes. In the long term not so much, as long as we're talking the SP floor.

I hereby predict TSLA will be over $5000 by 2030. (current share count)

How high it can go is quite impossible to predict indeed. Depends on how many humans have been replaced by Optimi ;)
 
The growth rate in this projection is 110% in 2023 compared to 2022 and 48,5% in 2024. By your definition PE should drop by over 50% in one year then...
And yet your P/E for 2022 is 63.....even though your earnings growth rate from 2022 is 2023 is 110%. It distorts all your numbers for share price. It fundamentally flaws all of your share price forecasting and P/E numbers.

And again, I can list stock after stock after stock after stock after stock and so on of stocks across all sectors of the market that have P/E's 1.5-2X their actual earnings growth and their projected earnings growth rates.

And further, the P/E should be a reflected of the next 5 years of earnings discounted down. There's simply no logic to TSLA having a P/E of 38, 32, and 30 across 2023, 2024, 2025 unless you think we're headed for another 2008 great financial crisis where P/E multiples compress every stock massively (actually worse than the GFC). If so, you should state that and that you also expect the overall market to plummet 50-75% from here. That's the only way the numbers have any logic here.

Not trying to harp on you. But saying TSLA is going to be a sole anomaly at mega market cap is just illogical.
 
Share price predictions are a dangerous game at the best of times, but with Tesla its a near impossibility.
6E2043AE-767E-4C26-976F-163E7F05A61A.jpeg
 
Are those share prices predictions after some sort of split every year or something? 🤔 Because I mean, 970$ for next year and 1,386$ by 2025 seems kind of low.

When I worked on 5 year Strategic Plan forecasts at 3 large multi-national companies, we had a Likely Scenario, a Downside Scenario and then an Upside Scenario. Looking at @PKEllefsen 's numbers I would say the financials look like the Likely Scenario while the P/E multiples (38, 32 and 30 in 2023, 2024 and 2025, respectively) reflect a downside scenario.
Edit: my opinion of course.
 
Discoducky is Googling stuff just like me, they are not an expert in autonomous vehicles and very few people in the world are right now.
I just gotta call this out.

This is a community. This isn't Twitter or some random place on the web. People develop reputations and credibility here. I consider myself "New", but you are wet behind the ears. It's ok to be new and not understand. It's not ok to be rude about other posters.

I didn't know Discoducky had experience with autonomous vehicles, but I do know that other members of the community would call him out on his if he made random claims like that. The thing that makes TMC a bit special is there are subject matter experts here. If you listen instead of shout and argue, you can learn a bunch.

It's ok to be skeptical and a bit contrarian. You just have to be careful about how you do it. This isn't the way.
 
Share price predictions are a dangerous game at the best of times, but with Tesla its a near impossibility.
PE is the least interesting value though unless you're going all in on leaps banking on PE staying at least 60 😅 PE isn't saying much of how they're actually performing (only how bullish people are for the next 5+ years).
And yet your P/E for 2022 is 63.....even though your earnings growth rate from 2022 is 2023 is 110%. It distorts all your numbers for share price. It fundamentally flaws all of your share price forecasting and P/E numbers.

And again, I can list stock after stock after stock after stock after stock and so on of stocks across all sectors of the market that have P/E's 1.5-2X their actual earnings growth and their projected earnings growth rates.

And further, the P/E should be a reflected of the next 5 years of earnings discounted down. There's simply no logic to TSLA having a P/E of 38, 32, and 30 across 2023, 2024, 2025 unless you think we're headed for another 2008 great financial crisis (actually worse than the GFC). If so, you should state that and that you also expect the overall market to plummet 50-75% from here. That's the only way the numbers have any logic here.

Not trying to harp on you. But saying TSLA is going to be a sole anomaly at mega market cap is just illogical.
Your logic doesn't make any sense.

According to my projections tesla is at their peak of their s-curve in 2021/2022. Their net income growth YoY for the following years are as follows: 2022 - 126%. 2023 - 110%. 2024 - 48,5%. 2025 29%. If we assume a somewhat following trend we can guess the growth rate will be going down towards 20% within the next following years. If we are to use the same PE as growth rate for 2025 (which is basically what i've used) 30 PE for 29% growth, then we cant have 110 PE for 2023, nor 48,5 PE which is the growth rate for the following year. With a PE of 50 in 2023 you would lose money
And yet your P/E for 2022 is 63.....even though your earnings growth rate from 2022 is 2023 is 110%. It distorts all your numbers for share price. It fundamentally flaws all of your share price forecasting and P/E numbers.

And again, I can list stock after stock after stock after stock after stock and so on of stocks across all sectors of the market that have P/E's 1.5-2X their actual earnings growth and their projected earnings growth rates.

And further, the P/E should be a reflected of the next 5 years of earnings discounted down. There's simply no logic to TSLA having a P/E of 38, 32, and 30 across 2023, 2024, 2025 unless you think we're headed for another 2008 great financial crisis where P/E multiples compress every stock massively (actually worse than the GFC). If so, you should state that and that you also expect the overall market to plummet 50-75% from here. That's the only way the numbers have any logic here.

Not trying to harp on you. But saying TSLA is going to be a sole anomaly at mega market cap is just illogical.

The fact that you're criticizing my PE for 2022 and 2023 doesn't make any sense. If you want to criticize my PE you at least need to do it for the 2025 numbers. And if the PE to growth ratio is the only thing you have a problem with then the 2025 numbers should be fine (as PE to growth is basically 1 to 1 for the 2025 projection). The problem with raising the PE by 20-50% for 2022 and 2023 is that you will lose money in 2025 if that PE is fine. 2022 and 2023 PE right now is good as they are if the 2025 PE remains unchanged. The only PE worth discussing is 2025 as the growth rate is starting to stabilize at this point.
 

Elon provided more background on the importance of moving to a "unified vector space" for FSD during Part 3 of his interview with Tesla Silicon Valley Owners Club in early June 2022:

(41:45) FSD Beta Development "Unified vector space" | TSVOC on Youtube


"We still don't have a unified vector space, where all the neural nets pour their output into a single, agreed-upon vector space for both fixed and moving objects.​
"That's quite important. If the moving-objects neural net and the static objects network disagree on the position of a car, then you could say this car is in a non-drive space position because of a disagreement between the neural nets.​
"And if they're not running at the same frame rate you can get a time error as well."​

Hearing this level of architecture planning and progress makes me even more confident that Tesla is on the right path to solving FSD, that they understand the task, and they know how to provide solutions within the current development effort. I am confident in their approach.

Paging @Ohmster

Cheers!
 
Last edited: