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.
Not that TA is everything...but the 20DMA has just passed through the 100DMA; successfully completing the 20:100, 20:50, & 20:200 golden crosses. So long as TSLA continues to trend above $250 SP, we should see the final 50:100 golden cross by Q4 earnings call, upon whichtime we should finally see the Energy suprise coup de grâce #BurntHairAnyone ?
Screenshot_20231219_162612_Chrome.jpg
 
Thoughtful (Extremely Hopeful) Bot Speculation Randy & Brian. Does anyone else love hearing these guys spitball like I do?

Q: Can someone here confirm Bot is in fact currently being used in a Tesla factory? (I thought this was confirmed months ago by a reputable source here)


Pretty sure there was wild speculation about that with nothing to support it.

And seemingly contradicted by Tesla themselves just a week ago saying " it plans to soon start using the robot in its own manufacturing operations"



Also in October this year they were posting job listings to hire people to eventually help it do that

"The Robot Software Engineering team is responsible for designing, building, and integrating various types of mobile robots and vehicles to facilitate the movement of people and parts within and between all Tesla facilities worldwide. This includes the management of AMRs (Autonomous Mobile Robots), Vertical Storage Systems, conveyor system and the potential usage of Tesla Bot on the factory floor."
 
We seem to be back to tracking to the Stephenson Trend Line (blue) with a two year parallel lag (red). Onward and upward.
Since share price performance follows an S curve as opposed to linear trend line, TSLA should be able to reclaim the blue line by 2025 with the red trend line forming the bottom.
It sure feels like TSLA wants to move up.

Screen Shot 2023-12-19 at 6.54.18 PM.png
 
We seem to be back to tracking to the Stephenson Trend Line (blue) with a two year parallel lag (red). Onward and upward.
Since share price performance follows an S curve as opposed to linear trend line, TSLA should be able to reclaim the blue line by 2025 with the red trend line forming the bottom.
It sure feels like TSLA wants to move up.

View attachment 1001162
I just want ATH so I can order a Cybertruck (or perhaps a new X if the rear doors are too small).
 
Thoughtful (Extremely Hopeful) Bot Speculation Randy & Brian. Does anyone else love hearing these guys spitball like I do?

Q: Can someone here confirm Bot is in fact currently being used in a Tesla factory? (I thought this was confirmed months ago by a reputable source here)

Obliged in advance
IMO training is the likely bottleneck and something we have no real hard data on.

I believe Optimus will be training in Tesla factories next year, and other robot brands will be training in customer factories in 2024. But there is a difference between starting training and completing training,

Optimus has perhaps already completed some training tasks, but most of the training done so far that we know about seems to be "proof of concept" and "testing the hardware design".

My (uneducated) guess would be something like an average of 3 months and 2,000 hours worth of computer time to train on a single task, but Tesla could perhaps train 50 tasks in parallel if they had sufficient staff and sufficient training hardware.

They want to build enough bots to support the maximum possible rate of parallel training , but there is not much use in building bots that can't be trained and can't be put to work. There is no point in selling a bot to a customer that the customer can't train. There is not a lot of value in building up bot inventory because production can probably always ramp faster than training can ramp.

Beyond factories, Tesla can also use Optimus in service and delivery.

By the end of 2024 if Optimus could perhaps be trained to at least 10 tasks well, we will probably see small numbers of bots in factories, or service centres doing those tasks.

The relevant question is if Tesla can scale up the rate of training a lot of training hardware seems to be in the pipeline, staff are needed to train the bots but the closer the training can get to "pure vision" the less skill or knowledge the trainer needs. Ultimately we want the trainer simply showing or telling the bot what to do, with the bot being trained off video. Even if this is possible, it still takes time, perhaps some skill to manage the NNs, and likely many iterations.

Eventually training data from earlier tasks will aid in training later tasks, training may become faster and easier.,
 
We will soon have posts about Lending Tree analysis shows ‘Tesla has the highest accident rate among all “. They do disclose that the data comes from Lending Tree applicants.
Everyone should understand that such data is deeply skewed towards people with poor credit and/or poor driving records. In other words their data is sourced to produce deeply negative selection.

of course this will have wide uncritical coverage.
I do not quote it here simply because I refuse to give them any more clicks.
 
Reading multiple views here about 2024...
Am I missing the mark by thinking about supply vs demand as the ultimate driver of 2024 vehicles sold? Forget CT for a moment, but Tesla's vehicle margins have been pretty low and the trajectory has been downward. I feel like that is a significant piece of the puzzle regarding Berlin, Austin, Shanghai and Fremont increasing production rates. Why push them harder and higher if the demand is still ~soft and it would require further reducing prices and margins to move the increased production? Don't current wait times and discounting actions indicate that S&D are currently well balanced?

Another thought is while the TM3 Highlander will increase demand, it will cost reduced margins on the current supply of TM3s to clear them out and will cost Fremont some production down time to switch over...so in reality, it's not purely positives...
 

Day 07: Seven Swans a-Swimming | Cybertourdeforce​

Part of 12 Days of Christmas - Tesla Edition a series (c) by the Artful Dodger, Dec 2023

Over this Yuletide season, I will post a daily installment focusing on Tesla products, past, present, and future (please note that I will express major themes as short-hand bullet points, so I can Yule just in tide ). Here's the series so far:

Day 01: A Partridge in a Pear Tree | Roadster Proof of Concept
Day 02: 2 Turtle Doves | S/X Fraternal Twins go Mainstream
Day 03: 3 French Hens | Model 3 Bets the Company
Day 04: 4 Calling Birds | Model Y Built at Four Factories
Day 05: Five Golden Rings | Semi Breaks Physiks
Day 06: Six geese a-laying | Megapack To Excel

Intro to Part 7: Cybertourdeforce

Cybertruck (CT) is the embodiment of a dream Elon had many years ago: to build the best truck possible w/o artificial limits ('physics is the law, everything else is a suggestion'). The vision for CT was to make it a better truck than a Ford F-150, and a better sportscar than a Porsche 911. CT would be the technological tour de force that sets Tesla apart from all other automakers, and firmly establishs them as the innovation leader in the auto industry.

1. First Principles Design

  • CT would be the first 21st Century truck - U.S. Market full of 70-yr-olds
    • yes, the Ford F-1 pickup debuted 76 years before Cybertruck
    • if you took off its bodywork, you'd still recognize the bones beneath
  • from aerodynamics to bodywork to powertrain, batteries and electricals, Cybertruck would showcase Tesla engineering prowess in every area
  • Elon and his engineers created the best truck they could imagine; if the Market likes it, we'll build two. Or even three.
  • CT camps, it swims, its the Official SpaceX Truck of Mars: Do it all, better.
Lesson 1: No compromises, Wave good-bye to the past

2. Shaping the Future - the 7 secrets of the the Cybertruck

  1. Paintless bodywork - All owners see is shiny stainless steel; all accountants see no billion $ paint shop
  2. Bullet-resistant - Trucks are supposed to be tuff, but often never leave the pavement and can't take the abuse of a real apocalyse (or worksite)
  3. 48v low-voltage bus - How to decrease the amount of copper by half, and provide the power demanded by modern electronic accessories
  4. Steer-by-Wire - Transport aircraft stopped using cables and hydralic lines for control surfaces 5 decades ago, let's bring cars into the present
  5. 4-wheel-steering - Tame the unmanageable wheelbase, but keep the utility
  6. Homelink - big battery unused 16 hrs per day, 50 wks/yr? Integrate it with into your home power ecosystem instead
  7. Range Extender - for those times when you need to tow, baby, tow!
Lesson 2: If you got it, flaunt it (best-of-show)

3. Built it Right: Alien Dreadnaught 2.0

  • Tesla believes in automation in the factory as the primary way to increase productivity (which supports both high wages for workers and low prices for customers)
  • The Fremont NUMMI factory was inherited from a GM/Toyota joint venture, and was barely suitable for purpose to build the Model 3. Elon called Fremont the Alien Dreadnaught v0.5 (Shanghai is 1.0 and Austin is 2.0)
  • Over the few past days, Tesla gave an unprecedented access and an inside look via a hosted tour of the new CT assembly line, specifically showing how sub-assemblies like doors and stampings are made. 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)
  • it is extremely telling that Tesla chose this show'n'tell format along with at least 3 other long-form interviews with senior engineers and designers since the Nov 30 Delivery Event (they're like a proud papa showing off CT, baby)
Lesson 3: How to eat an elephant? 1 byte at a time

Conclusion:​

Cybertruck represents not only the future of Tesla Manufacturing, but also the future direction of the auto industry. Why the future for the industry? Because Tesla will drag them along as they go (beginning with NACS, then gigacastings, soon with 48 volt electrics, eventually with FSD safety). If other carmakers are serious about even having a future, they'll be quick to follow Tesla's lead, which is now unassailable. Well done, Tesla Team (I knew you had it in you).

Next up: Model 2: Space-age tech brought down to planet Earth (for the masses)

Tomorrow's Topic:

Day 08: Eight Maids a-Milking | Model 2 World Car​

 
Last edited:

It case anyone wondered.....

It's just bad insurance data. You can tell that something's wrong when the same report says that Mercury, a brand that hasn't sold a car since 2010, has the lowest accident rate.

The issue is that these are only the accidents reported to insurance. If a Model S and a Mercury both get in accident that requires repairs, which driver is more likely to report it to their insurance?
 
Last edited:
Thoughtful (Extremely Hopeful) Bot Speculation Randy & Brian. Does anyone else love hearing these guys spitball like I do?

Q: Can someone here confirm Bot is in fact currently being used in a Tesla factory? (I thought this was confirmed months ago by a reputable source here)

Obliged in advance

The idea Tesla might ship 100K bots next year is preposterous. I could easily see a couple hundred working in Tesla factories being tested and iterated upon, but Tesla themselves are targeting 2027 to start selling Optimus to customers, so that's the earliest I'd expect to see Tesla sell any bots at all.

On a related note, the more number crunching I do on Optimus revenues, the more excited I get for future valuations of TSLA. I could easily see bot sales becoming the company's main product within 10 years. It's like Tesla cars and energy (and even FSD) were just stepping stones to Tesla's real business: selling robots.