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.
So the general opinion of the internet was that Sandy Munro was wrong suggesting a knitted or woven carbon overwrap for the Tesla motor. This reply discusses why he suggested such and shows a picture that suggests Tesla has already been working on it.


I suspect Sandy had already seen that photo of the woven motor wrap when he made his comment that he would like to see Tesla move to a woven wrap.

I further suspect that photo is outdated, an early prototype that didn't work out because it expanded too much.

Which would mean that Sandy was fooled by a leaked photo. I mean, look at how crude that thing looks compared to the production version that was wound on a machine custom modified to wrap rotors. Notice the weave of the woven wrap is not continuous. Remember, the job of the carbon fiber is to create a pre-loaded force on the rotor to prevent expansion. How do you do that with a woven wrap? It even looks like there are visible radial tool marks from a measuring device used to measure expansion as the rotor was spinning at high rpm which means the photo is of a prototype used for testing purposes.
 
Man on the street report (and why the H aren't there more of these???)

Yesterday stopped by the local sales and service center. It was just as busy as last time, almost every possible inch of parking taken up by cars, vehicles double parked. Nobody even glanced at me when I parked in front of other cars, basically double parking. There were no longer vehicles out on the grass towards the street, there sure needed to be, perhaps a code enforcer stopped by. THought there was a lawn service trailer open until I saw a truck inside the trailer, maybe it was hauling away trades.

I went into the showroom, only a red 3 on display, a handful of people around, seemed to be waiting on service. Nobody really available to help me with a few questions for my delivery Mar 24-28 if account is telling the truth. I could see some on computers in a locked office, lady came out holding paper and went in the back. Tesla employees coming and going from the lot, seemingly looking for a car, perhaps dropping one off or taking it back for prep or service. I had commented to one guy about how busy they were and he said it has been like that all week.

Finding no help, I went back out in the lot, tracked someone down, got my simple questions answered, and went on my way.
1. I asked about Tesla trade in prices vs Carmax, got a non committal answer that sometimes Carmax offers are better.
2. New cars are being delivered complete- nothing is missing from them.
3 FSD does not roll over automatically (if purchased), one has to get back in line. DARN, we love it.
4. Hitch comes with receiver only.
5. No more shiny piano black consoles.

As a formerly employed hard worker, it used to bug me to have others sitting around. When visiting stores today, it bugs me to not have help when employees are gathered chatting and not helping. Makes me think they feel too confident in their jobs.

Everybody in uniform at TSLA was abuzz with worker bee energy!
 
Don’t see how you can be so bearish in regards to a company whose products are so difficult to criticize.

TL;dr: The pros: all the very important things we all know about…performance, tech, value, ease of use.

The negatives: the silly things we all put up with…he doesn’t live near a service center, panel gaps, he’s a boomer and thinks buttons are luxurious so he prefers Audi and BMW ‘luxury’


Good to hear both sides.. even from a bonehead. I agree with most points he makes.
Don’t agree with his price targets. Hopefully he changes his mind once he gets his Y. Build quality seems so much better on our model Y.
 
When you first invested in TSLA you were here:
View attachment 779781

When TSLA split you were here:
View attachment 779782

You are now here:
View attachment 779775

When Tesla Network launches:
View attachment 779778

When Bot starts cleaning your couch:
View attachment 779779

When Bot starts cleaning your multiple Mars couches:
View attachment 779780
Good post, but I think this one more accurately captures my current state:
images.jpg
 
So the general opinion of the internet was that Sandy Munro was wrong suggesting a knitted or woven carbon overwrap for the Tesla motor. This reply discusses why he suggested such and shows a picture that suggests Tesla has already been working on it.
I was one of those who thought he was wrong, and as @StealthP3D points out he still may be.
 
Wow, this is not a great sign for F investors. One of the 5 or 6 impassable moats is the dealer model.


“We’re going to bet on the dealer franchise system,” Farley said. “That’s a different bet than I hear from others. But we’re going to do it by asking them to specialize.”
Ugh.
 
I've never heard anyone who's known Elon for so long to speak so frankly about him. Flat out stating that Elon was not the founder of Tesla, (even after Elon sued to claim that title, which I always thought was petty of him), and Elon's difficulty in admitting when he's wrong, specifically in regards to lidar. His argument that cheap and compact lidar will make it inevitable is interesting though I hope he's wrong and Elon is correct that it's not needed.
 
Interesting to see who is pushing this:

 
Only Options Market Makers have the exemption which allows them to sell naked short shares (Rule 201 "Madoff Exemption" to SEC Regulation SHO). For all other market participants, naked short selling is illegal.

If your theory is true, then this must be conducted by organizations with these market priviledges. That narrows down the list of suspects to about 28 TSLA Options Market Makers

To be very specific:
And the crucial detail:

I have bolded and underlined the crucial point. To understand one needs to carefully read and absorb every singe thing. Very, very careful understanding of the 'Madoff Rule" shows that @Artful Dodger 's infamous 28 can run naked shorts indefinitely without rule violations by simply borrowing shares for a few specific hours every couple weeks, thus complying with the letter of the rule. In this case the 'letter' and 'intent' are the same. Further nobody is actually positioned to enforce violations anyway.
Rule 204 – Close-out Requirement. Rule 204 requires brokers and dealers that are participants of a registered clearing agency[8] to take action to close out failure to deliver positions. Closing out requires the broker or dealer to purchase or borrow securities of like kind and quantity. The participant must close out a failure to deliver for a short sale transaction by no later than the beginning of regular trading hours on the settlement day following the settlement date, referred to as T+4. If a participant has a failure to deliver that the participant can demonstrate on its books and records resulted from a long sale, or that is attributable to bona fide market making activities, the participant must close out the failure to deliver by no later than the beginning of regular trading hours on the third consecutive settlement day following the settlement date, referred to as T+6. If the position is not closed out, the broker or dealer and any broker or dealer for which it clears transactions (for example, an introducing broker)[9] may not effect further short sales in that security without borrowing or entering into a bona fide agreement to borrow the security (known as the “pre-borrowing” requirement) until the broker or dealer purchases shares to close out the position and the purchase clears and settles. In addition, Rule 203(b)(3) of Regulation SHO requires that participants of a registered clearing agency must immediately purchase shares to close out failures to deliver in securities with large and persistent failures to deliver, referred to as “threshold securities,” if the failures to deliver persist for 13 consecutive settlement days.[10] Threshold securities are equity securities[11] that have an aggregate fail to deliver position for five consecutive settlement days at a registered clearing agency (e.g., National Securities Clearing Corporation (NSCC)); totaling 10,000 shares or more; and equal to at least 0.5% of the issuer's total shares outstanding. As provided in Rule 203 of Regulation SHO, threshold securities are included on a list disseminated by a self-regulatory organization (“SRO”). Although as a result of compliance with Rule 204, generally a participant’s fail to deliver positions will not remain for 13 consecutive settlement days, if, for whatever reason, a participant of a registered clearing agency has a fail to deliver position at a registered clearing agency in a threshold security for 13 consecutive settlement days, the requirement to close-out such position under Rule 203(b)(3) remains in effect.
 
So there's 2 issues here.

First is, Tesla never recognizes revenue for undelivered features.... that $ actually sits on the books as a liability until they deliver. So it's not like this is padding their profits and would have to unwind that if they had to refund it.

Second, the only thing left to deliver to anyone who bought after March 2019 is level 2 autosteer on city streets. That's the only undelivered thing on the list of features promised during the sales process for FSD since then.

They still owe considerably more to the pre-3/19 buyers, but those make up an increasingly small % of the fleet-- and they only paid ~3k for FSD anyway.







It was 3 years ago.

They've made minor edits since- but that's when the product changed from a generalized promise of (at least) L4 self driving, to a promise of specific features... all at L2... and all of them already delivered by now other than the autosteer on city streets bit.


(and to be clear, I fully expect if Tesla does successfully get a working system beyond L2 it'll be delivered to buyers of both versions of FSD, even if they only technically owe it to the pre-3/19 ones-- the significantly narrowed promises post 3/19 only are relevant for: Revenue recognition purposes... and avoiding additional legal liability if they aren't able to get beyond L2 on those cars)
I didn’t realize the revenue sits on the books as a liability. Smart move and directly answers my concern. Thanks for the info. When they realize it it will make for a crazy good quarter I suspect.
 
FSD beta v11 is even BIGGER and MORE IMPORTANT than that. V11 is "Single Stack", which means Tesla will use one unified neural net for both city and highway driving.

Better for the tech forums, but FSD is not "one unified neural net' it's a bunch of them, sometimes with one feeding results to another. The single stack thing is just that they'd not be running an entirely different bunch of them for highway use vs city streets use, they'd use the same bunch for both.


Once proven, this single stack will become the basis for standard autopilot deployed on all cars sold in N. America.

All new ones certainly.

What's interesting there is it implies those still with HW 2.x will stop getting improvements/updates, since the HW isn't capable of running that code, and with AP1 as history it's unlikely they'll put much/any effort into future updates of an abandoned codebase. It's a relatively small % of the fleet at his point, and might encourage more of them to buy FSD to get the free driving computer upgrade too if the highway stack sees significant improvements from it that they otherwise can't get.



Why is this so important? Because it means that once "single stack" becomes the production version, then ALL Tesla cars will be gathering data for FSD

This... misunderstands a lot.

First- again HW2.x cars can't run this code, so it'd actually REMOVE some folks from the useful data gathering pool since today those cars provide data for the legacy highway code and they wouldn't in a single stack world. But they're a minor % of the fleet so not a huge deal.

Second- for HW3.x cars, all cars can already gather FSD data. That's what shadow mode is.

Karpathy and others have explained this numerous times-- that the entire fleet (that can run the code) provides data for them.

Green has explained how it works on twitter a number of times if you want a bit more of a dive into the technical details.... but the really short version is they can send campaigns out to any or all cars, regardless of if they paid for FSD or are using FSD, to collect whatever Tesla is looking for at the time, and stuff will run in the background to look for and collect it- regardless of the active FSD state of the vehicle.


It's a pretty important thing to understand for why Teslas fleet collection data set is so large and powerful- I'm kind of surprised you didn't know this.
 
The EPA fuel economy site just added a new variant of Model Y (the 123 MPGe one). I suspect this is the new Austin 4680 variant. The efficiency is impressive if that’s the performance trim.
8B4FE032-4925-416E-A06A-6F7ADEDA8A1E.jpeg


Edit: We’ll now I’m wondering if it’s a new medium range version. 279 miles range. New LFP battery? Hmmm…
6A22E246-A0DE-48A2-89A0-19263D00592C.jpeg
 
Last edited:
Wouldn't it also say "Performance" if it were?
Don’t know, it doesn’t say “long range” either. Maybe they’re using it for both? I expect they’ll prioritize Performance deliveries from Austin for a while. We’ll know for sure in a month.

Edit: Or maybe Medium Range version with LFP battery?
 
Last edited:
Tesla changed their wording on the FSD on the order page many months ago. In fact I think it was almost a year ago. The new wording only puts Tesla on the hook for reaching feature complete, not full autonomy.

Red-read the description of the FSD
I'm curious, does this mean Tesla must honor early Model 3s FSD with Full Autonomy? Somehow I doubt it, but would hate to trade something in with solid FSD contract for a diluted L5 version.
 
I've never heard anyone who's known Elon for so long to speak so frankly about him. Flat out stating that Elon was not the founder of Tesla, (even after Elon sued to claim that title, which I always thought was petty of him), and Elon's difficulty in admitting when he's wrong, specifically in regards to lidar. His argument that cheap and compact lidar will make it inevitable is interesting though I hope he's wrong and Elon is correct that it's not needed.
I do think Elon is right regarding LIDAR and I am frankly surprised that he being a board member of Tesla doesn’t really understand how Tesla’s approach will win. One of the biggest problems with having a LIDAR is that you have one more sensor that provides you with conflicting data. For example, if there is a plastic bag on the road, Lidar will tell you “oh my god there is a massive object in front of the car” and wouldn’t know whether it’s a plastic bag or really a big piece of stone. Many people just assume computers can somehow magically decide what object is what or decide to take which data from which sensor ( camera or Lidar ). It doesn’t work like that computers are dumb. Ok you may say, I will build advanced AI to help computers make such decisions. Then the counter is that, if you can really make such an AI that is smart enough to handle complicated sensor fusion, then you can surely make an AI that can just use vision data that can get to human level perception. Either way sensor fusion is dead - why waste compute and resources on hardware that’s essentially a distraction.

Basically with a camera, you can train it identify whether it’s a rock or a plastic bag. Furthermore, LIDAR is no help for various complex scenarios such as when road is covered with snow, deciphering letters and symbols written on road itself and on boards on side of road.

That’s why when any autonomous solution company says they want to do “sensor fusion” , it’s a red flag in my book. Many people don’t seem to understand.

End of the day, FSD is a software problem. It’s not a hardware problem. You go back and design hardware to support your software goal rather than starting with hardware first and go writing software around it.