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

[⚡️ELECTROMAN⚡️]

I think you guys are all drunk already this Saturday night. Not that we don't have plenty of reasons to celebrate.

 

[MC3OZ]

Tesla Cybertruck, built on Earth for duty on Mars.

Designed by Martians, built by Texans...

 

[Knightshade]

That won’t be a requirement when vehicles become self-driving. It’s that they aren’t self-driving is why we have traffic jams and traffic cops.

You’re not thinking broadly enough, but rather that only one thing changes (that cars are self-driving) when the entire world changes with them.

Just being self driving doesn't remove the need for traffic cops if the signals are out.

Hence why the self driving cars will have to recognize not just signals that are out, but be able to understand, and correctly respond to, various hand signals from said traffic cops.

Waymo has had this recognition/ability for over a year now, so I'd certainly expect it to be included in any L4 version of FSD from Tesla.

 

[heltok]

Tesla Cybertruck, built on Earth for duty on Mars.

It’s not built for Mars at all. Tesla Model M will likely look totally different, have different batteries, use different materials for body, have different shape and different features etc. Temps are different, gravity is different, air friction coefficient is different, road surface is different, ”roads” are different, needs are different etc.

But armored APC-look might be similar. Something like this:

 

[Paul_SF]

It’s not built for Mars at all. Tesla Model M will likely look totally different, have different batteries, use different materials for body, have different shape and different features etc. Temps are different, gravity is different, air friction coefficient is different, road surface is different, ”roads” are different, needs are different etc.

But armored APC-look might be similar. Something like this:


View attachment 610698

some people do seem to think Mars is just Arizona, but a little colder.

 

[insaneoctane]

Impromptu slogan contest for Cybertruck.

How about:

Ugly Piece of Unyielding Metal

Here's 2 versions :
SUV -
Stainless Utility Vehicle
Stainless Unyielding Vehicle

 

[Krugerrand]

Just being self driving doesn't remove the need for traffic cops if the signals are out.

Hence why the self driving cars will have to recognize not just signals that are out, but be able to understand, and correctly respond to, various hand signals from said traffic cops.

Waymo has had this recognition/ability for over a year now, so I'd certainly expect it to be included in any L4 version of FSD from Tesla.

You’re also thinking too small. The cars will communicate with each other.

 

[Knightshade]

You’re also thinking too small. The cars will communicate with each other.

You are thinking too imaginary.

AFAIK Elon has never suggested V2V as being needed (or even in the plans) for FSD- the feature complete promised shortly contains no such ability (nor any suggestion of one being necessary). More on that at the bottom in fact.

Also right near the turn of the 21st century the FCC set aside 75Mhz of spectrum for a V2V standard (DSRC).... earlier this year since almost nobody had bothered using it they proposed new rules that will take away most of it- largely to use for more wifi spectrum.



Even if they did have such ability though it wouldn't solve the actual problem of neither car having any indication of when it's ok to go or not because the signal is out and there's a human cop giving hand signals about which cars get to go or not in the example originally raised.

So in short- the system's gonna need to be able to visually recognize, understand, and respond to such hand signals from a human traffic cop.

Again- Waymos system already does this- so I've no doubt Teslas will too.

Or as a smart man once pointed out:

Once you solve cameras for vision, autonomy is solved; if you don't solve vision, it's not solved

 

[Bunky]

You are right Elon has never suggested V2V as being necessary. However, when you have mastered FSD and have a mature fleet of self-driving cars, it would be strange for them to not communicate.

....also, I think the broader point was that if we get to that point and the cars communicate with each other, signals would not be necessary.

 

[gabeincal]

Crazy, eh?

I'm a little embarrassed about my initial TMC reaction to it as it was unveiled. I was pretty sure Musk was being a prankster and playing a mean trick on us. How could he jerk our emotions around like this? Shattering our dreams in an instant. Certainly, the "real" Cybertruck would be revealed shortly and we could all have a good laugh about it. Musk was grinning ear to ear like a father proud of his child. I was stunned, and not in a good way. It was like a bad dream and there was a sickening feeling in the pit of my stomach. This had to be a cheap stunt made from cardboard and silver paint. I expected the ropes above would shortly rip off that ugly silver carboard, or the cardboard will fall away to the floor, revealing the sleek beauty underneath. I waited in anticipation. My head was spinning as the realization settled in that this really was the "real" Cybertruck and there was no mock bodywork to take off. I posted my disbelief on TMC. This was the monstrosity that Musk was grinning ear to ear about!

Then, things began to change. Musk started describing the exoskeleton and the design started to make practical sense. He talked about how tough the material was and how it needed to be folded to fabricate an exoskeleton from such a hard material without breaking the production presses. I was just starting to understand how light and rigid this truck would be. He described the amazing accessories that were standard on every model. Then he mentioned the price.

I was quickly warming up to this thing, mostly because I finally understood the tough, no-nonsense body made of a material that was unparalleled in the automotive world. I've always appreciated vehicles that were uncommonly rigid and light and this took it to a new, incomparable level. This was a truck that would not only handle heavy loads without complaint, but it would also handle twisty roads like a sports car. It would handle parking lot grocery carts, over-grown trails and would knock alder saplings out of the way with no headlight lenses to shatter, no fragile paint to scratch. It was starting to look exceedingly durable, modern and beautiful. Incredible even. I had to have one and placed my order as the presentation was starting to wrap up. Yes, I had just made the biggest and quickest 180 of my life. I couldn't live without this thing. Now, I wanted it more than I did when I could only imagine how awesome it would be!

I've always believed that form follows function. People who think the Cybertruck is ugly don't understand what makes a good truck (or they don't understand how the Cybertruck is built). Tesla will sell everyone they can make to those who do. Because it's beautiful when one understands how it works. People who spend $80K on a sluggish, gas-guzzling, over-weight, top-of-the-line F-150 King Ranch or Dodge Ram Laramie Longhorn with a flexy frame and fragile body work and windows will become the laughing stock of the truck world. Those kind of trucks are effeminate next to the Cybertruck (and not in a good way). Trucks are not supposed to be effeminate. The buyers of those inferior trucks will dwindle until all that's left are old men with brains that have become rigid and fossilized from years of breathing carbon monoxide, eating cheap pork chops and gravy for breakfast and drinking cheap liquor, and bottom of the barrel coffee while they listen to Fox News and nod their faces. They do these things so they can afford their next fill-up, their next oil change, their next truck payment. Their trucks are the only thing that makes them a man. They will stick to "real" trucks that burn real fuel made from old dinosaurs. It will cost them more for this experience but that's who they are and they will be dead and gone soon enough.

Oh man. What a read... Thank you!

 

[gabeincal]

Obviously this doesn't work when you need a big whack of cash for a big expense right now, but for month to month living expenses, this can be a nice alternative to simply selling shares periodically for those living expenses.

Well how about selling a $1000 CC for Jan ‘23? Pays $7.1k per contract. Sell 14 of ‘em for $100k to use as instant cash (or essentially earning about 1k per week for the next 2 years guaranteed and paid upfront).

Then either part with 1,400 shares in 2 years or see it expiring below strike and take the free premium. Or buy it back at 5% price when the stock / IV drops again.

 

[Knightshade]

You are right Elon has never suggested V2V as being necessary. However, when you have mastered FSD and have a mature fleet of self-driving cars, it would be strange for them to not communicate.

....also, I think the broader point was that if we get to that point and the cars communicate with each other, signals would not be necessary.

If vision is solved I guess I'm not entirely clear what they get having extra hardware used, or software written, just to talk to each other directly?

The cars already feed traffic data back to the mothership and it's used for routing on the whole fleet...As vision gets perfect it's likely going to also report back things like accidents, debris on the roads, potholes, etc.... all of which makes far more sense to report back to the mothership for the whole fleet to be told about rather than having one specific car talking to one other specific car over and over to spread that data....

So if vision is already solved, V2V gets you... what, specifically? I'm open to there being a useful answer to that but vision really covers a lot.

They certainly can't make any assumptions about what to do at a non-working light just talking to each other- at least not until every vehicle on the road is in on the same conversation....and also incapable of being manually driven by someone NOT in on the conversation....

V2V seems like one of those things that sounds great in theory, less great in practice....especially if you've got vision solved.

See again the original V2V standard has been around over 20 years now and basically nobody's bothered doing much with it.

That's not even getting into the nasty stuff hackers could get up to if your car starts relying on what another car 'tells it' rather than what it actually sees for itself or gets securely from Tesla- but that's getting pretty far down a rabbit hole that likely belongs over here someplace-

Autopilot & Autonomous/FSD

 

[Causalien]

How about.

By Texans for Martians

 

[feigen66]

 

[heltok]

So if vision is already solved, V2V gets you... what, specifically? I'm open to there being a useful answer to that but vision really covers a lot.

V2V has been dreamed about in the robotics community a lot. It is certainly not needed for FSD and pretty high hanging fruit. But I has some merits. Exactly what will be communicated and how we will see, but expect that in 20years this will be reality:

For now it might be infrastructure in control at the intersection, but at some point the extra infrastructure will not be needed.

 

[LN1_Casey]

Doing my part to help Q4. And help further with my Model S which should be delivered maybe in two weeks.

I already have a hotwheels Model S, and I'll get the model Model S on Friday-ish, and then my real Model S the week following. Slowing increasing in size!

 

[CyberDutchie]

Mobileye partnering with startup Luminar to develop a robo-taxi fleet:

Luminar and Intel’s Mobileye team up to develop a fleet of robo-taxis

The article says Mobileye aims to bring down the hardware cost of the FSD package to under $5K by 2025.

Does anyone know how that compares to Tesla's FSD hardware cost? Do they have a chance against Tesla, or will it be too little too late?

 

[AZRI11]

WTF is this BS. I'm going on Twitter to inform Elon

 

[Mo City]

Mobileye partnering with startup Luminar to develop a robo-taxi fleet:

Luminar and Intel’s Mobileye team up to develop a fleet of robo-taxis

The article says Mobileye aims to bring down the hardware cost of the FSD package to under $5K by 2025.

Does anyone know how that compares to Tesla's FSD hardware cost? Do they have a chance against Tesla, or will it be too little too late?

One thing we know for sure: Tesla won't be standing still between now and 2025. Moving targets are very hard to hit.

 

[Artful Dodger]

Meanwhile back on the farm:

Cheers!