Poll: Do you have, or plan to get, an electric car?

[betarhoalphadelta]

The news is a few days old, but what the heck happened that the whole Supercharger team at Tesla got fired?

https://electrek.co/2024/04/29/tesla-conducting-more-layoffs-including-entire-supercharger-team/

I haven't had the opportunity to use the supercharger network yet, but everything I've heard about it has been positive.

Hopefully Ionna and other networks were prepared to hire, because a lot of talent just became available.

I don't understand much about what Musk does, nor why people seem to value TSLA so highly. The below is something I put together just shy of 2 weeks ago, so the numbers are likely a little off now.

Let's take a look at a couple stocks and their 12-month forward PE ratio:

• Tesla / TSLA 62.9
• Ford / F 6.22
• General Motors / GM 4.79
• Toyota / TM 10.27
• Honda / HMC 12.98
• Volkswagen / VWAGY 2.54
• Merc / MBGYY 5.33
• BMW / BMWYY 5.87
• Rivian / RIVN (N/A, losing money so no earnings)

Why in the world would anyone think TSLA is worth that multiple? The forward EPS is $2.58. That's Ford territory ($2.08 forward EPS). Toyota's is over $22.

Now, some people say TSLA shouldn't be valued as an automaker; it should be valued as a tech stock. I personally think those people are bonkers, but so be it.

The darling of the entire tech world right now is NVDA which produces the GPUs needed for AI. Red hot. Forward PE multiple of 36.11 on $25.07 forward EPS.

It makes no rational sense.

 

[Kent88]

Just saw a video with Sandy Munro spouting off about how the layoff makes business sense. Kind of sounded like Munroe thought of the Supercharger network as a completed product. I can't see it that way. Yes, the hardware is ahead of the majority of their competitors, and they almost definitely have more plugs than any of their competitors, and I'm sure that those plugs are places very strategically.

But they've just opened up their network to thousands of non-tesla vehicles, are planning on becoming available to thousands more soon, and they do have competitors in ChargePoint, EVgo, ElectrifyAmerica, and others. Also, it isn't clear to me, but it sounds like some of the people laid off were also in charge of maintaining the equipment.

Also, the more I learn about DC fast charging the more I find that software is a huge factor in charging speed. I've learned about the software industry and taken more than a couple programming classes, and one thing I kept hearing is how the maintenance phase of your software is crucial. Customers need you to keep up on the product you sold them, or they're going to find a different product with a company that will back it.

Have we reached a point where charging networks are mature enough that it's no longer beneficial to have the kind of competitive advantage that Tesla has enjoyed for the past (how long has it been?)? I'm skeptical.

 

[mac_1103]

Seems like it was a Musk moment as opposed to a business decision. Word is he did it to prove a (stupid) point to the head of the team.

Not to defend Musk and his moments, but IME most of the points that owners feel the need to prove are relatively stupid.

It makes no rational sense.

Forget it, Jake, it's Wall Street.

 

[Bilsch]

My guess is the advances in FSD have made Elon think about switching strategies from being just a car manufacturer to a company that provides autonomous transport as their core business. If he has cracked it then all the other things they currently do would be a waste of Tesla's time.

 

[betarhoalphadelta]

My guess is the advances in FSD have made Elon think about switching strategies from being just a car manufacturer to a company that provides autonomous transport as their core business. If he has cracked it then all the other things they currently do would be a waste of Tesla's time.

That's 100% true.

And 0.1% likely

 

[Bilsch]

That's 100% true.

And 0.1% likely

I guess it seems more likely to me, someone using FSD v12.3.4 on a daily basis.

 

[Airborneguy]

I use it almost every day but have to take over frequently. The #1 issue being ridiculous lane changes when I’m cruising with the flow of traffic. It never did that in prior versions. I wish we could turn off lane changes completely.

 

[Bilsch]

I use it almost every day but have to take over frequently. The #1 issue being ridiculous lane changes when I’m cruising with the flow of traffic. It never did that in prior versions. I wish we could turn off lane changes completely.

Interesting. I haven't had to intervene since getting v12
Do you have it set on assertive or chill ?

 

[TheDudeLebowski]

Saw this rolling down my road after lunch

 

[Airborneguy]

Interesting. I haven't had to intervene since getting v12
Do you have it set on assertive or chill ?

Chill and I click “limited lane changes this drive” every time I use it.

 

[Rodent]

My guess is the advances in FSD have made Elon think about switching strategies from being just a car manufacturer to a company that provides autonomous transport as their core business. If he has cracked it then all the other things they currently do would be a waste of Tesla's time.

Waymo is far ahead of Tesla on autonomous vehicle technology. Not only is Waymo enabling passenger vehicles, but they're also leading the autonomous semi-truck game.

Waymo technology is fully autonomous - no human driver is required behind the wheel.

In contrast, Tesla technology is classed as driver-assist - it requires a human driver behind the wheel.


https://waymo.com/

 

[Kharnynb]

the fact that the tesla system still uses cameras instead of lidar etc, means it's a no-go for fully automatic, too many issues with weather etc.
As long as these systems require a driver to be alert, i don't really see the point, all it does is give morons a reason to not pay attention and cause accidents when the system does go wrong.

 

[Kent88]

https://www.pbs.org/newshour/show/b...uld-lead-transition-from-fossil-fueled-flight

 

[Bilsch]

the fact that the tesla system still uses cameras instead of lidar etc, means it's a no-go for fully automatic, too many issues with weather etc.
As long as these systems require a driver to be alert, i don't really see the point, all it does is give morons a reason to not pay attention and cause accidents when the system does go wrong.

Or maybe the data says otherwise.

 

[day_trippr]

lol! I totally read that backwards. Carry on!

 

[whoaru99]

Or maybe the data says otherwise.

That looks impressive, but it appears to be Tesla vs everything else. It's not really apples to apples with the point of LIDAR systems vs [visual] camera systems.

I'm not a great statistical guy but for sure I'm always doing the "Hmmm" thing when data is presented that shows such a lopsided result. I don't doubt there is data behind it but how the data is used/presented is a different matter. In the words of Twain(?) "There are lies, damned lies, and statistics."

 

[mac_1103]

But the question is how many of those miles were driven by morons?

 

[Kharnynb]

Or maybe the data says otherwise. View attachment 848709

comparing modern cars with bonus features vs everything else on the road is not really disproving what I said, as long as automatic driving isn't fully hands off, it seems more of a gadget that might cause bad drivers to get distracted and watch movies/phones instead of the road and there is enough info online about camera style systems not working well in fog and rain compared to lidar systems.

 

[betarhoalphadelta]

comparing modern cars with bonus features vs everything else on the road is not really disproving what I said, as long as automatic driving isn't fully hands off, it seems more of a gadget that might cause bad drivers to get distracted and watch movies/phones instead of the road and there is enough info online about camera style systems not working well in fog and rain compared to lidar systems.

This is where I am. I actually think Autopilot is pretty capable. But it's not full L5 autonomy. It's high-end L2 or low-end L3 as I understand it. Which might still be safer than human drivers, but it's not L5.

L5 is when we're talking society-changing capability. And I'm not sure Tesla is on the road to that with a pure vision system.

 

[Kent88]

Until I can plug in an address, hit start, and fall asleep, autonomous driving isn't interesting to me.

There are a few safety features that could come out of it that I appreciate.

 

[betarhoalphadelta]

Until I can plug in an address, hit start, and fall asleep, autonomous driving isn't interesting to me.

Yep. And add in that I can be >0.08 in the front left seat with ZERO legal risk.

 

[Airborneguy]

Well technically you could do that now if the law didn’t stop Tesla from removing functionality that does in fact work.

I’m an FSD complainer, but my issues are related to driving style. I am fully confident that the car can get me from point A to point B safely without my intervention. It will just be slow and very different from my preferred style of driving. Of course it can’t because of the “steering wheel nag” but that’s a mandated feature. It’s not necessary.

And soon, I’m hearing it will be gone anyway.

 

[betarhoalphadelta]

As ridiculous as it might sound, the bar to be cleared isn't that FSD be safer than average human drivers. After all, 90% of humans are above-average drivers, if you ask them.

You have to remember that if you crash, it's your fault. If your car crashes while on Autopilot, it's Tesla's fault, and you can sue the $#!+ out of them. Or, you could, if it was actual true full autonomy. But it's not, it's ADAS. L2 where you have to remain in control because ultimately YOU are responsible.

The bar to be cleared to L5 isn't going to be a computer that's better than a human. It's going to be a computer that is orders of magnitude better than a human. So much better that the computer can do everything, in every situation, and the human is no longer legally responsible for the outcome.

(BTW I'm not moving goalposts here--I've always believed that autonomy is an all-or-nothing thing. It's not like I'm saying that now despite the fact that Tesla continues to get better and better. Until they're 100% there, they're not there.)

 

[Bilsch]

The bar to be cleared to L5 isn't going to be a computer that's better than a human. It's going to be a computer that is orders of magnitude better than a human. So much better that the computer can do everything, in every situation, and the human is no longer legally responsible for the outcome.

This is where Tesla has everyone else beat due to the 1.3 billion miles and counting of data where they learn from every conceivable situation. IMO AI and data is all it takes.

I also don't think it's going to be that difficult for a computer to be an order of magnitude better than human drivers. 3700 auto deaths per day worldwide speaks volumes to that point.

 

[mac_1103]

IMO AI and data is all it takes

So you don't see the sensory method as limiting in any way? (serious question; not rhetorical)

I also don't think it's going to be that difficult for a computer to be an order of magnitude better than human drivers.

I think you're missing his point - he said orders of magnitude, plural, and I'm pretty sure he meant several. Not because one order of magnitude wouldn't be a massive improvement and an undeniably good thing, but because liability will shift from individual drivers to deep pockets car manufacturers. This does seem like a case where you have to be essentially perfect.

 

[betarhoalphadelta]

This is where Tesla has everyone else beat due to the 1.3 billion miles and counting of data where they learn from every conceivable situation. IMO AI and data is all it takes.

I also don't think it's going to be that difficult for a computer to be an order of magnitude better than human drivers. 3700 auto deaths per day worldwide speaks volumes to that point.

This is when/how people overstate what we're capable of with "AI". Tesla was promising 1M robotaxis on the road in 2020. Didn't happen. Musk as recently as Jul 2023 was stating that he thought Tesla would reach L4 or L5 perhaps as soon as the end of the year--it's May 2024 and it still looks to be high-end L2 or low-end L3.

Maybe they truly are on the verge. I guess this would be different than the other thousand times Musk predicted it, then...

And this:

I think you're missing his point - he said, orders or magnitude, plural, and I'm pretty sure he meant several. Not because one order of magnitude wouldn't be a massive improvement and an undeniably good thing, but because liability will shift from individual drivers to deep pockets car manufacturers. This does seem like a case where you have to be essentially perfect.

Full autonomy means I no longer have legal liability for what my car does. I punch in an address and go to sleep in the back seat and if the car is at fault for an accident, that's Tesla's fault, not mine.

 

[Kent88]

 

[Rodent]

or today you can (if you're in one of the available service areas) open your Waymo app, identify where you are and where you want to go, and then wait for your driverless JLR I-PACE BEV to arrive at your pick-up location.

https://waymo.com/waymo-one/

Tesla is waaaay behind on the autonomous driving technology that's live and in use today.

If you're ever near Mountain View, CA and want to see the Waymo fleet in action, venture yourself over to the GoogleX campus on Mayfield Ave and take in the view of empty I-PACE BEV vehicles coming/going throughout the day.

 

[sibelman]

actually charging by use

This is hugely surprising. Was your electricity service a flat rate based solely on your panel's capacity until recently? Or is it tiered by capacity (main fuse size) and charged by usage? In the U.S., we generally pay a small base fee, but the bulk of the bill is usage-based.

 

[Bilsch]

or today you can (if you're in one of the available service areas) open your Waymo app, identify where you are and where you want to go, and then wait for your driverless JLR I-PACE BEV to arrive at your pick-up location.

As long as it’s inside specially pre-mapped geo fenced zones under certain conditions in a car that cost $300k a pop. Not way ahead at all.

P.S. That and the couple people Waymo has over.

 

[Kharnynb]

This is hugely surprising. Was your electricity service a flat rate based solely on your panel's capacity until recently? Or is it tiered by capacity (main fuse size) and charged by usage? In the U.S., we generally pay a small base fee, but the bulk of the bill is usage-based.

there's a base monthly fee by panel fuse(3x25A now, want to go to 3x35), then there is a "carry fee" per mwh. these are by the local network owner, you get no choice, but they are bound by the law on how much they can charge and how much profit they are allowed to make on "net fees".
After that you have to get an actual electricity provider contract, you can pick any provider, and any contract, there's a lot of competition and contracts can be all the way from "market prices per hour" to set prices over a 2 year contract.

 

[Bilsch]

So you don't see the sensory method as limiting in any way? (serious question; not rhetorical)

No since Humans manage with just two eyes and a lot of the time not even that.

 

[mac_1103]

No since Humans manage with just two eyes and a lot of the time not even that.

But I thought your point was that humans suck at this.

 

[Rodent]

As long as it’s inside specially pre-mapped geo fenced zones under certain conditions in a car that cost $300k a pop. Not way ahead at all.

P.S. That and the couple people Waymo has over.

For reasons I cannot share due to NDA, let's just say that I have 'direct experience' with the technologies mentioned and stand by my statement from facts of what is as of 1Q2024.

 

[Bilsch]

For reasons I cannot share due to NDA, let's just say that I have 'direct experience' with the technologies mentioned and stand by my statement from facts of what is as of 1Q2024.

That would be great if you are correct.
I'm hoping for someone, anyone to do this well and bring with it the huge societal benefits of autonomous vehicles. Not the least of which being cheap rides for me to and from the bar.

One of the reasons I wish it to be Tesla is their tech does not require specially prepared and mapped areas. If Waymo is the winner then where I live we'd be lucky to get the service 5-10 years after all the more popular places were completely covered. And considering the cost of their vehicles probably not exactly cheap either.

 

[mac_1103]

Not the least of which being cheap rides for me to and from the bar.

I have a feeling it won't be cheap. You know, demand and all.

 

[Kent88]

I have a bad feeling that real, and recognized as safe, automated driving systems will probably be software as a service that one pays a monthly fee for. It won't be like Microsoft Office from the 90s where you can say you own it (even though you kinda didn't), it'll be like office 365, which I believe you never really install on your device, you basically rent it, and you just use it through a browser.

 

[Kharnynb]

I have a bad feeling that real, and recognized as safe, automated driving systems will probably be software as a service that one pays a monthly fee for. It won't be like Microsoft Office from the 90s where you can say you own it (even though you kinda didn't), it'll be like office 365, which I believe you never really install on your device, you basically rent it, and you just use it through a browser.

i'm afraid you're right with that, there is a huge push in all fields including automotive to go to a pay for services system, luckily on software, there's more choice still, like libreoffice.

 

[sibelman]

basically rent it

The non-free software world has been moving towards subscription models for years. Even if only for security related updates, one may not be best served by the (late, lamented) buy-once pay-once model.

Free software like libreoffice (which I love!) is a different matter. I give them money every year.

 

[Kent88]

Also a LibreOffice fan since back when it was OpenOffice, and a Linux user since the mid/late aughts