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

[kbM3]

Ridealong: Porsche's Taycan Turbo Absolutely Redefines Speed and Acceleration

Review of the Taycan for those interested.

I believe that at least the top line version will have better performance than either Model S or 3P (note: performance is more than just acceleration).

From my perspective here are the pros:
1) Size: In between Model 3 and Model S. I would like more luxury than my 3, but don’t want a car as big as the S.
2) Performance.
3) Customizability - If it follows Porsche tradition, there may be thousands of different combinations and one can choose exactly the options desired.
4) I love the looks of the Mission E, so if they come close?

Cons:
1) If it follows Porsche tradition, the base model will be decontented and Porsche will bleed you on the options.
2) Probable lack of efficiency and therefore range.
3) No Supercharger
4) Probably way less tech, OTA...
5) Probably inferior autonomy.

In summary I would not choose over Tesla, but there are definitely some things I like.

 

[kbM3]

I have been thinking how the day regulators, developers, and the public unanimously decide "FSD has been achieved". How is this suppose to work?

I am wondering how any company besides Tesla having any authority on the matter? When you only have 10 million or 15 million miles to show for, how can this data be used to determine safety and completion when Tesla can just give regulators 10-20 billion miles to look at?

I mean in the grand scheme of things, having 10 million miles of safety record compared to 20-30 billion is like me in my car with my hands off the wheel for 10 seconds and claim I have achieved autonomous driving because during those 10 seconds I had zero crashes.

And if regulators tell Waymo "I need at least 1-5 billion real world miles in order to tell us how safe your FSD really is"...how will they achieve that if it took them 10 years to hit 10 million miles?

IF the 10M miles had zero accidents and zero safety related interventions, I would think that would be sufficient, at least in the same geography.

One problem we’re already seeing with Self-Driving cars is frustrations they cause for other drivers. I wonder if this will end up being a topic for regulation?

 

[Bladerskb]

Ha, he just sets up a bunch of straw men to knock down. The reason to use radar was never a temporary stopgap until camera’s could work. It uses a wavelength that can cut through common obstructions that cameras can’t see through. And the reason to use radar over LiDAR has nothing to do with cost and everything to do with the fact that LiDAR has none of those benefits.

The guy has no idea what he’s talking about. I guess I shouldn’t be surprised when he says LiDAR uses “lasers combined with radar”

Funny because that's exactly what Elon said.

 

[Christine69420]

Review of the Taycan for those interested.

Thank you - I'm glad for every EV made. One less fossil stinkbomb I have to drive behind.

But how many can they make given the global battery shortages?

Batteries must be one of Teslas greatest advantages.

 

[Bladerskb]

Many credible posters here are doubting what FSD progress the Tesla AI team could possibly have achieved in the 3 years since they split with MobilEye - an automotive vision pioneer with 30 years of history.

Here's a comparison of AP1 (the last MobilEye iteration) and AP2:

Tesla Autopilot 1 vs. AP2: Incredible Improvements On Difficult Road

The Tesla testing takes place on a two-lane road with plenty of sharp turns, blind curves, dips, and elevation changes. Most of the time, you can see by the lane markings that passing isn't allowed on this road, which generally indicates that it's a challenge. Later in the video, there's a complete lack of lane lines.

"As you can see, Autopilot 1 struggles constantly on this drive, and I can assure you, it used to be WORSE on this road. I have previous tests with AP1."

"On this test, AP2.x absolutely NAILS IT! Successfully navigated the test roads"
​

I believe that's what a grounds-up, first-principles approach and the "Tesla data advantage" of a fleet of hundreds of thousands of vehicles gives you in just 3 short years.

And this AP1 to AP2 comparison is still on a HW2.5 basis and not the very latest firmware - HW3 gives ~21x more processing power.

I believe we've seen nothing yet in terms of FSD magic ...

You are comparing a 2014 system with access to one camera versus a 2019 system with access to 8 cameras.
It took Tesla years to catch on to a 2014 EyeQ4 and beat AP1 (arguable since the detections still isn't as accurate & stable).

Why don't we compare the underlying neural networks, their development and deployment? With all the bragging Elon did on Autonomy Day, I went to analyze what camera based neural networks Tesla actually had in production as of Feb 2019 versus what Mobileye had as of Q4 2017.

This proved what every one in here was saying, Tesla is clearly 3-4 years ahead...not

With all the talk on vision you would think that Tesla was leading the field. Their entire talk was actually what Amnon shasua was talking about in 2015 and 2016. Tesla Vision looks like they are still playing catch up. Yet most people here believe Elon will have a Level 5 car that works anywhere in any weather condition on any road condition in about 9 months.

It should be noted that mobileye is doing all of this while using a FRACTION of compute power. Its all about efficient and accurate neural networks.

 

[Creekstalker]

Batteries must be one of Teslas greatest advantages.

Relative to every other EV manufacturer, this is correct.

Relative to Tesla's need to produce a profitable quantity of EVs right this minute, batteries are Tesla's biggest disadvantage. Sounds like this is getting fixed though.

 

[Pras]

Relative to every other EV manufacturer, this is correct.

Relative to Tesla's need to produce a profitable quantity of EVs right this minute, batteries are Tesla's biggest disadvantage. Sounds like this is getting fixed though.

I disagree. Tesla can be profitable at the current scale if that is what it aims to. Battery costs are not as high (note battery vs just cell cost).

Tesla lost money last quarter as many pointed, due to many one offs. Slower sales because it hadn’t launched in all market. Add to that poor planning for that quarter.

But most importantly Tesla is continuing to expand (superchargers, Servicing, factories) and is spending a lot on tech that hasn’t provided much revenue (robotaxi / FSD, TN, solar roof, truck/Y tooling etc). If Tesla slows down it will be profitable. But that is not the goal.

Tesla is very much a growth company and the last cash infusion should remove the need to be profitable at the cost of growing.

 

[kbM3]

You are comparing a 2014 system with access to one camera versus a 2019 system with access to 8 cameras.
It took Tesla years to catch on to a 2014 EyeQ4 and beat AP1 (arguable since the detections still isn't as accurate & stable).

Why don't we compare the underlying neural networks, their development and deployment? With all the bragging Elon did on Autonomy Day, I went to analyze what camera based neural networks Tesla actually had in production as of Feb 2019 versus what Mobileye had as of Q4 2017.

This proved what every one in here was saying, Tesla is clearly 3-4 years ahead...not

With all the talk on vision you would think that Tesla was leading the field. Their entire talk was actually what Amnon shasua was talking about in 2015 and 2016. Tesla Vision looks like they are still playing catch up. Yet most people here believe Elon will have a Level 5 car that works anywhere in any weather condition on any road condition in about 9 months.

It should be noted that mobileye is doing all of this while using a FRACTION of compute power. Its all about efficient and accurate neural networks.

Weird post.

All of us are very aware that MobilEye is far and away the leader in presentations and videos.

 

[Cobbler]

You are comparing a 2014 system with ...

Where can I buy a car TODAY with the MobileEye suite that gives me a better drive assistance than Autopilot?

 

[Oldman]

JFYI, the Tesla Semi has similar weight to comparable Class 8 trucks, which is ~17,000 lbs plus 200-300 gallon of fuel which is another 1,400-2,100 lbs of weight: a modern tractor trailer can easily weigh ~20,000 lbs.

Diesel engines, the steel chassis to distribute all that force plus fuel weighs a lot.

That the Tesla Semi is going to weigh too much has been a TSLAQ lie from day 1.

The semi battery pack will weigh about 10,000 pounds

 

[Singuy]

The way it will work in US is - self regulation. As the Florida law shows, minimal legislation that just makes the company liable for any accidents. Assumption being no company wants bad publicity and millions in law suits so they won’t operate a FSD car unless they are ready.

So then how will any company ever be "ready" if the future is a unpredicteable outcome? And how can any company claim they are ready if there are no standardized metric to go by? It seems that FSD is a moving Target and mathematically 99.99999% with 100% as an unachievable scenario dictates that the company with the most miles/one accident is the winner..and the company with not enough miles(whatever enough is) will be challenged.

For example, Uber prior to their first fatality can be falsely extrapolated as fatality free..until a future event happened making it not fatality free. So now it's measured in fatality per x miles until the next fatality which maybe the next day or the next decade..who knows. But without billions of miles collected like Tesla, it's like they are stuck in the past while Tesla has the ability to better predict future events.

Like I said, Tesla each day worth of miles shouldnt be used to extrapolate safety and yet how can any company extrapolate future predictions using only one day worth of Tesla miles?

 

[hcdavis3]

Bob Lutz on CNBC this morning continued his diss of Tesla, saying that Tesla is not a tech company. Rather Tesla is a car company that uses batteries. It took a little while for the dinosaurs to disappear may take a bit longer for the Lutz dinosaur to disappear.

 

[Tslynk67]

Surprisingly well written article from business insider It's time for Tesla to go into stealth mode for the rest of 2019

He obviously reads this thread as it's exactly what I suggested over a week ago. Just deliver cars, lots and lots of cars...

 

[Tslynk67]

Ridealong: Porsche's Taycan Turbo Absolutely Redefines Speed and Acceleration

Review of the Taycan for those interested.

I believe that at least the top line version will have better performance than either Model S or 3P (note: performance is more than just acceleration).

From my perspective here are the pros:
1) Size: In between Model 3 and Model S. I would like more luxury than my 3, but don’t want a car as big as the S.
2) Performance.
3) Customizability - If it follows Porsche tradition, there may be thousands of different combinations and one can choose exactly the options desired.
4) I love the looks of the Mission E, so if they come close?

Cons:
1) If it follows Porsche tradition, the base model will be decontented and Porsche will bleed you on the options.
2) Probable lack of efficiency and therefore range.
3) No Supercharger
4) Probably way less tech, OTA...
5) Probably inferior autonomy.

In summary I would not choose over Tesla, but there are definitely some things I like.

What were we saying earlier?

 

[elasalle]

China is raising tariffs on $60 billion of US goods starting June 1 (just headline)

 

[pnungesser]

No, it’s easy and it’s being done.

https://www.propertycasualty360.com/2012/06/04/state-farm-offers-driver-discount-for-verified-odometer-readings/

Besides self-reporting, insurers may get data from cell phones, oil change shops, smog checks, etc. The state could follow suit.

Tesla could do it even more easily, reporting OTA.



Weight could and should be part of an odometer-based road tax. In the long run this would be good for Tesla.

I would think it would not be difficult for the states to get odometer readings. Before getting a tag renewal, cars need to get their annual safety/emissions test which is forwarded to the state. If that doesn't include odometer readings, it would be an easy add. When you get your tag, road use taxes could be automatically calculated based upon milage and vehicle type.

 

[Cobbler]

We need a realitycheck here and stop discussing the future or the past. Here's the present:
SP is dropping fast we need positive news on the Maxwell acquisition fast and more signs of shitload cars being sold!

Then we'll have media shitstorm with the autopilot death and the pedo-thing.
Short term: hold tight

 

[CyberDutchie]

He obviously reads this thread as it's exactly what I suggested over a week ago. Just deliver cars, lots and lots of cars...

The only thing I believe would benefit Tesla is advertising. Expect lots and lots of competitor ads as they bring their EV’s to market touting “new” ADAS capabilities, “new” levels of performance and safety, “new” software features, over the air improvements, etc. that have been sold by Tesla since day one.

Majority of viewers will believe what they are fed in the commercials, that’s just a fact.

 

[EinSV]

Ridealong: Porsche's Taycan Turbo Absolutely Redefines Speed and Acceleration

Review of the Taycan for those interested.

I believe that at least the top line version will have better performance than either Model S or 3P (note: performance is more than just acceleration).

From my perspective here are the pros:
1) Size: In between Model 3 and Model S. I would like more luxury than my 3, but don’t want a car as big as the S.
2) Performance.
3) Customizability - If it follows Porsche tradition, there may be thousands of different combinations and one can choose exactly the options desired.
4) I love the looks of the Mission E, so if they come close?

Cons:
1) If it follows Porsche tradition, the base model will be decontented and Porsche will bleed you on the options.
2) Probable lack of efficiency and therefore range.
3) No Supercharger
4) Probably way less tech, OTA...
5) Probably inferior autonomy.

In summary I would not choose over Tesla, but there are definitely some things I like.

The performance of the $140K Taycan “Turbo” (more with options) will be comparable to the $59.5K Model 3. With options it will cost $80-$100K more for similar performance.

No Autopilot. No Full Self-Driving. No SuperCharger network. Probably less range.

And smaller, less practical interior with what sounds like a cramped rear seat — “The Taycan is a four-seater with just about enough space for two adults in the back,” according to the review.

The base Taycan models coming out in late 2020 will likely drastically underperform the current 3P and cost $30-$50K more, have less range, etc.

 

[Pras]

China is raising tariffs on $60 billion of US goods starting June 1 (just headline)

Any impact to Tesla?