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

[ZeApelido]

Musk absolutely busting on Buffet on twitter, lol

 

[replicant]

Me too! After years of ridiculing him, I always wondered what took so long.

If you still want to read his drivel then just use another browser without signing into your account.

No need for that, just open an incognito window (Ctrl+Shft+N). The same works for most paywalls.

 

[Beltsbear]

The pedestrian basically stepped in front of the car at the last second. I don’t think Waymo software or the driver were at fault.

The Waymo car could have easily stopped. The pedestrian was in the wrong place illegally crossing but was tracked well in advance. Self driving cars will kill a lot of people if we adopt that type of system that kills people who do stupid things.

 

[SW2Fiddler]

The Semi can make a killing even if it doesn't have any FSD at all. The economics of electric trucks are overwhelming vs. diesel.

Furthermore: Even if you have to pay a human driver with a CDL to sit in every truck, you can probably pay less if the truck mostly drives itself, since it's a less stressful job.

I’m gonna have to dust off the CB radio if truckers are going to be reacting / interacting with AIs. I’d subscribe to Bubba & Rosie blog or YT channel in a heartbeat, uh, duty cycle.

 

[ggr]

Apple processor in ipod and iphone historically always needed less memory compare to samsung for same performance, so custom hardware and software(Tesla) always need less resources then general purspose GPU.

I guess it depends on what you mean by "custom hardware and software".

This is not a completely apropos counterexample, but I was involved in the Supercomputer industry back in the 1980s. There was the "big iron", dominated by companies like Cray, CDC, IBM, Fujitsu, Amdahl. Then there were a number of "mini-supercomputers", startups using different architectures and specialized hardware. Most of them you'll never have heard of: Elxsi, CHOPP, Pyramid, nCUBE, and a few others. What killed them? "The attack of the killer micros". Basically, machines with lots of commodity microprocessors. (I was involved in a project at IBM Research that replaced a vector mainframe with a rack full of Riscsystem 6000s. Got into a lot of political trouble. Deep Blue was a rack full of RS6000s too, which is how I got to meet Gary Kasparov. Long story.)

Companies like Intel (especially), Motorola, National Semiconductor, at the time could throw more money and effort into making better microprocessors than these other companies could improve their specialized hardware. At any given instant, their lead was only 18 months over a machine with a bunch of micros at 1/10th the price (Sequent, Sun, others.)

Fast forward to a decade ago, same thing happened with general purpose micros versus dedicated GPUs, we still need a CPU but it controls an array of GPUs. Fast forward to now, and we see dedicated (but soon to be commodity) NN machines taking part of the market that was dominated by GPUs and vector instruction sets.

I see us at a fork in the road. Who will be the big player in NN processor chips? Who will be the Intel, or NVidea, Bitmain, for NNs? There are a bunch of incumbents who have recognized the space being important (Intel, NVidea). There are a bunch of new startups (LMGTFY). Then there's Tesla. If it was any other company/entrepreneur I'd say that Tesla was not going to succeed in this area, because they would be the supercomputer manufacturers facing the killer micros, and one of the startups would win. But they're not, they are (in this area anyway) a well-funded startup with some of the best talent and a captive market. I think they will lead, not get crushed. I certainly hope so. BUT, to lead, they have to make the chips available to others too; their captive market isn't big enough, and I don't just mean other car manufacturers. They have to open to other entire industries. In other words, they succeed if their hardware is no longer "custom", rather "commodity".

Corollary: Intel and NVidea will not win this race. Incumbents never do, too much baggage. If it isn't Tesla it'll be a different startup. Hmmm, must start looking around and following some of those startups.

 

[Pras]

Musk absolutely busting on Buffet on twitter, lol

Except for Elon, nobody dares to question Buffet (CNBC prints every word from Buffet as letters from bible). Buffet is an investor, he will of course prefer to buy a safe investment than starting a risky venture. So no doubt why he called Tesla's venture in car industry as very risky and now the same comment for insurance. Buffet would rather buy AAPL a decade after they built the first iphone, or AMZN after they have already dominated the US retail industry.

 

[elasalle]

.

Musk absolutely busting on Buffet on twitter, lol

Gieco has no moat

 

[ggr]

The Waymo car could have easily stopped. The pedestrian was in the wrong place illegally crossing but was tracked well in advance. Self driving cars will kill a lot of people if we adopt that type of system that kills people who do stupid things.

Before continuing on this tangent, let's at least clarify that it was an Uber car, not a Waymo one.

Mod: let's not continue on this tangent. --ggr.

 

[Remus]

It's not a lie.
But I understand why people might think that. It comes from a misunderstanding of the development of this tech.
A lot think it's about "programmation", they think the software dev are basically writing code for the car to operate in different environments.

It doesn't work this way. They're not writing any code, they're building a Neural Net, and then training it by feeding it data. Then they correct the NN by watching how many times it did the right thing or wrong thing. The Neural Net adjust itself.

The most basic way to think how a Neural Net is working in the car is basically : there is a straight line, and then a turn. The computer (car) will drive, and when the turn comes, it goes straight ahead and doesn't turn. The developers are just going to point to the Neural Net that it was bad.
Next time, the computer drive, and then instead of going straight ahead, it will turn in the wrong direction. The developers are just going to point to the Neural Net that it was bad.
....
Until the Neural Net does the right thing.

It's called Reinforcement learning.

Reinforcement learning is kind of machine learning that doesn't require labeling (unsupervised learning). It's usage is currently limited at playing games including chess and go.

They are NOT using reinforcement learning or else they don't need the human labour for labeling

 

[Krugerrand]

A few years ago in Fremont CA where Mission Blvd crosses Warm Springs Blvd very close to the Tesla factory a red light camera took a picture of me throwing up my arms and running out of the way of a Mack Truck that was running a red light while I had a pedestrian crossing sign illuminated in my favor.

I just hope they gave that truck driver one large sweet ticket.... But I would love to have a copy of that photo.

Clarification of my funny vote. Not funny you were almost the bug, very funny visual in my minds eye of you with your arms up in the air as a Mack comes bearing down on you caught on traffic camera. Dude who reviews those...omg, I think I want that job. Yes, admittedly I have a dark sense of humor sometimes. Ok, most times.

 

[Remus]

Shortsville Times: Tesla Board of Directors Approve Musk Suicide

They have to personally test the changes before giving it to Elon

 

[Remus]

Except for Elon, nobody dares to question Buffet (CNBC prints every word from Buffet as letters from bible). Buffet is an investor, he will of course prefer to buy a safe investment than starting a risky venture. So no doubt why he called Tesla's venture in car industry as very risky and now the same comment for insurance. Buffet would rather buy APPL a decade after they built the first iphone, or AMZN after they have already dominated the US retail industry.

Let's not forget IBM.

What's considered "safe" is not safe in a highly disruptive environment, which he appears to understand very little. With young people no longer give a dam about their parents' brand, his investment won't have the same return as before

 

[Krugerrand]

Completely agree. Conservative deadline for media & investors, more aggressive deadline internally. Why is that so hard? Especially after countless predictions about timelines have proven to be way too optimistic.

Who says it’s not? What if the internal deadline is more aggressive than announced? In fact, maybe all public deadlines are less aggressive than internal ones? How would we know?

 

[dc_h]

What do you prefer?
A) Elon saying 12months and delivering in 24months
B) Elon saying 24months and delivering it in 30months
C) Elon saying 36months and delivering it in 36months

Because deadlines will determine how hard the teams are working. I as investor is very happy with Elon doing A, the shorts will claim that they were right and that Elon is a fraud, let them. What matters in the end is profit, not accuracy of predictions.

I’m reminded a bit of George Washington. He lost almost every battle, but persisted to win the war. With failures like growing only 75% a year, rather then 100 or 150% or getting FSD done a year or two later then planned, but years ahead of the rest, those are failures we can all be happy about down the road. It may hurt the stock in the short run, but even short term, the stock is approaching 1 times forward one year revenue. It is engineering for lower earnings to lock in a very long term market leadership and possibly a near monopoly in the transportation as a service market. Keep failing Tesla and keep overpromising your way to 500+ billion market cap.

 

[elasalle]

Looking forward to announcement of MY factory in next few weeks.

With new round of funding, hope they can fast track the MY schedule a bit. (6-3 mths)
Also, hope this causes PANA to up it's game and commitment to more battery output in GF1.

For Q2, 90K production was re-iterated in investor call, so hope it works out as planned.

 

[Krugerrand]

Elon’s propensity is to give impossible deadlines both externally and internally and he has done this for decades. I believe he thinks the employees would not treat an internal deadline that is tougher than the external one with the same urgency.

Cons of public hyper-optimistic deadlines:
1) Stressful for employees
2) Possibly shortcuts are taken to hit deadlines
3) Stressful for investors.
4) Reduces Elon’s credibility.

Pros:
1) Employees put in tons of unpaid overtime.
2) Employees come up with stunning innovations. E.g. lessons learned from the impossible 5k / week by 12/17 deadline may have caused wasted effort and capital expenditures, but building the tent may have created innovations that will be replicated over the next dozen Gigafactories.
3) New research indicates employees (especially top caliber ones) are motivated by impossible deadlines. E.g. Getting to Mars in 5 years (or full autonomy next year) inspires way more than getting to Mars in 20 years or full autonomy in 8 years.

I submit that Elon understands the pros and cons very well and has judged that the pros outweigh the cons. In fact I believe it is a key factor in all of his companies successes.

Do you really believe Elon has operated for decades in an unwise fashion?

I think it would be fair to say that Elon is of the opinion and belief that it has to happen now not decades from now, and thus the urgency of the mission plays as big a part in the aggressive deadlines as does Elon’s personality, perhaps more.

 

[Krugerrand]

Maybe it is, maybe it isn't, but better to give the outside world a conservative timeline and then beat it.

Why? Because that makes certain people (you?) feel better? Or it helps your investment? Or it’s expected or what you’re more comfortable with?

See where I’m going with that?

 

[ReflexFunds]

Reinforcement learning is kind of machine learning that doesn't require labeling (unsupervised learning). It's usage is currently limited at playing games including chess and go.

They are NOT using reinforcement learning or else they don't need the human labour for labeling

I think Tesla may be using or may be planning to use reinforcement learning in its simulations (but simulations aren't currently its key training method).
This would be like AlphaGo (and not like AlphaZero). 1) Begin with supervised learning based on high graded human driving actions collected from the fleet. 2) Use reinforcement learning with a reward system to train the driving policy. 3) Train a value network to predict the value of each decision.

The difficulty is that the real world is far more complicated than any computer game. The movements of other people, cars and objects are extremely difficult to model when building your simulation. The simulation training also doesn't help you solve cases the software engineers couldn't imagine and didn't design for. Reinforcement learning in simulation is only really going to work if you already know the case you are trying to solve. This is why Waymo's simulation first approach isn't likely to work, and why in fact Tesla also has the advantage in simulation.

Tesla's fleet will be able to identify and collect data on far more rare problem scenarios. For these scenarios it can use the data collected to feed good driving examples into a supervised learning model in a simulation and improve the driving policy through reinforcement learning. It can also use the real world data to feed a neural network to much better predict and simulate the decisions of other cars and pedestrians. Waymo is using this approach to try and train the other moving objects in its simulations, but I don't see how they can make much progress with this without the volume of data only Tesla has access to.

 

[kbM3]

I think it would be fair to say that Elon is of the opinion and belief that it has to happen now not decades from now, and thus the urgency of the mission plays as big a part in the aggressive deadlines as does Elon’s personality, perhaps more.

Exactly. During WW II, industry rallied to ridiculous levels and achieved miraculous improvements, because of the dire circumstances and pressure.

Elon is waging war against Climate Change.

 

[AlMc]

Exactly. During WW II, industry rallied to ridiculous levels and achieved miraculous improvements, because of the dire circumstances and pressure.

Elon is waging war against Climate Change.

Good analogy except that with WW2 just about everyone was actively trying to help. No one was denying that war.

With Climate Change there are many deniers and many pacifists that either try to block or do not care about the climate war.