Tesla, software and disruption • Hacker News Board

Re: Tesla, software and disruption

#81

Post by annerajb » Sun, Sep 02, 2018, 2:09 AM UTC

akira2501 wrote: post #33

frenchie4111 wrote: post #26

Something missing from this article: Tesla doesn't necessarily need to solve the vision only SLAM/Autonomy problem itself. If anyone solves the problem, and is willing to sell it. Tesla can immediately license it and turn their fleet of 200k+ Autopilot 2 cars into self driving cars overnight.

> Tesla can immediately license it Aren't they already licensing the second generation from Nvidia on Nvidia hardware? > Autopilot 2 cars into self driving cars overnight. That's presuming that the current sensor arrays, computing power and their arrangements will be compatible with the new technology.

No. Tesla does not use Nvidia software libraries afaik. They do license the socs and gpus on a board designed by tesla. Based on the drive px1 reference design. Being changed in 6-8 months to a custom made asic made specifically for self driving by Tesla.

Re: Tesla, software and disruption

#82

Post by annerajb » Sun, Sep 02, 2018, 2:10 AM UTC

PAClearner wrote: post #29

mrfusion wrote: post #25

Earlier quoted context omitted.

Speaking of that. I’d think we could possibly train on simulations. We’d have the 3d information plus the rendered 2d camera view.

I don't know if people do this for vision yet but I know some people doing this for RL for robotic control.

Google has a simulation for their cars that does this. Nvidia has a hw suit for this purpose too and for testing autopilot hardware by feeding it data from another rack/server

Re: Tesla, software and disruption

#83

Post by zaroth » Sun, Sep 02, 2018, 2:11 AM UTC

dragontamer wrote: post #52

seldon999 wrote: post #49

Earlier quoted context omitted.

Do you know how to read a financial statement? They have 26% gross margins on the Model S. Model 3 just turned gross margin positive in Q2. Your response is boring and boneheaded. Next.

I'm not talking gross margin. I'm talking raw and simple profits. If I wanted to talk about margins, I would have said "margin" in my post earlier. I said profit. To be more specific, how about the words "Cash Flow Positive" ?? In any case, Tesla has only had ONE profitable quarter since its IPO, and has never had a profitable year.

Tesla is not a company that wanted to nor needed to be cash flow positive until now. They raised cheap capital and applied it to grow their business faster than bootstrapping would have allowed.

Which is why the point about margins is so crucial. No one anticipated such great margins on an electric car. The tear down analysis demonstrates that the $35k standard-range Model 3 is not a unicorn, but will be profitable to make and sell once the production lines have any slack at all at the higher price points.

Re: Tesla, software and disruption

#84

Post by annerajb » Sun, Sep 02, 2018, 2:13 AM UTC

henrikschroder wrote: post #46

amluto wrote: post #15

Earlier quoted context omitted.

> It is the fact that I felt that this is derived from the way Tesla designed the car, it is intuitive just like when the first Iphone introduced. I beg to differ. The iPhone was intuitive in that all the (built-in) apps were nicely laid out so you could easily interact with them while you are looking at the phone and giving it most of your attention. Kind of like a Tesla is nicely laid out so that you can mostly eas…

> To Tesla’s credit, their on/off/drive/reverse control is vastly superior to anyone else’s. They got it from Mercedes, as well as all the other steering wheel stalks (lights, window washers, cruise control), as well as the driver door window and mirrors control pad. Good on them for selecting the best on the market, but it's not their invention.

I don't think his talking about the physical button and the model 3 one uses a custom made one with a different layout. Their old one at one point was a Mercedes one they later switches to another design don't remember from which OEM.

Re: Tesla, software and disruption

#85

Post by annerajb » Sun, Sep 02, 2018, 2:15 AM UTC

empath75 wrote: post #61

I don’t get why Tesla added the self driving stuff at all. Just seems like a lot of wasted investment that keeps the price of the car high. They could have released a much cheaper model 3 without it, and likely deal with less bad press from their cars killing their owners.

It's not mandatory to buy the self driving car. They want it on the car regardless of the customer buys it or not. They have other uses for the hardware.

Re: Tesla, software and disruption

#86

Post by twic » Sun, Sep 02, 2018, 2:16 AM UTC

> There’s an old car industry joke that you can see the organization chart of a car company in the dashboard, and also see that the steering wheel team hates the gear stick team.

So, Conway's Law then! Can anyone tell us more about this car industry adage?

Re: Tesla, software and disruption

#87

Post by ReverseCold » Sun, Sep 02, 2018, 2:17 AM UTC

kjksf wrote: post #75

Animats wrote: post #42

Earlier quoted context omitted.

Especially since Tesla doesn't make cells. They buy from Panasonic. Panasonic competes directly with Tesla in home and bulk energy storage.[1] [1] https://na.panasonic.com/us/energy-solutions/battery-storage...

That product is not made by Panasonic but by Pika Energy ( https://www.pika-energy.com/harbor-battery/ ). Never heard of Pika before? Me neither. Which is why they are not as much a threat to Tesla's Powerwall as you claim. When you order from Tesla ( https://www.tesla.com/powerwall ) you get some inkling of how much it'll cost. When you order from Pika ( https://www.pika-energy.com/purchase/ )? "Call us". I yet have…

The other solar providers are usually done the traditional car way. You find some local company (akin to a car dealership) to install your panels, and they'll tell you how much it costs.

EnergySage does give you some insight to the price before you buy though.

Basically, you don't buy the battery from Pika energy, you buy it from a local dealer.

Re: Tesla, software and disruption

#88

Post by techstrategist » Sun, Sep 02, 2018, 2:36 AM UTC

kjksf wrote: post #76

IBM wrote: post #37

Earlier quoted context omitted.

There's even less of a moat for Tesla in selling batteries to utilities than there is for their cars. And SolarCity was heading to zero. It was acquired by Tesla because Elon had to bail himself out or the myth of Elon Musk as Tony Stark would have been undone. But also because he along with his cousins would have lost a lot of money. That's compounded by the fact that Elon is significantly personally leveraged and a…

At which point making a better product at lower prices stopped being a moat? SolarCity was a totally different business (installing traditional solar panels but mostly the business was financing those installation) and Tesla is winding it down. The future of energy in Tesla is Solar Roof, Powerwalls (residential storage) and Powerpacks (utility-scale storage). In storage Tesla has significant advantages on cost becau…

There may be reason to question whether the solar roof product is as it appears: https://mansionengineer.com/2018/08/10/elon-musk-tesla-and-t...

Re: Tesla, software and disruption

#89

Post by Latteland » Sun, Sep 02, 2018, 2:38 AM UTC

IBM wrote: post #37

pcprincipal wrote: post #28

I read a lot of articles on TSLA and usually fail to see any mention of their power business. Granted, this is a small part of revenue right now ($1.1BN for FY17 of $11.8BN total), but growing fast and with some highly visible data points (Puerto Rico, South Australia). They are going to win multi-billion-dollar contracts if the US or other countries decide to rebuild their grids. As a TSLA shareholder, this opportun…

There's even less of a moat for Tesla in selling batteries to utilities than there is for their cars. And SolarCity was heading to zero. It was acquired by Tesla because Elon had to bail himself out or the myth of Elon Musk as Tony Stark would have been undone. But also because he along with his cousins would have lost a lot of money. That's compounded by the fact that Elon is significantly personally leveraged and a…

This sense that his whole set of companies are some kind of scam that is some fakey leveraged house of cards with fake revenue, growth, and future potential is wrong, and tired. Any company can lose out long term by not innovating and growing, whether that's GM, IBM, Amazon, or Tesla. Tesla is in their innovative growth phase. And they make actual products that are years ahead of their competitors. There's no other solar roofs. Amazon was seen as a company that couldn't make a profit, but everyone can see now that it was a clever set of choices about long term vs short term.

Sure, there's plenty of things tesla could do better, like make reliable production targets, not have to kill themselves. But they do keep growing their production and sales. They make generally excellent products, their customers like them. Their stock is crazily highly valued, but at the same time their company is leading a market, crushing basically the entire automotive industry. There's no need for a conspiracy of financial chicanery to explain how they got here - smart financing, use of incentives that were available to any other company (look how much Nissan got) that were intended to grow a new market. And look, there's a new market for evs.

Re: Tesla, software and disruption

#90

Post by cagenut » Sun, Sep 02, 2018, 2:44 AM UTC

twic wrote: post #86

> There’s an old car industry joke that you can see the organization chart of a car company in the dashboard, and also see that the steering wheel team hates the gear stick team. So, Conway's Law then! Can anyone tell us more about this car industry adage?

yea I loved seeing that but a quick googling didn't net me anything more.