4 ms·

Gotta love HN comments about Tesla.

Gudoguru 7y ago
Agreed
navigatesol 7y ago
It's not just HN, it's everywhere. Anything that comes out of the mouth of Elon is an instant fact, and is repeated as fact until it's proven false. Then it was merely a prediction or plan. He just went on stage and told investors that their non-existent, full self-driving robot taxi fleet will cut the cost of personal transport by 90% over Uber, and will make the car owners $10-$30k+ ("in some cases more!") in annual income. It's a pipe dream, but now it's talked about like it's already happening. Remember Solar City and the solar shingles? Their solar deployment is 25% of what it was the quarter he made that announcement. "Synergies!". Meanwhile, they are winding down the company. In a year people will forget it was ever a thing.
- loceng 7y ago
  Did you watch the presentation relating to the Tesla chip release - along with watching the fully self-driving vehicle video? They had the lead chip engineer explain their improvements, and they had their AI lead explain why they're at the lead for this - both very experienced and articulate people in their fields. I don't understand how you say it's a pipe dream unless you don't comprehend what they're sharing. If they're outright lying about the operations per second speeds (TOPS) they announced then that's one thing, however I doubt they would. Another thing to realize is Tesla will be self-interested, as they should be to some degree, so how much profit they will allow non-Tesla owned vehicles to operate in the Tesla Network could heavily be dictated by them - so I wouldn't necessarily go out and buy 20 vehicles to inject them for full-time Tesla Network usage, however some people may take that gamble and then it also fronts money to Tesla. Even if the improvements are 1/4 of what they announced, it is significant. Re: Solar City - they've had to redirect battery and other resources to focus on the vehicles. You're making a lot of assumptions. It might help to balance that out with they're working very hard, have very intelligent people on board, their work and effort has already lead them to this level of success, they're working on creating a whole ecosystem on limited resources that they have to manage, and they have the semi and other vehicles lined up for next year - which they already have pre-orders in for as well. I'd say you're a bit too hyper-focused on certain things, and no, I'm not suggesting to not be skeptical and understand they may not meet their goals and on time the 100% that gamblers-as-investors are betting on them to have.
  - navigatesol 7y ago
    >Did you watch the presentation relating to the Tesla chip release - along with watching the fully self-driving vehicle video? Yeah, I did. It was very interesting, but it was also Computer Vision 101. Karpathy essentially explained how neural nets work to a non-technical audience. It was nice. But did you not assume this was exactly what they were doing? I certainly did. Do you think no one at Waymo/Google/Uber/Apple/Lyft/Ford/GM/VW understands this or is trying to do the same or similar? Tesla does have a fleet data advantage, but Elon's "simulated data won't work" is nonsense. >If they're outright lying about the operations per second speeds (TOPS) they announced then that's one thing, however I doubt they would. They weren't lying about the specs, but it isn't industry leading, like they claimed. So now they have a second best chip that they have to produce themselves in small volumes. >Re: Solar City - they've had to redirect battery and other resources to focus on the vehicles. You're making a lot of assumptions. No, you are making a lot of assumptions. I am looking at the data. Since the solar shingle reveal, solar revenue has decreased every quarter. It is now 25% of what it was then. The reviews on Yelp and other sites are terrible. The Buffalo plant is in trouble, and hasn't fulfilled employment requirements. What does any of that have to do with batteries going to the Model 3 in volume in 2018? Again, your whole argument revolves around "Tesla says..."
    - loceng 7y ago
      Of course others are working towards the same. That's the point and value of competition - and what Elon wanted to cause (whether you take it up as clever PR or genuine care) other manufacturers to start producing EVs. The point is in a highly competitive environment every incremental improvement is potentially a huge advantage in getting to market and scaling faster; Elon referenced the exponential aspect of this, especially in terms of the amount of sensor data they're already capturing and able to analyze compared all competitors combined. Citation needed for it not being industry leading, for their specific use case - hoped for TOPS along with maximum energy usage considerations? TOPS alone isn't the only factor, doing operations efficiently is required to keep energy usage as low as possible. Unless the lead engineer and whole chip team is pulling a fast one over Elon's head - he's not an idiot and not going to move forward in a direction he's not confident in. I wouldn't say ~1MM chips per year is small volume, and that's fine that they're producing in small volume - especially because they have the next version coming out in I think Elon said 2 years. They're likewise not building the facility to produce them themselves, they're outsourcing it - I think they said to Samsung in Austin, TX? I'm not making assumptions, and wasn't referring to that data you were referencing. Elon has said they had to divert resources from battery-related production; yes, it could be a convenient false story, however we can agree they're under a lot of pressure and have limited resources - so it's believable enough. My whole argument doesn't revolve around "Tesla says" of your generalized statement - however it does try to put down my argument to elevate yours. If you don't want to give any amount of trust to what Tesla says - yes, they've not been honest/straight-forward in all situations which isn't ideal - then the foundation for your arguments aren't taking into account a lot of the proof points. Just like now I believe you're being genuine in your belief, and that we just have different experiences, understandings, analysis - I suppose it's possible you have financial interest in Tesla failing and haven't disclosed that, however I won't make that assumptions - and will assume you're speaking in good faith. I don't really get what you're defending or your point is? Do you think Tesla will completely fail? Do you believe their stock is overvalued? That all auto manufacturers are now investing shittons of capital towards EV vehicles and infrastructure is perhaps the strongest signal that the market is there, and Tesla is leading it - and Elon believes they're 3-7+ years ahead of everyone else; and during even only 3 years they get get much further ahead in the race.

Agreed

It's not just HN, it's everywhere. Anything that comes out of the mouth of Elon is an instant fact, and is repeated as fact until it's proven false. Then it was merely a prediction or plan. He just went on stage and told investors that their non-existent, full self-driving robot taxi fleet will cut the cost of personal transport by 90% over Uber, and will make the car owners $10-$30k+ ("in some cases more!") in annual income. It's a pipe dream, but now it's talked about like it's already happening. Remember Solar City and the solar shingles? Their solar deployment is 25% of what it was the quarter he made that announcement. "Synergies!". Meanwhile, they are winding down the company. In a year people will forget it was ever a thing.

loceng 7y ago
Did you watch the presentation relating to the Tesla chip release - along with watching the fully self-driving vehicle video? They had the lead chip engineer explain their improvements, and they had their AI lead explain why they're at the lead for this - both very experienced and articulate people in their fields. I don't understand how you say it's a pipe dream unless you don't comprehend what they're sharing. If they're outright lying about the operations per second speeds (TOPS) they announced then that's one thing, however I doubt they would. Another thing to realize is Tesla will be self-interested, as they should be to some degree, so how much profit they will allow non-Tesla owned vehicles to operate in the Tesla Network could heavily be dictated by them - so I wouldn't necessarily go out and buy 20 vehicles to inject them for full-time Tesla Network usage, however some people may take that gamble and then it also fronts money to Tesla. Even if the improvements are 1/4 of what they announced, it is significant. Re: Solar City - they've had to redirect battery and other resources to focus on the vehicles. You're making a lot of assumptions. It might help to balance that out with they're working very hard, have very intelligent people on board, their work and effort has already lead them to this level of success, they're working on creating a whole ecosystem on limited resources that they have to manage, and they have the semi and other vehicles lined up for next year - which they already have pre-orders in for as well. I'd say you're a bit too hyper-focused on certain things, and no, I'm not suggesting to not be skeptical and understand they may not meet their goals and on time the 100% that gamblers-as-investors are betting on them to have.
- navigatesol 7y ago
  >Did you watch the presentation relating to the Tesla chip release - along with watching the fully self-driving vehicle video? Yeah, I did. It was very interesting, but it was also Computer Vision 101. Karpathy essentially explained how neural nets work to a non-technical audience. It was nice. But did you not assume this was exactly what they were doing? I certainly did. Do you think no one at Waymo/Google/Uber/Apple/Lyft/Ford/GM/VW understands this or is trying to do the same or similar? Tesla does have a fleet data advantage, but Elon's "simulated data won't work" is nonsense. >If they're outright lying about the operations per second speeds (TOPS) they announced then that's one thing, however I doubt they would. They weren't lying about the specs, but it isn't industry leading, like they claimed. So now they have a second best chip that they have to produce themselves in small volumes. >Re: Solar City - they've had to redirect battery and other resources to focus on the vehicles. You're making a lot of assumptions. No, you are making a lot of assumptions. I am looking at the data. Since the solar shingle reveal, solar revenue has decreased every quarter. It is now 25% of what it was then. The reviews on Yelp and other sites are terrible. The Buffalo plant is in trouble, and hasn't fulfilled employment requirements. What does any of that have to do with batteries going to the Model 3 in volume in 2018? Again, your whole argument revolves around "Tesla says..."
  - loceng 7y ago
    Of course others are working towards the same. That's the point and value of competition - and what Elon wanted to cause (whether you take it up as clever PR or genuine care) other manufacturers to start producing EVs. The point is in a highly competitive environment every incremental improvement is potentially a huge advantage in getting to market and scaling faster; Elon referenced the exponential aspect of this, especially in terms of the amount of sensor data they're already capturing and able to analyze compared all competitors combined. Citation needed for it not being industry leading, for their specific use case - hoped for TOPS along with maximum energy usage considerations? TOPS alone isn't the only factor, doing operations efficiently is required to keep energy usage as low as possible. Unless the lead engineer and whole chip team is pulling a fast one over Elon's head - he's not an idiot and not going to move forward in a direction he's not confident in. I wouldn't say ~1MM chips per year is small volume, and that's fine that they're producing in small volume - especially because they have the next version coming out in I think Elon said 2 years. They're likewise not building the facility to produce them themselves, they're outsourcing it - I think they said to Samsung in Austin, TX? I'm not making assumptions, and wasn't referring to that data you were referencing. Elon has said they had to divert resources from battery-related production; yes, it could be a convenient false story, however we can agree they're under a lot of pressure and have limited resources - so it's believable enough. My whole argument doesn't revolve around "Tesla says" of your generalized statement - however it does try to put down my argument to elevate yours. If you don't want to give any amount of trust to what Tesla says - yes, they've not been honest/straight-forward in all situations which isn't ideal - then the foundation for your arguments aren't taking into account a lot of the proof points. Just like now I believe you're being genuine in your belief, and that we just have different experiences, understandings, analysis - I suppose it's possible you have financial interest in Tesla failing and haven't disclosed that, however I won't make that assumptions - and will assume you're speaking in good faith. I don't really get what you're defending or your point is? Do you think Tesla will completely fail? Do you believe their stock is overvalued? That all auto manufacturers are now investing shittons of capital towards EV vehicles and infrastructure is perhaps the strongest signal that the market is there, and Tesla is leading it - and Elon believes they're 3-7+ years ahead of everyone else; and during even only 3 years they get get much further ahead in the race.