fechter said:
OT since it was a Model S, but the video is pretty scary
https://www.youtube.com/watch?v=jznDBN_FBH0
They're suspicious that it was some kind of staged event.
Looks like a classic YouTube lipo pack-stabbing fire video, except car-sized and with no one there to stab it.
The thing that really makes this video special is the fact the cameras were recording BEFORE the fire started, there have been a fair amount of "Parked Teslas" bursting into flames, just none recording the incident from the very start.
Very same Tesla model being the "Model S" fire happened Today, the guy was just driving and noticed something was wrong and pulled over where the car went up in flames. He called the fire brigade immediately but still burned significantly.
https://www.autoblog.nl/nieuws/tesla-in-de-fik-tijdens-rit-op-een-n-weg-te-culemborg-123635
The belief from people who are betting against Tesla is that there are so many Model-S and Model-X mystery fires that their only real question is why doesn't it happen to the Model-3. Apparently, the model-3's have NOT suffered from spontaneous combustion problems like their older counterparts. So in this area, the Model-3 is a winner.
------------------------------------------------------------------------------------------------------------------------------------------
Today was Tesla's "Autonomy Day", where they go over their latest work in self-driving technology.
CNBC which is normally ridiculed for pushing Elon/Tesla's hype was far more in the middle this time around, maybe years of seeing presentations designed to appeal to sharemarket investors makes it easy for them to see it for what it is.
https://www.youtube.com/watch?v=PlILEXbe-cA
[youtube]PlILEXbe-cA[/youtube]
My biggest reason why I haven't believed in non-LIDAR self-driving solutions is the amount of computing-processing power it takes for a mostly optical camera (non-LIDAR) only based driving system to really understand the world around it.
Tesla has put together a custom processor to try and achieve the goals of mostly just optical camera based driving, and their presentation video is far more convincing than their previous stuff.
But Tesla came out and said they have been putting their new processor design in all new Tesla's for roughly the last month now and are still claiming they are more or less 1 year away (which really means 2-3 years away) from FSD.
To me one of the bigger giveaways was when this lady reporter asked about the next released FSD computer and what will it be able to do that the new current one can't do, and the head engineer neural-net processor guy said the most hilariously honest thing to help Elon answer the question, that is he tries to whisper to him "safety"
https://twitter.com/wolfejosh/status/1120527933863530499
https://youtu.be/Ucp0TTmvqOE?t=6194
^Full Youtube Tesla "Autonomy Day" video here.
So what this really means is that while their new FSD computer might be a remarkable leap over what they have had before, to really get true safety FSD level driving it won't happen until they have their next generation even more powerful FSD computer is completed, IMO.
I really think Elon had hoped to come out with even more hardcore promises of timelines rather than just "next year" but while their new FSD hardware is a big improvement he couldn't say "its here" which I believe he really wanted to be able to do so because of the falling "deliveries" numbers. The stock is obviously going to about to get hammered in my opinion.
It seems like its taken Tesla years to finally work out exactly what they really need for really reliable non-LIDAR optical camera based driving and they are inching towards that goal, but aren't there yet.
I think if other companies want to copy the optical-based zero-LIDAR FSD approach it won't take them that much work to catch up but definitely "appears" Tesla has a lead. I actually do now agree that LIDAR is just too expensive and messy long term, I am convinced a cheap LIDAR technology will never come, and that is its biggest problem.
what self-driving cars really do need is a lot of camera optic inputs and an incredibly powerful computer working out what's going on, so far some of the Tesla crashes on autopilot have been so dumb its just scary, https://www.youtube.com/watch?time_continue=135&v=nWaz2jer8Ek
so they still have a lot of work to do.
https://www.marketwatch.com/story/elon-musk-is-just-another-car-salesman-2019-04-22
While folks might like what Elon is doing, it's important to remember just how much "it's about the money" all this can be argued about.
Elon Musk’s CEO pay-to-worker ratio is the highest ever recorded — how it compares at other tech companies
https://www.bizjournals.com/sanjose/news/2019/04/02/tsla-median-pay-musk-compensation.html
Consumer Reports: Tesla Must Prove Safety Before Claiming “Self-Driving” Ability
https://advocacy.consumerreports.org/press_release/consumer-reports-tesla-must-prove-safety-before-claiming-self-driving-ability/
“But instead of treating the public like guinea pigs, Tesla must clearly demonstrate a driving automation system that is substantially safer than what is available today, based on rigorous evidence that is transparently shared with regulators and consumers, and validated by independent third-parties.
https://www.youtube.com/watch?v=7b_y6dSkFpk
[youtube]7b_y6dSkFpk[/youtube]
