Why is dazpacho still posting here?

Shilling 101

Not at all: have not voiced an opinion on $TSLA stock price.

Tesla Model 3 production at Gigafactory 3 to start in November, says Morgan Stanley

"Morgan Stanley’s team expects 'Tesla will be the leading luxury EV player in China.'

They expect that Tesla will be able to produce 35,000 to 40,000 vehicles at Gigafactory in 2020 and ramp-up to 60,000 units for 2021.

Tesla CEO Elon Musk said that Tesla wants to produce 3,000 Model 3 vehicles per week at Gigafactory 3 in Shanghai by the end of the year.”

https://electrek.co/2019/07/17/tesla...organ-stanley/

I'm not sure how you could sleep at night after that.

• Sebastian Thrun worked with Lidar for 20 years
• He led Waymo
• He seems to not trust/like the founder of the non-Lidar startup (pronto.ai)
• In spite of that he says Lidar may not be necessary for FSD

At the very least you should acknowledge it's possible for a non-Lidar system to work. Having a strong conviction that it is impossible is ridiculous in light of the above.

I think you would benefit from the self-driving nanodegree offered by Udacity (which was founded by Thrun, incidentally). Would be great if you could report back on what you learn, too.

https://www.udacity.com/course/self-...odegree--nd013

You'll first apply computer vision and deep learning to automotive problems, including detecting lane lines, predicting steering angles, and more. Next, you'll learn sensor fusion, which you'll use to filter data from an array of sensors in order to perceive the environment. Then, you'll work with a team to program Carla, Udacity’s real self-driving car.

Tesla CEO Elon Musk said that Tesla wants to produce 3,000 Model 3 vehicles per week at Gigafactory 3 in Shanghai by the end of the year.

And he's got such a good record on this type of prediction, right?

Entry level python skills.

Or a 20 year career in Lidar + leading the Waymo project at Google (Sebastian Thrun, arguably the most experienced person alive in self-driving cars).

Basically the same thing right?

If you want to learn more about self driving (and the viability of the non-Lidar approach), check out the nanodegree on Udacity, or for a more basic introduction, the primer (Intro to Self-Driving Cars).



https://youtu.be/kMRUfARa6Uo

I'm not following delivery numbers tbh. How'd it pan out last quarter?

iirc last quarter he lowered prices like 4 times to meet delivery numbers. will be interesting to see the accounting tricks they use to get close to breakeven

WTF is this? Ban this piece of **** already.

He's either too stupid to know that shills don't outright talk up the stock price (in which case he's so stupid that anything he says is worthless), or he's deliberately being disingenuous in the most obtuse way possible.

Then he does what shills do: quoting days old news yet again.
And to top it, off, highly disingenuously left off the following information:

Here's what Morgan Stanley and Jonas really thinks about Tesla when talking to their private high net worth clients:

One of Tesla's biggest bulls just turned around and trashed the company on a private call with Wall Street

Your "expert opinion" on the impossibility of FSD without Lidar is contradicted by the guy who started Waymo and led it for longer than anyone else.

No comment?

It's not contradicted at all. WTF is wrong with you? I've said repeatedly that vision will work with near human level intelligence in an AI. The trouble is that's many years from being viable, and Musk has (while lying of course, like everything single autonomous driving prediction he's failed at) promised FSD feature complete in 6 months and 1 million robotaxis in 2020.
This is just you being disingenuous again. I've repeatedly said that non-Lidar systems will work just fine in 10 years once the technology for machine learning gets closer to human level intelligence. They don't work on current or near future technology however, and can't, because machine learning is highly inadequate for this task right now.

The bolded is the key word. Says it all really. It's going to be solved in 6 months according to the liar and fraud, Musk, but it "may" be solvable without Lidar on some unspecified timeframe? LOL @ you

LOL, so the guy trying to sell expensive courses on self driving gets himself in the media and says it "may" not be necessary oh and by the way, try my course? How naive are you are?

Once again, show me one working example of a level 3 (let alone level 4) autonomous driving system without Lidar? It doesn't exist. Musk certainly doesn't have one - he has ZERO miles tested on public roads in the very state where his team is.

I've started a GoFundMe for TS to learn the basics of self-driving engineering.

This would be a good investment for everyone as he'll no doubt pop back into the Tesla thread and update everyone with his newfound knowledge and journey.

It's USD $2154 for 6 months access. Let's go!

https://www.gofundme.com/f/5fas4f-to...32XMldUGCtWUuU

Here's an interview with Thrun. Amazing guy. Also head of Kitty Hawk (https://kittyhawk.aero): electric personalized flying drones.

Unforunately you for you I have a history in this very thread of being 100% right about Tesla's self driving progress vs Musk (and his fan's) record of being 100% wrong.

I laid out the evidence in detail why Tesla are nowhere in level 3, let alone level 4, as early as 2015, and because of that none of Musk's fraudulent promises like



promising in 2 years something that hasn't happened 3.5 years later would come to fruition. I was spot on the money. Perhaps Musk should get the autonomous driving microdegree?

This is just weird.

LOL - descapito just deleted his post above quoting Musk after the owning I gave him below:
Ok. Since you like quoting him, let's have a look at Musk's history of predictions in this space:

January 2016: Not only says a Tesla will cross the US unaided, but that it will be available publically in their summon feature in ~2 years:



Result after 3.5 years: Tesla can't even navigate across a city, let alone across the US. Zero demonstrations of even level 3, let alone level 4. Summon crashes into the garage wall.

Jun 2016, over three years ago:

Result: Tesla is still on level 2 and still hasn't even fixed basic dangerous flaws in its level 2 software, like slamming into parked objects or not being able to see giant trucks blocking the road (they repeated their 2016 driver decapitation from this inability in 2019). There are zero autonomous testing miles on public roads in the very place where their team is based.

In January 2017, 2.5 years ago, he's asked

His response:



This has still not happened 30 months later. This isn't some grand prediction; this is a customer asking when he's going to see actual different features for a FSD package he paid $5000 for on Musk promises.


November 2018 promises Advanced Summon in 6 weeks:



37 weeks later (a miss of 600%), it's still not here.

This assclown is now promising to solve level 5 autonomy in 6 months, right before a major capital raise no less and just when the stock is down 30+% for the year?

You're living in the past man.

It's 2019 and Tesla is selling a hundred thousand units of FSD capable hardware per quarter.

Wake up and smell the lithium.

Yes, that's precisely how Musk keeps brain dead worthless losers like you in his corner. He makes grand predictions over and over, fails at them, then makes more. When he does this repeatedly, people who aren't worthless losers start to see the clear and obvious pattern that the guy is lying. People who are, fall for it. I call it the "donkey-carrot" effect and with Tesla/Musk it's strong.

Tesla has never sold 100K cars in a quarter and they certainly won't this quarter either. The cars are also not FSD capable as I catalogued above; the hardware is grossly insufficient. For example, this is their sole camera covering 40 degrees of vision; it can't even handle sun flares. This is not suitable for FSD:

Lithium is what mentally ill people take.

*YAWN*

The guy has been wildly optimistic and I have never believed his claimed timeline about FSD until now. Not that it matters but I do believe he believed it, which is problematic in some ways, but also not without advantages.

It will be pretty obvious in hindsight that you were wrong on this one.

Exhibit A: a single screenshot of a low-resolution youtube video. Proves a lot! Weird that you keep showing that. You claim FSD without Lidar is possible but rate limited by the state of AI (as if you know what state of the art AI is, lol). Yet you keep posting this Youtube screen grab. Why is it relevant? Fusing the various sensors on a Tesla is obv sufficient for FSD providing the ANN is ready.

gg (showers for you)

Here's an accurate representation:
https://player.vimeo.com/video/192179726

Eight surround cameras provide 360 degrees of visibility around the car at up to 250 meters of range. Twelve updated ultrasonic sensors complement this vision, allowing for detection of both hard and soft objects at nearly twice the distance of the prior system. A forward-facing radar with enhanced processing provides additional data about the world on a redundant wavelength that is able to see through heavy rain, fog, dust and even the car ahead.

Yeah, when you're balls deep in Elon love it's a major downer to actually be presented with his history of 100% failure at autonomous driving predictions. Your participation here is bringing all this info, it's great!

Sure you haven't. And what has changed now? Musk is still claiming what he always has. The stock was just tanking and he needed a new carrot to raise in the face of a demand disaster a few months ago. It's what he does. But in terms of evidence, there is zero new evidence that he is any further in autonomous driving.

Yes, giving "wildly optimistic" false timelines to customers (who pay $5000 for "coming soon" FSD vaporware fraud) and investors does have its benefits. Glad you can see that!

LOL. People have said this to me on:
- My predictions from years ago that he'd fail horribly at his autonomous driving timelines
- My claim that $420 "funding secured' was a fraud.
I posted the whole video. It shows the horrible quality of the side cameras. You can't have reliable autonomy with sensors that flawed.
It's a response to your parroting of Musk's false claim that current Teslas have all the hardware necessary for FSD. That video shows they clearly don't.
How do you do sensor fusion when there's only one sensor? You're definitely the dumbest person on 2p2.

That camera I showed where the image is destroyed by the sun's lens flare is the only camera for 40 degrees of vision. The side sonars only go for 8m. Their camera setup is obviously not viable for FSD and no "sensor fusion" is possible on that image.

Everything has changed. Obviously.

Time will prove you wrong on this one. Just you wait.

CSB. Extremely irrelevant (it wasn't me) and lol logic. It could have been glorious on some runouts and you're being outcome oriented.

lol sample size.

also doesn't show the impact on FSD - which probably works fine in spite of that dubious lens flare.

Tesla is well aware of lens flare and all over it.

No it does not. See above. Stupid!

For that to be true, you would have to acknowledge that you are sub-human. I'm not expressing an opinion about whether or not you are, but if you were, this is exactly the kind of **** you might post.

TS on selecting sources: trawl the internet for the worst possible example, disregard any concern for reliability, wildly over-extrapolate from a single (unreliable) source, claim it proves everything when anyone with 10% of a brain would know it proves absolutely nothing.

LOL! This is a preprogrammed route using nVidia's Drive unit and software. And even here on simple preprogrammed country roads it fails, freaking out several times. Prepping for this video is the reason for Tesla'a public horrifically bad disengagement data (it's the one test they've done on public roads) of one disengagement every 3.2 miles (4000x worse than Waymo).

Notice the complete lack of 3D mapping, and of intelligent object tracking that Lidar provides. The only thing they have here is half baked object recognition and lane keeping on very clearly marked roads, the two simplest problems in autonomous driving....they're showing tech that existed 10 years ago in this video.

Thanks for posting this, we had a good laugh when it was put out a while ago.

There are other videos in this thread of Tesla Autopilot with no driver engagement which proves you wrong. Also lol sample size.

I'd love to see you run a clinical trial.

No I don't notice the lack of 3D mapping, how could you possibly notice that from this video alone, are you joking or challenged?

So if there's a single instance of autopilot navigating a short preprogrammed trip successfully, we're at full autonomy? Okay...

I could put a brick on the gas pedal and a brace on the steering wheel and make a car drive without a driver for a while.

That’s not fully autonomous driving.

When you sign up to buy a Tesla, is there some legal stuff you have to sign with the FSD feature to ensure that customers can't go after them later for fraud? There's 0% chance this is going to happen on any existing Tesla models. People are throwing their money away unless they are happy with the enhanced cruise control stuff.