Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

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

This site may earn commission on affiliate links.
So does that mean I'll get FSD in "a few days" (10.69.2.1)? or shortly after AI day (10.69.3)?

My interpretation is that 10.69.2.1 will go out to employees in ~3 days, limited Beta a day after that, wide current Beta ~3 days after that, and then scores 80+ a day or two after that.

So 1-2 weeks from now, for us folks presently in the queue.
 
Our contribution to Q3 earnings:

We picked up our Model Y on Sept 30, 2021. Just got notification for annual Premium Connectivity subscription, which of course we accepted. So under ASC 606 revenue recognition, that's $99/365 * 1 day, or $0.03, recognized in Q3. You're welcome.
That ain’t a contribution ;) /s. /jk

Congratulations!!
 
Our contribution to Q3 earnings:

We picked up our Model Y on Sept 30, 2021. Just got notification for annual Premium Connectivity subscription, which of course we accepted. So under ASC 606 revenue recognition, that's $99/365 * 1 day, or $0.03, recognized in Q3. You're welcome.
We have a new 3, our third (we sold the other 2), coming next month. That’s the estimated delivery date, at least.

We have 5 cybertrucks on order. I expect to take delivery of all of them as long as tesla honors the FSD price we locked in in fall of 2019.
 
Most likely the car sales will go down in The Netherlands as of these months, as well as in other European countries that depend a lot on natural gas.
Things are not good here with respect to energy: the prices of electricity and natural gas have gone through the roof.

A lot of households here use natural gas for their heating, as natural gas was for many years plentiful and cheap from gas fields in our country.
Also, electricity was produced a lot with natural gas, as it produces less CO2 than coal fuelled power plants.
However, because of earth quakes in the province of Groningen as a result of the gas production, this production had to be lowered significantly.
The switch was made to importing gas and the shutting down of natural gas from Russia has made natural gas very expensive.

The government is now contemplating a price ceiling on energy bills of households, as things will likely become worse because of the continuing ending of relatively cheap fixed price energy contracts that many people still have at this point.

I think that the result will be that a lot of people will hold on to their existing car for a longer period; also they will look at the risen price of electricity versus the price of gasoline as fuel for a car.
 
Last edited by a moderator:
I noticed that since we went to vision...FSD on highway tends to phantom brake at the tops of hills. Not sure why it does this, but it does it enough that I rest my foot on the accelerator on every hill now.

My take would be that we as drivers are really way overconfident when driving up to the top of a hill - we expect all to be rosy, because otherwise there would have been signs warning about a dangerous turn right after the top. In reality we really are driving blind, there could be a cow lying right in the middle of the road, so it makes sense to slow down and be prepared for anything.

But statistically nothing happens - takes a bit of a leap of faith or a lot of driving experience.

So I think this phantom braking near the top of hills is a good thing. It's a perfect trap too for TSLAQ's ... too easy to create prepared accidents.

IF the driver is so confident in statistics, let HIM take the responsibility of keeping up the speed.
 
I think that the result will be that a lot of people will hold on to their existing car for a longer period; also they will look at the risen price of electricity versus the price of gasoline as fuel for a car.
It is going to be a rough transition. There will be shocks we’re not anticipating on top of what you’ve mentioned.

Be nice to the gas heads you know. They’re not bad people and they don’t know what’s coming. It’s gonna hurt a lot.
 
My take would be that we as drivers are really way overconfident when driving up to the top of a hill - we expect all to be rosy, because otherwise there would have been signs warning about a dangerous turn right after the top. In reality we really are driving blind, there could be a cow lying right in the middle of the road, so it makes sense to slow down and be prepared for anything.
You should work in ML/AI. This type of thinking could save your company millions or billions. The gap between theory and applied is what you should be getting paid.

Correct me if I’m wrong, you awesome nerds, but cannot NNs learn on their own any possible algorithm? And perfectly map any mathematical function?
 
I do not misunderstand the technology. It is indeed amazing. But right now on city streets it has zero utility.

That's not even close to true for me. Even if I were to make every turn -- which I don't, though I do take over for a number of them -- just taking the mundane bits out where I'm following some line of other drivers for miles through suburban and country roads and stop lights and whatever... that's very nice for me. That's not the part of driving I enjoy.

(It's also not "the city" but it is the part "where highway AP doesn't apply and the +5mph top speed of non-highway AP pretty much cripples it.")
 
You should work in ML/AI. This type of thinking could save your company millions or billions. The gap between theory and applied is what you should be getting paid.

Correct me if I’m wrong, you awesome nerds, but cannot NNs learn on their own any possible algorithm? And perfectly map any mathematical function?

I am awesome and a nerd.

They can learn to perfectly map any mathematical function in the defined range which you feed it data. Hilariously it may require a decent model to train *dumb* functions like logarithm or multplication. But that is not the power of modern NNs. Classification isn't even the power of modern NNs. Heck when you have known processed input features, other classical machine learning algorithms may perform better.

The power of deep learning is the ability to automatically create features from raw data streams, bypassing the steps of manually handcrafting features by the engineer. You don't code in an algorithm for how to track a car - you create the architecture to do so and the optimization will train the model to learn what convolutions to apply to optimally track the car. This is extremely powerful and can scale dramatically.
 
FSD on highway tends to phantom brake at the tops of hills
How I explained it to myself when I was on my 14,200km round trip across Canada/US is that if there are traffic ahead not visible until you reached the crest, then there will likely be phantom braking, especially at dusk or night. But if no traffic then not.
 
Electric Viking Youtube on the new Nio ET3, with comparison to the Model 3

He starts the video with: “the key question is. . . can this car save the company?”

According to the video, ET5 has:
- better styling / more interior and exterior colors
- wider and longer than M3
- M3P-like performance even on the low end model, since dual motors are standard on both
- Optional 700km battery swap for longer trips
- 3 minutes to swap battery
- “Nio House” - a social club
- built for top safety rating
- auto-pilot like driving assist
- 15% higher price than like-for-like Tesla
- Nio doEd not make the type of profit margin that Tesla makes

He does not talk about all the downsides to battery-swapping discussed at length on this forum. I think the reason is to get Nio fans to like the video and to subscribe to his channel, then perhaps later he will convert them over to buy a Tesla or some $TSLA instead.

0C72CD89-25F7-46E9-ABCB-19697BD03E16.png
 
Last edited:
  • Informative
Reactions: M|S|W and wipster