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.
My uncle the dentist was the only owner of a Mercedes Benz in my family growing up. They were expensive cars with luxurious comfort that my dad could only dream about.

So my jaw dropped when I learned that the electric Mercedes EQB will be a lot cheaper than the equivalent Model Y when it launches in Norway this spring. The EQB is a 7-seater all wheel drive SUV which costs NOK 477,000. A 5 seater Model Y AWD costs about NOK 593.000 and a 7 seater will probably cost even more.

So a difference of at least $13,000. The Model Y has much better range and twice the number of horse powers. And don't forget the autopilot. But still? Mercedes as the cheaper option? What are they doing? 🤔 🤓

Source in Norwegian: Mercedes ypper til priskrig – 100.000 under Tesla
A ramped up Brandenburg factory will change the European market. While interesting, the situation you describe might be temporary. And I note the EQB is not yet available.
 
I don't know if you're bs'ing us by implying that the AMD system can only work at idle or max power?
I don't know how you have concluded that. Atom is like a 10w processor with like a 3watt idle power because it's for ultra thins, tablets, and cell phones. Amd embedded processor is 45watt at max power, and the gpu is like 175watt at max power. At idle the gpu sips power while amd is also sipping power, but they are not sipping power at 3watts like the atom, and most likely not even idling at 10watts, but 30 wattish combined. I gave what the min and max are, and they will use whatever power management sees fit in between.
 
Not quite. Conplete loss of a cell results in a step change in capacity due to the reason you call out:

For example: a 46p96s 400V pack will lose roughly 2% capacity if a cell drops out (1/46).
An equal kWh 23p192s 800V pack will lose roughly 4% capacity (1/23).


Indeed, in the 12V lead acid case of 1s6p, a cell loss is 100% energy loss (1/1)

(Excluding an advanced BMS that can transfer energy between and group(s) of cells at a rate equal to or greater than the capacity imbalance divided by the full pack depletion time.)
I wonder about 4680 cell capacity and the loss of one cell relative to the number of cohort members and if it limits the max system voltage (reduction in cohort members).
 
More expansion in Shanghai:


The filing showed that expansion will allow Tesla to add 4,000 staff, taking the number of people the plant can employ to 19,000. There will be no change to the models the factory currently produces, according to the filing, which also did not specify how much production capacity will be increased.
 
More expansion in Shanghai:


The filing showed that expansion will allow Tesla to add 4,000 staff, taking the number of people the plant can employ to 19,000. There will be no change to the models the factory currently produces, according to the filing, which also did not specify how much production capacity will be increased.
That seems a solid uptick for Shanghai! another 20-25% production?
 
  • Like
Reactions: Discoducky
Maybe someone can add some context for people like me not so deep into processor tech - I guess a new chip is always better but maybe some details as to why this is important to Tesla? Thanks.
I'd say it's nice to have a waypoint added and reroute in less time.

Prior builds in the past were less demanding but it seems the users always want better features and faster response times.

Like say you want to have your split-screen navigation + a podcast + you just went to change an unrelated setting (like heat) during a waypoint addition/reroute.

Slower CPU might choke a bit and give you a delayed response on the reroute or maybe the UI just feels laggy.

Also from the supply aspect I wouldn't trust Intel as much as I trust AMD. It's just plain good news to see TSLA and AMD working together because both companies see the mutual benefit of working together.
 
He also gave Tesla a $157 trillion valuation by 2030, which would make me a billionaire 🤣

When projecting that far out it's necessary to take share dilution into account. And the likelihood that you might sell too soon thinking it had gone too high, too quickly. If through some amazing process Tesla did manage to reach $157 trillion by 2030, I'm gonna guess problem #2 would be more likely to keep you from reaching $1 billion than problem #1. ;)
 
All I am referencing is the jump from .4 to .5. I am sure at some point there will be at least one wholesale rewrite of the vision architecture and a few rewrites of the path planning architecture.

I don't lose any sleep over these rewrites, and in fact think they're great ways to not get stuck at local minima like Jim Keller says.

On the other hand, Tesla's software tooling, labeled data sets, test cases, simulations, hardware, all are going to be substantially more evolved. Once you're there, getting to better models gets quicker and quicker. I believe that the rate of improvement in FSD may be another S curve and we are just about getting started.

Can't wait for 2023. Or honestly even v11 of FSD Beta.
I second everything you said. We have FSD in our Model Y - which also started on 10.4 and now has 10.5. We had the car drive us from Irvine to San Diego- over 65 miles today without a single disengagement and application of accelerator once as the only intervention! That was starting FSD as soon as pulling out of the garage and taking back control after entering the parking lot of our destination. The one place that accelerator was applied was on the ramp to enter the freeway which has a lot of construction going on - the car was very slow. Luckily no one was behind us so we let the car do its thing with some push. After that it was smooth sailing all the way to the destination parking lot. Here we disengaged quickly because the car was going too fast for a parking lot.

This is my husband’s car, mine is still not on FSD as I am refusing to turn on the insurance monitoring. My husband definitely has more patience than me, so he qualified with 99% score. I know I won’t get that kind of score.

This was the first time I was with him for the long drive - I felt it was great! The drive felt easy since we were not doing most of the driving. I still think robotaxis are 1-2 years out - but for personal use this is already good enough.
 
No the comparable trims are a difference of NOK 37,000........They're playing with their words and I'm surprised you fell for it.

The Model Y NOK 593,000 is with additions for which they do not add those same additions to the Mercedes (that you know on a Mercede's are going to be insanely expensive). A fair comparison is actual Model Y price of NOK 535,000 verses the NOK 498,000 Mercedes........and given it's a cold climate, that 100 miles less range for the Mercedes will be a deal breaker for many there.

My point - perhaps badly made - was that Mercedes has gone from being the car everybody wanted but could not afford - to being an "also ran" using lower price to compete with the car people really want.

Which back in the 70's was what Fiat, Simca and Peugeot offered to my dad when any Mercedes or Volvo was too expensive.