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.
Because you want a boat or a new house?
Because no financial adviser would advise to have all of your money in a single stock?

This is normal behavior for every almost executive in every industry. Stock is a large portion of their pay. If you need money, you sell stock. If you follow a proper investment plan, you sell stock. For many of them, the sales will be automatic.

Gracias is not an executive but one of three VC Directors from pre-IPO era (other two are Jurvets0n {on LoA} and Ehrenpries):

Screenshot (63).png


Gracias has been a steady seller on somewhat of an irregular basis:

Screenshot (67).png


Screenshot (65).png
 
Both are machine learning based not similar enough for you? Tesla's new SoC happens to be a specialized NN computer! NN just happens to be the most important part of the system!

The underlying NN structure and the training mechanism are essentially the same. Actually natural language processing is a much simpler problem than object recognition based on video.
Ummm... apache is a web server and it is written using traditional programming techniques. So is Word. They are not even remotely similar.

And natural language is close to being a provably unsolvable problem (talking general case). For all its corner cases, autonomous driving is nothing like the issues present in natural languages. Parsing synthetic languages (like programming languages) is fairly trivial which makes translation between such languages fairly straightforward as well. Computer generated natural language translation is... a good way to be very confused, embarrassed or both. Maybe its my perspective from being a programmer and studying a breadth of natural languages (asian to middle eastern to european to long dead), but I just don't see how "both use neural networks" has any particular relevance.

But as both programming languages and natural languages are very OT I won't continue further. I was just wanting to make sure I hadn't missed something.

Cheers!
 
It is just 18Wh/cell * 0.5 * 365 = 3.3 GWh. Although it is not clear exactly how Tesla defines the cell's capacity, most likely between 17 and 18Wh i think.
FWIW, EPA docs imply 17.73 Wh per cell.

So reducing this scrap rate could maybe take Panasonic to 27GWh. I'd guess the rest of the shortfall vs 35GWh capacity is ramping up cathode production/supply, reducing line downtime and ramping up staffing.
I think it's based on upgrading the 10 old lines to match the output of the 3 new lines. Carsonnight says the new lines do 400k cells per day vs. 300k for the old lines.

The comment below also suggests to me that Panasonic is actually manufacturing Cathodes at GF1 and not just purchasing from Sumitomo. If this is correct, this is the first I've heard of cathode production on-site. I could be wrong though, I'm not an expert on cell manufacturing and this could relate to a different part of the process.
"On several occasions something has fallen into one of the 16-foot mixers — which contain a blend of chemicals including volatile lithium — inside of the plant"​
Panasonic's machines coat aluminum sheets with a wet slurry of solvents and NCA powder to make cathodes. They buy processed NCA powders from Sumitomo Metal Mining. They also have machines that coat copper sheets with a wet slurry of solvent and graphite (and maybe a little silicon) to make anodes.

Panasonic then bakes these electrodes in huge ovens to evaporate the solvents. Maxwell's Dry Battery Electrode coating process "glues" the powders directly onto the metal sheet. This eliminates the need for the toxic solvents, the huge drying ovens, etc. Assuming it works at scale.

If they're still only at 84% yield, there's still a lot of money left to wring out of the system.
Indeed. I doubt Tesla buys their scrap, though, so the savings should go to Panasonic,
 
So now reports (Fred Alert) of a Model S/X refresh with new motors and 2170 batteries as soon as June. Potentially even with an interior refresh at the same time. With this I think Tesla would easily ramp S/X demand back to 25k+ per quarter.

Panasonic better hurry up and ramp GF1 cell lines to capacity! If this report is true, I wonder what Panasonic will do with its 18650 cell lines in Japan. Anyone think these old 18650 lines could be adapted to 2170? It looks like half the 18650 lines were not operational in Q1 given S/X production levels, so maybe they had a chance to start adapting half the cell lines already.

If this story is untrue, expect Elon to deny it quickly. This will further kill S/X sales until the refresh is released.
 
Last edited:
Apparently lots of EV competition coming in the future,
However, What may clearly differentiate tesla further is a superior FSD,
All else being the same. Superior FSD implies pricing power.

If the car is truly an appreciating asset, it’s “game set match, strictly
Club rules” ( bond, James Bond)

To date we haven't seen any EV that begins to approach any of Tesla's current models, mostly because they're bastardised ICE with a motor and some batteries. Where do I see Tesla's advantages:

- Frunk, both for cargo and safety
- Trunk, just blows all other EV's into the weeds - OK, they need an M3 form with a hatch, but that will partially be resolved with the MY
- Range, nothing else gets close
- Performance, nothing else gets close
- Software/Hardware integration, ditto
- OTA - is anyone else doing this, at all?
- Charging infrastructure, guess what...
- And yes, FSD, that's not to shabby ether and improving all the time

Where they do need to improve, and I've said this constantly for the last 5 years, is the fit, finish and quality of interiors, but that's it, just that.
 
So now reports of a Model S/X refresh with new motors and 2170 batteries as soon as June. Potentially even with an interior refresh at the same time.

Panasonic better hurry up and ramp GF1 cell lines to capacity! If this report is true, I wonder what Panasonic will do with its 18650 cell lines in Japan. Anyone think these old 18650 lines could be adapted to 2170? It looks like half the 18650 lines were not operational in Q1 given S/X production levels, so maybe they had a chance to start adapting half the cell lines already.
The key word used in his story is "battery architecture". I think they may use the 18650's and place them in modules designed like the Model 3 pack internals. The chemistry can stay the same as the Model 3 (if it's not already) they just need better cooling to push higher charge rates. It seems from earlier reports that the supply of 2170s is an issue right now.. Can't see them making that worse in Q3.