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

General Discussion: 2018 Investor Roundtable

This site may earn commission on affiliate links.
Status
Not open for further replies.
I'm assuming GA1 is Model S/X, but why is there no mention of GA2 anywhere?

My best speculation is that the GA2 line is unfinished/abandoned/idle. Maybe it was an early design of the production line, and it didn't work as intended, and then the easiest way forward was to focus on GA3 as well as install a new line in a tent. If so, that at least indicates that the factory should be able to do 10k/wk without the tent, once GA2 is updated to the latest and greatest design. But also suggests that a significant amount of capex spent on Model 3 production equipment was wasted.

Maybe they are in the process of reconfiguring it to be like GA3, or GA4? (Or even better a combination of the best of both of them.) I suspect that even in that case most of the capex isn't wasted, it will just get re-purposed and help eliminate future capex. (Like the capex from the abandoned automated conveyance system that was re-used to make GA4.)

Or Maybe GA2 is related to S and/or X and isn't related to the Model 3 at all?
 
Last edited:
Maybe they are in the process of reconfiguring it like GA3, or GA4? (Or even better a combination of the best of both of them.) I suspect that even in that case most of the capex isn't wasted, it will just get re-purposed and help eliminate future capex. (Like the capex from the abandoned automated conveyance system that was re-used to make GA4.)

Or Maybe GA2 is related to S and/or X and isn't related to the Model 3 at all?
Tesla has said Model 3 has three general assembly lines, so I can't see GA2 being for anything but the Model 3:


But, sure, it may be shut down for updates, and I'm certainly not saying all the capex was wasted. It wouldn't surprise me if some of the machinery is now in the tent, for one thing. But it's always more expensive to redo something than to get things right the first time around.
 
Last edited:
  • Like
Reactions: neroden
Hard to deny the took out all the stops to get to 5000. Doesn’t mean it’s not sustainable, but process improvements are needed. Karpathy and every other exec will quit if they are working full time on GA4.
...
I queried my intern kid and got a confusing emoji and then pic of Converse instead of Red Wings, so that intern is back in Deer Creek. Interns have to be lowest on totem pole so, perhaps GA4 has mostly assembly associates now. So maybe Karpathy is back to Software 2.0 or hand-coding neural network weights. I think factory will fill out with new hire assembly workers and engineers will return to cubicles while production process improves and numbers increase. It's just a matter of weeks.
 
Last edited:
"Take a look at the Diesel smoke-stained exterior walls of every major urban center in Europe and Asia."

It's from centuries of burning wood and coal, but the narrative is compelling.

Not according to the very sharp guide we had at the Colosseum in March! ;) Could even be traced to the rise in popularity of Diesel.
 
GA4:
  • Is linear
  • Is set up in half a tent
    • The other half apparently feeds GA4 and carries away waste (arteries, veins)
  • Restricted to be 2-dimensional because tent (instant skin)
    • But you gain quick setup and alteration
    • You also gain visibility
  • 2-D can be folded
    • Elevate supply path out of the way of waste & other supply arteries - already part of design of prior GA (don’t ask me which one, I haven’t taken my tour yet)
    • Getting into 3rd D takes a building (bones)
      • But can be set up in an existing building if you have figured out what goes where
    • Folding can also be switchbacks in line to optimize use of 2D plane
      • This also helps condense in 4D (time efficiency) by colocating compatible organs, er, assembly stations

What’s-his-twitter ridiculed Tesla in his spy drone videos where for a couple of minutes the line wasn’t moving and folks may have even been standing and looking. But if what you regard is a flattened out map of what you intend to evolve into a 4D machine Machine...
Mustering your AI jocks and your Chip Design chicks and such, to come imagine convoluting it while a sudden army adjusts, documents and measures it all... so that it can be replicated ... could be the Best use of a whole lot of those minutes.
 
June sales advance 5.2%; SAAR soars

June (auto) sales advance 5.2%; SAAR(seasonally adjusted annual rate) soars

U.S. light-vehicle deliveries, boosted by healthy deals, America's surging appetite for light trucks and an extra weekend of sales, rose 5.2 percent in June as the auto industry closed out the first half of 2018 on a high. The SAAR for June came in at 17.47 million, up sharply from June 2017 and May.
 
The Model S Finally Has Competition--From The Model 3--And That's Not Good for Tesla

I'd argue:
  • Tesla is working on a refresh (also they need to solve the overheating issue for electric GP , a problem which does not seem to exist in model 3).
  • Q1 deliveries probably include lots of overflow from december 2017 and thus not directly comparable.
  • MS/MX sales numbers are nowhere near comparable to M3 to claim any cannibalization (they could be higher if tesla offered something like P130D)
  • The goal of Tesla is to advance EV adoption so whether it is S, X, 3 or Y, it really does not matter.
 
Last edited:
You're right; Musk effectively told us that the exit run rate is higher than 5000/w. How? Simple mathematics. For a full week, 7 days, the lines produced 5000 cars. But we also know that they were working that whole week to speed up the line. So at the beginning of the week, they were producing at less than the full rate, and to make up the difference, at the end of the week they had to be producing at greater than the average rate. He also said they'd be at 6000/w in July. You can't get there by producing less than 5000/w. So your innuendo is simply incorrect.

Your theory somewhat checks out if we use a dubious source for the daily rate:

skabooshka on Twitter

I believe Tesla produced at least 4,969 Model 3s in the final 7 days of Q2.
6/24: 547
6/25: 623*
6/26: 698
6/27: 811**
6/28: 856
6/29: 701
6/30: 734

* Missing June 25, I use the midpoint of 6/24 and 6/26.
** Missing data for half of one shift, this number is extrapolated.

(#) Skabooshka is a notorious bear. If he is seeing these numbers, for the life of me, I don't understand why he is bearish!! :rolleyes:
 
Status
Not open for further replies.