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

[ggies07]

Robotaxi is the wildcard if it can be enabled before end of year on the current fleet...whenever it reaches some inflection point like the 7M vehicle mark.

I liked your post for the first part, but I just got my car upgraded to HW3 and have had FSD for a day and I'm not impressed. I've watched dozens of vids and expected it to be like those, but I've had to disengage way too many times:

-it didn't follow the new straight and turn lanes in the new construction area
-it doesn't stay in the middle of the lane, but instead rides the right hand part almost on the line (autopilot has better handling)
-it was in the farthest 2 lane turning lane and it started to go as the straight lane lights turned green
-goes too fast in a neighborhood

This is not ready and will obviously need v5 hardware....otherwise I wouldn't get in a robotaxi.

Not smooth at all and I will not be paying the monthly fee.

This is all after I was sitting there listening to Elon talk about interventions yesterday at the meeting...

This software does not make me feel comfortable.

 

[Fred42]

On PV & Roof Panels, and Waymo:

• Let’s wrap up discussion of rooftop solar, hurricanes and energy incentives here soon, please. I am being lenient because I’m in the midst of augmenting my own system this very moment (just crawled down from the roof trying to figure out how I’m going to solve the next of a never-ending set of issues), so right now I’ve more than the normal amount of Mod-flexibility on that topic.

• And we’ve had Waymo discussion of other FSD-like discussion than this thread should be subject to. I really hope to see Nomo.

I have gathered from prior posts that there are others who qualify better than you as a spring chicken. I'm wondering if it's a good idea to work on the roof and climb or crawl up and down. Could you have contracted with someone to do this work?

If you don't go up a ladder, then you won't fall off a ladder.

 

[EQC_]

Because Elon bet the house on 4680 and lost.

This is incorrect. It seems you are ignoring or misunderstanding the context and intent of Tesla's plans and progress for 4680.

I suggest it is more accurate to say: Elon hedged with 4680's, and the ramp up of the 4680's was slower than expected.

But, last time I tried to explain, and asked what other strategy would have been better, you replied with:

The smart strategy is to go with what you know until a replacement is ready. ...

Which is also incorrect, because Tesla didn't and doesn't have any other in-house battery manufacturing, and they can't just force suppliers to ramp up production faster -- but, Elon has frequently begged suppliers to do just that.

So, because Tesla knew it needed more batteries, and it couldn't force suppliers, Tesla Hedged with the 4680 plan.

Hedged. As in: provided a fall-back plan...because Elon and Tesla could not be 100% reliant on suppliers to ramp batteries fast enough. 4680 would simultaneously give Tesla an additional battery supply line, however small, that they had control over. And, it would also provide some guidance and motivation for their suppliers to ramp up and keep costs down if needed.

There are several key facts that a statement like "bet the house on 4680 and lost" just completely ignores:

1. Elon has always publicly begged suppliers to ramp up battery production. Even on Battery Day, while talking about 4680, Elon made it clear that even when 4680 was fully ramped to huge numbers at some point in the future, Tesla would still buy any and all qualified cells from other manufacturers.
   
   1. So, nothing about 4680 should have slowed down the other suppliers in any way. In other words: 4680, even while being slowly ramped, has provided Tesla with more US-made cells than they would have had otherwise.
      1. I don't see a negative there...
   2. We also know that other manufacturers now have 4680 plans with factories in various stages of planning and building. I believe both Panasonic and LG are working toward 4680 cells. So...when those are ready, that will be even more cells for Tesla, and the extra nice thing is that those cells are in a format that Tesla decided was ideal for their products. Even better.
      1. I don't see a negative there either...
2. Tesla proved they would buy "any and all qualified cells" from suppliers, by purchasing and continuing to purchase big quantities of diverse supplies of cells, and doing the testing and engineering to make it work.
   
   1. I believe Tesla buys cells from suppliers including Panasonic, Samsung, LG, CATL, and BYD.
   2. Various form factors.
   3. Various chemistries.
   4. Clearly, Tesla is using all options...and 4680 is an ADDITIONAL option. Extra options aren't bad.
3. Tesla had no other in-house battery manufacturing that they could have ramped up on their own.
   
   1. If "go with what you know" was supposed to mean "Tesla should have just ramped up the 2170's they already used," that is impossible. Tesla didn't and doesn't make the 2170's. Panasonic makes them. Tesla doesn't hold the IP or patents or manufacturing knowhow on the 2170's they buy, and we have heard that the Panasonic side of the Gigafactory where the 2170's are made is kept relatively secret from Tesla employees.
   2. Any in-house Tesla batteries would have been a development process (just like the 4680's). And, logically, any Tesla-made batteries following a form factor they already buy from a supplier would have probably required navigating all sorts of extra patents and IP, and potential disgruntlement from that supplier.
4. Elon and Tesla can't just force their suppliers to ramp up or duplicate existing battery production lines faster. Period.
   
   1. Panasonic has been slow to ramp 2170's at GigaNevada, but they do ramp. Gradually, they either add production lines or improve existing lines to increase the production rates and/or the GWh output.
   2. But, Tesla can't just make a supplier invest faster. That's not how business works. Every company has its own philosophy on investment and risk evaluation...

Now, even though the ramp has been slow, we have plenty of clues that it is speeding up now. While Elon stated a new record of 1300 Cybertrucks per week, even assuming an average of 1000 Cybertrucks per week is already over 6 GWh per year of 4680's. Tesla is looking to target 2500 Cybertrucks per week by the end of the year...so potentially a rate of 12-15 GWh per year of 4680's. And, of course, we know that Tesla is ramping existing production lines while also building more lines.

When viewing those numbers, keep in mind that Tesla's only other supplier of US-made cells is Panasonic...and I believe all of Giga Nevada is roughly 42 GWh per year.

So, Tesla's "lost bet" on 4680's has already provided them with 15% more US-made cells than they would otherwise have. And, by the end of the year, that might be more like 25-30% more US-made cells...with plans to keep ramping up from there.

 

[SPadival]

It really seems crazy to me that they still don't make the model 3 in Berlin. Its the smallest car in the lineup and NOT made in Europe?

Because demand for SUV/CUV > Sedan/small car (unless you’re competing with the Dacia Sandero )

 

[dl003]

The 2170 and 4680 form factor. The materials have gotten cheaper but moving the capacity to 2170 and 4680 will take time.

Disappointed no one asked about the Model 3 tax credit. We need to make sure this question is part of the quarterly call in July. Very disappointed if they are not urgently working on a battery pack/cells for the US model 3 that will get the credit for both RWD and LR both for the credit and for geopolitical reasons.

The 3 is using 2170. The answer is currently Panasonic isnt making enough 2170 to supply the 3 production in the US.

4680 is going into Semi and CT and they are trying their best but currently no excess capacity for much else.

So Im gonna ask again: for everyone who is disappointed with a so-called *intentional delay* in the 25k car: how do you suppose Tesla will get the batteries for it?

 

[The Accountant]

I liked your post for the first part, but I just got my car upgraded to HW3 and have had FSD for a day and I'm not impressed. I've watched dozens of vids and expected it to be like those, but I've had to disengage way too many times:

-it didn't follow the new straight and turn lanes in the new construction area
-it doesn't stay in the middle of the lane, but instead rides the right hand part almost on the line (autopilot has better handling)
-it was in the farthest 2 lane turning lane and it started to go as the straight lane lights turned green
-goes too fast in a neighborhood

This is not ready and will obviously need v5 hardware....otherwise I wouldn't get in a robotaxi.

Not smooth at all and I will not be paying the monthly fee.

This is all after I was sitting there listening to Elon talk about interventions yesterday at the meeting...

This software does not make me feel comfortable.

I am surprised when I see FSD comments like yours. I drive FSD in Florida and in NY/CT.
In Florida I have not had one disengagement or intervention. I believe I have about 2,000 miles there with 90% on city streets. In my Florida city, roads are well marked and signage is good. Most area streets are grid-like .. . .very little construction.
In NY/CT I occasionally get interventions or disengagement but they are usually due to some odd edge case. The old neighborhood roads of New England can be very unusual at times and traffic patterns can be difficult for a human to interpret. But even here, I only get an intervention/disengagement 1 out of 10 trips (with v12.3). I have probably logged 3,000 miles on FSD v12 in NY/CT including Manhattan.

I realized that when I first started driving with FSD, I disengaged often because the car did not drive the way I would drive.
My wife would say, "why did you disengage? . . I would give her some complaint and she would tell me the car was doing fine.
As an Accountant, I can be a little controlling
I learned to let it do it's thing. It drives a bit closer to the right hand side of the road than I would, but it's fine (it has 8 eyes I tell myself).

On the speed issue, are you using the setting "Automatic Set Speed Offset"? I am very impressed with this setting. The car seems to know when to go below the speed limit in my town near school zones for example. Goes slightly above the speed limit where I would.

Keep using FSD and report back - I'm interested to see if you still find it lacking after a few more miles.

btw: one thing I can't seem to get to work, is having FSD enter my driveway. It always "arrives at my destination" in front of my house. I've seen videos of FSD taking the car right into the driveway but so far unsuccessful for me.

 

[SPadival]

I liked your post for the first part, but I just got my car upgraded to HW3 and have had FSD for a day and I'm not impressed. I've watched dozens of vids and expected it to be like those, but I've had to disengage way too many times:

-it didn't follow the new straight and turn lanes in the new construction area
-it doesn't stay in the middle of the lane, but instead rides the right hand part almost on the line (autopilot has better handling)
-it was in the farthest 2 lane turning lane and it started to go as the straight lane lights turned green
-goes too fast in a neighborhood

This is not ready and will obviously need v5 hardware....otherwise I wouldn't get in a robotaxi.

Not smooth at all and I will not be paying the monthly fee.

This is all after I was sitting there listening to Elon talk about interventions yesterday at the meeting...

This software does not make me feel comfortable.

Elon said he’ll add Boomer/Millenial mode to FSD soon (2 weeks)

https://twitter.com/x/status/1796715946918228360

 

[EQC_]

Elon said the market is awash with batteries. More variants would have avoided this slump. Not everybody wants to drive the same car. Isn't Elon being paid 1,000x other auto execs to figure out batteries?

No.

If Elon gets paid*, it will be to cover the production and market-cap milestones that were described in the 2018 compensation package, and achieved through Elon and the Tesla team's efforts.

Currently, Elon doesn't have a pay package for his ongoing work. Any future compensation packages have not been agreed to yet, as far as we know.

As far as the 2018 plan, there was and is a written out plan, explained rather clearly in Tesla's blog on the subject from January of 2018:

Tesla Investor Relations

PALO ALTO, Calif., Jan. 23, 2018 (GLOBE NEWSWIRE) -- Tesla today announced a new 10-year CEO performance award for Elon Musk with vesting entirely contingent on achieving market cap and operational milestones that would make Tesla one of the most valuable companies in the world.

ir.tesla.com

Put simply, Elon would earn stock options, in tranches of 1% of outstanding shares, each each time he pushed Tesla to simultaneously achieve an additional "Operational Milestone" from the list and an additional "Market Capitalization" milestone. The market cap milestones required Tesla to increase from the then current $50 Billion, through 12 $50-billion steps, up to $650 billion. You can also see that the Operational milestones included similarly huge increases in Revenue and/or EBITDA.

For reference, Tesla has indeed grown tremendously since then, even ignoring those specific metrics. The deal was made public in January of 2018, and just before that, Tesla announced the Q4 2017 and full-year 2017 production figures:

Tesla Investor Relations

PALO ALTO, Calif., Jan. 03, 2018 (GLOBE NEWSWIRE) -- In Q4, Tesla delivered 29,870 vehicles, of which 15,200 were Model S, 13,120 were Model X, and 1,550 were Model 3.

ir.tesla.com

• In Q4 of 2017, Tesla production totaled 24,565 vehicles, of which 2,425 were Model 3
  • At the time this deal was written, the Model 3 was at the very early stages of the ramp...deep in Production Hell.
  • Today, Tesla's Cybertruck ramp is currently well ahead of where the Model 3 ramp was when this deal was written. Roughly 1000 per WEEK, versus the Model 3 at 2500 per quarter (or roughly 200 per week).
  • And Tesla's Model Y numbers today are roughly 25,000 per week, or . So, Tesla produces more Y's in a week today than they produced all vehicles in an entire quarter of Q42017.
• In Q4 of 2017, Tesla delivered 29,870 vehicles, of which 15,200 were Model S, 13,120 were Model X, and 1,550 were Model 3.
• For all of 2017, production and deliveries were roughly 100,000 vehicles total.
• Tesla Energy was barely a spark back then.
• GigaShanghai didn't even break ground until January 2019...a year after this deal.
• Berlin...Austin...etc. etc.
• Critics of Elon and Tesla today really only draw attention to Elon's "antics" and the "ugly" truck. That's also a far lot better than the constant predictions of bankruptcy and back in the 2017/2018 timeframe.
  • Ha ha ha...Elon's "bankwupt" tweet didn't even come out until April 1, 2018...months after this compensation package went public.

So, for all the complaints and headlines about Elon's supposedly terrible management, bad decisions, and public antics...comparing Tesla today to the Tesla at the time this deal was written, I feel like the achievement is enormous and the success is huge.



*The word "paid" here is used loosely, and means "allowed to exercise the agreed upon stock options"...which I believe technically involves Elon paying Tesla, and receiving discounted shares in return...with the stipuation that he must also hold those shares for years. And, as is clearly stated in the blog: "If none of the 12 tranches is achieved, Elon will not receive any compensation"...or perhaps ONLY whatever minimum wage the company was legally required to pay him.

In contrast, while the other auto executives have a much smaller pay package on paper, they face zero risk, and are guaranteed their millions per year, plus likely bonuses, plus a likely golden parachute of more millions even if they fail miserably. Literally enough money for a family to live off of for a lifetime, even if they fail miserably and are pushed to walk away from the company. I'm not weeping for them...and "guaranteed millions of dollars" isn't really a reasonable comparison to "earning a 12% stake in the company if and only if you can drive that company to perform 10x better and be worth 10x more."

 

[Linkmeup]

I liked your post for the first part, but I just got my car upgraded to HW3 and have had FSD for a day and I'm not impressed. I've watched dozens of vids and expected it to be like those, but I've had to disengage way too many times:

-it didn't follow the new straight and turn lanes in the new construction area
-it doesn't stay in the middle of the lane, but instead rides the right hand part almost on the line (autopilot has better handling)
-it was in the farthest 2 lane turning lane and it started to go as the straight lane lights turned green
-goes too fast in a neighborhood

This is not ready and will obviously need v5 hardware....otherwise I wouldn't get in a robotaxi.

Not smooth at all and I will not be paying the monthly fee.

This is all after I was sitting there listening to Elon talk about interventions yesterday at the meeting...

This software does not make me feel comfortable.

my posts carry weight on this investor forum?!

 

[Webeevdrivers]

I am surprised when I see FSD comments like yours. I drive FSD in Florida and in NY/CT.
In Florida I have not had one disengagement or intervention. I believe I have about 2,000 miles there with 90% on city streets. In my Florida city, roads are well marked and signage is good. Most area streets are grid-like .. . .very little construction.
In NY/CT I occasionally get interventions or disengagement but they are usually due to some odd edge case. The old neighborhood roads of New England can be very unusual at times and traffic patterns can be difficult for a human to interpret. But even here, I only get an intervention/disengagement 1 out of 10 trips (with v12.3). I have probably logged 3,000 miles on FSD v12 in NY/CT including Manhattan.

I realized that when I first started driving with FSD, I disengaged often because the car did not drive the way I would drive.
My wife would say, "why did you disengage? . . I would give her some complaint and she would tell me the car was doing fine.
As an Accountant, I can be a little controlling
I learned to let it do it's thing. It drives a bit closer to the right hand side of the road than I would, but it's fine (it has 8 eyes I tell myself).

On the speed issue, are you using the setting "Automatic Set Speed Offset"? I am very impressed with this setting. The car seems to know when to go below the speed limit in my town near school zones for example. Goes slightly above the speed limit where I would.

Keep using FSD and report back - I'm interested to see if you still find it lacking after a few more miles.

btw: one thing I can't seem to get to work, is having FSD enter my driveway. It always "arrives at my destination" in front of my house. I've seen videos of FSD taking the car right into the driveway but so far unsuccessful for me.

Definitely not the case for us. At least when driving in Canada. When 12.3.6 came out the car was immediately capable of doing 3 and probably 5 more intersections that it got wrong before. Wrong lane, freezing up mid intersection etc. It was great and a huge improvement. But over a few weeks it dropped every one of them and now gets them anll wrong every time.

As well, any flashing advanced green signal for turning left it gets wrong. A robotaxi ride in Canada would get you to the first intersection with a flashing advanced green and then you are walking.

12.3.6 is still better than previous versions but the needle has moved very little. FSD is years away in my opinion. We are not even close.

 

[willow_hiller]

But over a few weeks it dropped every one of them and now gets them anll wrong every time.

Between you and @ggies07 it sounds like both of your cars need a camera recalibration. You can do it yourself in the service menu. FSD won't be available while it recalibrates, and it can take a couple dozen miles or more, but people with similar problems have reported it made a massive improvement.

 

[ggies07]

Between you and @ggies07 it sounds like both of your cars need a camera recalibration. You can do it yourself in the service menu. FSD won't be available while it recalibrates, and it can take a couple dozen miles or more, but people with similar problems have reported it made a massive improvement.

Thank you for this, very helpful. I literally watched it inch over to the right side of the line and then I started hearing bumps from driving over the lane reflectors, as it was that far in the middle of the lanes...I immediately turned it off.

Oh, also everyone, I'm 39, not a boomer!

 

[Skryll]

Definitely not the case for us. At least when driving in Canada. When 12.3.6 came out the car was immediately capable of doing 3 and probably 5 more intersections that it got wrong before. Wrong lane, freezing up mid intersection etc. It was great and a huge improvement. But over a few weeks it dropped every one of them and now gets them anll wrong every time.

As well, any flashing advanced green signal for turning left it gets wrong. A robotaxi ride in Canada would get you to the first intersection with a flashing advanced green and then you are walking.

12.3.6 is still better than previous versions but the needle has moved very little. FSD is years away in my opinion. We are not even close.

It's comical that when it enounters a flashing red it does stop go stop go - in California. Also one part of the freeway it always exists by accident right after entering because it fails to prioritize getting out of that lane. Sometimes hesitates when selecting lanes.

Then at other times it rock solid through narrow streets with random activity or goes town to town door to door with no intervention.

 

[GatorMeat]

That doesn't answer my question. If battery cells are readily available in the US to the extent they can support that 2-door car, why are we using Chinese batteries in the 3?

nobody said readily available 'in the US'

 

[GatorMeat]

Why don't you just put them back up once the new roof is on?

sure. but the installer said it would be another $5k at today's price and they charge $2k to travel (only 120 miles away)

 

[GatorMeat]

No.

If Elon gets paid*, it will be to cover the production and market-cap milestones that were described in the 2018 compensation package, and achieved through Elon and the Tesla team's efforts.

Currently, Elon doesn't have a pay package for his ongoing work. Any future compensation packages have not been agreed to yet, as far as we know.

As far as the 2018 plan, there was and is a written out plan, explained rather clearly in Tesla's blog on the subject from January of 2018:

Tesla Investor Relations

PALO ALTO, Calif., Jan. 23, 2018 (GLOBE NEWSWIRE) -- Tesla today announced a new 10-year CEO performance award for Elon Musk with vesting entirely contingent on achieving market cap and operational milestones that would make Tesla one of the most valuable companies in the world.

ir.tesla.com

Put simply, Elon would earn stock options, in tranches of 1% of outstanding shares, each each time he pushed Tesla to simultaneously achieve an additional "Operational Milestone" from the list and an additional "Market Capitalization" milestone. The market cap milestones required Tesla to increase from the then current $50 Billion, through 12 $50-billion steps, up to $650 billion. You can also see that the Operational milestones included similarly huge increases in Revenue and/or EBITDA.

For reference, Tesla has indeed grown tremendously since then, even ignoring those specific metrics. The deal was made public in January of 2018, and just before that, Tesla announced the Q4 2017 and full-year 2017 production figures:

Tesla Investor Relations

PALO ALTO, Calif., Jan. 03, 2018 (GLOBE NEWSWIRE) -- In Q4, Tesla delivered 29,870 vehicles, of which 15,200 were Model S, 13,120 were Model X, and 1,550 were Model 3.

ir.tesla.com

• In Q4 of 2017, Tesla production totaled 24,565 vehicles, of which 2,425 were Model 3
  • At the time this deal was written, the Model 3 was at the very early stages of the ramp...deep in Production Hell.
  • Today, Tesla's Cybertruck ramp is currently well ahead of where the Model 3 ramp was when this deal was written. Roughly 1000 per WEEK, versus the Model 3 at 2500 per quarter (or roughly 200 per week).
  • And Tesla's Model Y numbers today are roughly 25,000 per week, or . So, Tesla produces more Y's in a week today than they produced all vehicles in an entire quarter of Q42017.
• In Q4 of 2017, Tesla delivered 29,870 vehicles, of which 15,200 were Model S, 13,120 were Model X, and 1,550 were Model 3.
• For all of 2017, production and deliveries were roughly 100,000 vehicles total.
• Tesla Energy was barely a spark back then.
• GigaShanghai didn't even break ground until January 2019...a year after this deal.
• Berlin...Austin...etc. etc.
• Critics of Elon and Tesla today really only draw attention to Elon's "antics" and the "ugly" truck. That's also a far lot better than the constant predictions of bankruptcy and back in the 2017/2018 timeframe.
  • Ha ha ha...Elon's "bankwupt" tweet didn't even come out until April 1, 2018...months after this compensation package went public.

So, for all the complaints and headlines about Elon's supposedly terrible management, bad decisions, and public antics...comparing Tesla today to the Tesla at the time this deal was written, I feel like the achievement is enormous and the success is huge.



*The word "paid" here is used loosely, and means "allowed to exercise the agreed upon stock options"...which I believe technically involves Elon paying Tesla, and receiving discounted shares in return...with the stipuation that he must also hold those shares for years. And, as is clearly stated in the blog: "If none of the 12 tranches is achieved, Elon will not receive any compensation"...or perhaps ONLY whatever minimum wage the company was legally required to pay him.

In contrast, while the other auto executives have a much smaller pay package on paper, they face zero risk, and are guaranteed their millions per year, plus likely bonuses, plus a likely golden parachute of more millions even if they fail miserably. Literally enough money for a family to live off of for a lifetime, even if they fail miserably and are pushed to walk away from the company. I'm not weeping for them...and "guaranteed millions of dollars" isn't really a reasonable comparison to "earning a 12% stake in the company if and only if you can drive that company to perform 10x better and be worth 10x more."

Elon's contract runs 2018 through 2028.

 

[GatorMeat]

This is incorrect. It seems you are ignoring or misunderstanding the context and intent of Tesla's plans and progress for 4680.

I suggest it is more accurate to say: Elon hedged with 4680's, and the ramp up of the 4680's was slower than expected.

But, last time I tried to explain, and asked what other strategy would have been better, you replied with:


Which is also incorrect, because Tesla didn't and doesn't have any other in-house battery manufacturing, and they can't just force suppliers to ramp up production faster -- but, Elon has frequently begged suppliers to do just that.

So, because Tesla knew it needed more batteries, and it couldn't force suppliers, Tesla Hedged with the 4680 plan.

Hedged. As in: provided a fall-back plan...because Elon and Tesla could not be 100% reliant on suppliers to ramp batteries fast enough. 4680 would simultaneously give Tesla an additional battery supply line, however small, that they had control over. And, it would also provide some guidance and motivation for their suppliers to ramp up and keep costs down if needed.

There are several key facts that a statement like "bet the house on 4680 and lost" just completely ignores:

1. Elon has always publicly begged suppliers to ramp up battery production. Even on Battery Day, while talking about 4680, Elon made it clear that even when 4680 was fully ramped to huge numbers at some point in the future, Tesla would still buy any and all qualified cells from other manufacturers.
   
   1. So, nothing about 4680 should have slowed down the other suppliers in any way. In other words: 4680, even while being slowly ramped, has provided Tesla with more US-made cells than they would have had otherwise.
      1. I don't see a negative there...
   2. We also know that other manufacturers now have 4680 plans with factories in various stages of planning and building. I believe both Panasonic and LG are working toward 4680 cells. So...when those are ready, that will be even more cells for Tesla, and the extra nice thing is that those cells are in a format that Tesla decided was ideal for their products. Even better.
      1. I don't see a negative there either...
2. Tesla proved they would buy "any and all qualified cells" from suppliers, by purchasing and continuing to purchase big quantities of diverse supplies of cells, and doing the testing and engineering to make it work.
   
   1. I believe Tesla buys cells from suppliers including Panasonic, Samsung, LG, CATL, and BYD.
   2. Various form factors.
   3. Various chemistries.
   4. Clearly, Tesla is using all options...and 4680 is an ADDITIONAL option. Extra options aren't bad.
3. Tesla had no other in-house battery manufacturing that they could have ramped up on their own.
   
   1. If "go with what you know" was supposed to mean "Tesla should have just ramped up the 2170's they already used," that is impossible. Tesla didn't and doesn't make the 2170's. Panasonic makes them. Tesla doesn't hold the IP or patents or manufacturing knowhow on the 2170's they buy, and we have heard that the Panasonic side of the Gigafactory where the 2170's are made is kept relatively secret from Tesla employees.
   2. Any in-house Tesla batteries would have been a development process (just like the 4680's). And, logically, any Tesla-made batteries following a form factor they already buy from a supplier would have probably required navigating all sorts of extra patents and IP, and potential disgruntlement from that supplier.
4. Elon and Tesla can't just force their suppliers to ramp up or duplicate existing battery production lines faster. Period.
   
   1. Panasonic has been slow to ramp 2170's at GigaNevada, but they do ramp. Gradually, they either add production lines or improve existing lines to increase the production rates and/or the GWh output.
   2. But, Tesla can't just make a supplier invest faster. That's not how business works. Every company has its own philosophy on investment and risk evaluation...

Now, even though the ramp has been slow, we have plenty of clues that it is speeding up now. While Elon stated a new record of 1300 Cybertrucks per week, even assuming an average of 1000 Cybertrucks per week is already over 6 GWh per year of 4680's. Tesla is looking to target 2500 Cybertrucks per week by the end of the year...so potentially a rate of 12-15 GWh per year of 4680's. And, of course, we know that Tesla is ramping existing production lines while also building more lines.

When viewing those numbers, keep in mind that Tesla's only other supplier of US-made cells is Panasonic...and I believe all of Giga Nevada is roughly 42 GWh per year.

So, Tesla's "lost bet" on 4680's has already provided them with 15% more US-made cells than they would otherwise have. And, by the end of the year, that might be more like 25-30% more US-made cells...with plans to keep ramping up from there.

the slow ramp of 4680 is why 50% growth rate died

 

[Usain]

Assuming you have a network that can already talk to every car out there and can already send some commands, how much more is needed to talk to the RoboTaxi?

You need to write the software for fleet monitoring and for remote control when a robotaxi gets stuck. Judging from Elon's answer, they haven't done that yet. It looks like they haven't designed the back end fleet management system yet.

 

[AudubonB]

I have gathered from prior posts that there are others who qualify better than you as a spring chicken. I'm wondering if it's a good idea to work on the roof and climb or crawl up and down. Could you have contracted with someone to do this work?

If you don't go up a ladder, then you won't fall off a ladder.

Thank you for your thoughts. I guess you don’t know or remember how isolated we are? Nearest village 80 miles over the Alaska Range; nearest barely-qualified riggers 200 miles away.
I DO wear the very finest safety harness.

Usually.

Sometimes, any way. Especially today when I also had to jury-rig an extension platform……

And (not joking)I never do roof-work in the rain: all-metal Rooooooooooooooffssss….

 

[keydiver]

We also have to have an extra $1M policy if over 10kW

That's only if you GENERATE over 10kW. You can actually have up to almost 12 kW of panels. And, I have been told that you only need that stupid $1 million liability policy for the installation, and you can discontinue it later. FPL is the only one I know of that has that dumb limitation of anything over 10kW is considered commercial generation. But, how do they know how much you are actually making? They don't.