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.
Biggest problem with it is inconsistency. One second I am astounded by its brilliance, the next I’m frantically disengaging to save me/car/other from an imbecilic move.

Elon recently explained why this sometimes happens: It is because the various neural set up something of a "race condition" for the inputs to the path planning code. That is, if one necessary input to that process is late or missing, it can lead to an inappropriate decision in real time driving.

Elon said that merging all neural nets into a "unified vector space" is the answer that fixes the issue. I hope that this is the plan for FSD beta v10.13 in the near term. It's almost certainly required for "single stack", where the FSD engine will be used for basic autopilot functions in all FSD computer equipped Tesla cars (post Apr 12, 2019 for Model 3).

Elon Musk on Twitter: "@WholeMarsBlog Next release (10.12) is another step towards all NN’s using surround video & reconciling output to a unified vector space for control code. Improves complex intersections in heavy traffic. Many upgrades to core code, so taking longer to debug issues. Probably Wed/Thurs release." / Twitter (May 6, 2022)

Cheers!
 
Yesterday, despite net purchase of 1.075 million shares of Tesla by mutual funds, $TSLA still dropped by $36.77 so they can get more shares.

My guess is we might see a further drop in $TSLA today and tomorrow (with help from MMs) as they wrap up their end of quarter rebalancing with more buys in $TSLA. By Friday we should close near the Friday max pain.

Sources:
 

Attachments

  • Book1.txt
    4.6 KB · Views: 43
Last edited:
Elon recently explained why this sometimes happens: It is because the various neural set up something of a "race condition" for the inputs to the path planning code. That is, if one necessary input to that process is late or missing, it can lead to an inappropriate decision in real time driving.

Elon said that merging all neural nets into a "unified vector space" is the answer that fixes the issue. I hope that this is the plan for FSD beta v10.13 in the near term. It's almost certainly required for "single stack", where the FSD engine will be used for basic autopilot functions in all FSD computer equipped Tesla cars (post Apr 12, 2019 for Model 3).

Elon Musk on Twitter: "@WholeMarsBlog Next release (10.12) is another step towards all NN’s using surround video & reconciling output to a unified vector space for control code. Improves complex intersections in heavy traffic. Many upgrades to core code, so taking longer to debug issues. Probably Wed/Thurs release." / Twitter (May 6, 2022)

Cheers!

Anyone have an updated timeline for when FSD Beta 10.13 drops? Last I saw was '2 weeks' from Elon on June 14th, so if that timeline is still correct, it should be dropping very soon:

 
What if 4680's and single structural pack were super ambitious (but unnecessary) step for MY, but absolutely necessary for CT, Semi?
Could GF Austin - revert to just 2170's for MY and scale MY that way? From the get go, reverting to 2170s was the backup plan it seems.
With demand and pricing power, (& inflation) there seems no longer be a need to try to bring down price of MY?

the failure in production output for 4680 to scale for MY, might still be a great step in starting the CT, Semi programs early?

Whenever Elon or Drew mentions the difficulty in scaling battery production they always seem to be focused on raw materials.

For 4680s, Panasonic, LG and Samsung are also making cells.

For Model 3/Y Tesla can use 2170s and LFP packs from CATL and possibly BYD.

We don't know what type of pack the CT will have, but IMO it is very unlikely that the Semi will have a structural pack.

Whatever issues Tesla may or may not have with 4860 production and structural packs, they don't seem to be show stoppers. Some 4680s and structural packs are being made.

It is likely that the issues are yield, linespeed, reliability, problems that can be overcome with some time and effort.

Even assuming the worst case that the problems linger on beyond Q1 2023, there are other sources for cells and other ways of making battery packs.

If the problem is yeild and machinery reliability, Tesla has solved similar problems in the past.
 
This recession will be gone as soon as all the ports are working, China stop randomly shut down, chips are of abundance, and the entire world(minus Ukraine) goes back to full production. GDP is a measure of goods produced and services for a country. Key word here is GOODS PRODUCED. Supply chain disruption is the culprit for a decrease in goods produced and high inflation. Once this is restored, GDP will shoot up.

I don't know why people are freaking out as if the supply chain will never be restored back to pre disruptive levels. If this is your belief then inflation is not transitory and we are in the beginning of a major recession. I'm on the other side believing this will all blow over in a few months(and things already massively improved). Third quarter GDP will correlate what Tesla Q3 production will be. If you think it's flat or negative compared to Q2, then recession continues.
I agree, people are being idiots but you are also discounting the impacts of the Ukraine/Russia conflict which could continue to result in disruptions in food and natural resources for years to come. Will China change the Covid policy? Declare victory and stop?
 
I agree, people are being idiots but you are also discounting the impacts of the Ukraine/Russia conflict which could continue to result in disruptions in food and natural resources for years to come. Will China change the Covid policy? Declare victory and stop?

The war in Ukraine has made Fossil Fuels very expensive.

That is driving short term inflation, but it is also driving investment in alternatives, in the long run the alternatives are deflationary.

We also need to question where some assets and investments are already highly priced, not Tesla, but there are plenty of others I would question.

The way we make Ammonia from Methane currently is all wrong very expensive and unsustainable. A lot of agriculture is based on fertilizers using Ammonia.

Monash university in Australia has a process for making Ammonia which might become competitive.

But if fertilizer prices are high for a few years, that will feed into food prices, compounding the war in Ukraine.

In the long run my hunch is that the new Ammonia processes will be cheaper than using methane.
 
Last edited:
I agree, people are being idiots but you are also discounting the impacts of the Ukraine/Russia conflict which could continue to result in disruptions in food and natural resources for years to come. Will China change the Covid policy? Declare victory and stop?
If China continues current covid-policy, lock downs will continue recurrently. Covid-19 is not going to disappear.
 

Hopefully she won't mind - saving you all a few clicks - and saving /archiving just in case

... <snip> (Please do click the URL link to her Substack if you can, fun side details about that interview )

...
Elon opened the fudge I'd brought him from Bucc-ees, and a few horns honked in the background as we were in a factory. So I apologize in advance for those minor glitches. Note to self, air, and paper can be very loud.

What We Talked About

I had a lot of questions and although we didn’t get through them all, he invited me back to finish asking him at another time so there will eventually be a part 2. Here’s what we talked about in this episode:

What the light of consciousness itself meant to him. He always talks about it but what is it to him?

His thoughts on what would happen to the light of consciousness if/when humanity goes extinct. What he's looking for in the first few people who go to Mars. The role Starlink has played and will continue to play in disaster relief. The Musk Foundation and its role in disaster relief, XPRIZE, and carbon capture (here I brought up Project Vesta, peridot, and other companies using lab-grown diamonds as a form of carbon capture.)

We also talked about his favorite accomplishment of the Musk Foundation so far although he emphasized that there is a lot of work to be done and that it's hard to spend the money in a way that is well spent--that has the highest impact.

Other topics we spoke about included the declining birthrate, poverty as a cause (we differed in opinion on that), and one key solution to solving poverty and I agreed with him on that. That solution is literacy. I am very fortunate that my mother taught me how to read at a young age. The library was my safe place as a child and I love to read. However, I know people who don't know how to read or write and they are in my age group.

The education system has failed so many here. I remember being in a class where impatient teachers would force students to struggle to read out loud. And she would impatiently correct them if they couldn't read the word fast enough. Many of my classmates hated books and reading. I hated math and sciences--we hate what we don't understand and find confusing especially if we are being taught these subjects by someone who isn't passionate about their job.

I digress. Elon also shared some updates for Tesla’s legal and justice department (good news for Tesla) and we spoke of Tesla Insurance, my hope that it will come to Louisiana soon since we have the highest average cost of insurance in the nation, mining, Tesla Energy's impact with ERCOT and the Texas grid (I got tongue-tied. Texla.) And we briefly touched upon AI and general intelligence.

Elon emphasized that he has no desire for Tesla to go into the mining industry. He made that super clear to me and I just wanted to emphasize that for anyone curious or even hopeful.

Meeting Him and the Adventures I Experienced Along the Journey​

Boy, was this trip an adventure? Already my flight home is delayed by a day. But at least, for now, I am not stranded in Dallas again. Fortunately, I used to live in Dallas. I was married there and this was my first time back since the hell I went thru (the hell that eventually let me to Elon, oddly) and it was surreal. I have friends and family there and one of them picked me up from the airport and drove me to Austin.

And since this was Elon’s birthday weekend, I wanted to get him something nice. I had already gotten him some Cafe du Monde from the Baton Rouge Airport (they have a cute little Louisiana grocery section in their store). I picked up some fudge from Bucc-ees along the way and made a stop at Nature’s Treasures.

I got him a couple of nice things from there and one of the clerks also bought him a small gift (a small pyrite specimen) and gave me a discount when I told them who I was interviewing. So, shout out to Nature’s Treasures in Austin for the love. Gail’s daughter designed a cute tee shirt and they also gave him some goodies.

Meeting Elon Musk and experiencing his kindness in person made the flight adventures worth it. Despite the audio glitches, I hope you all enjoy the podcast.

And thank you, Elon Musk, for joining me. For part two, I’m going to ask you about your favorite mineral, pyrite—among other things.

The interview drops July 1, 2022.​

The interview will be available on my website, www.gettingstoned.online, and will also be available on Spotify and Apple Podcasts. Here’s a quick preview with Elon talking about Tesla’s Master Plan Part 3.


UPDATE: something big (?) will be announced by Johnna Crider after July 1st - and it isn't in the interview

Someone please recap all events likely to possibly shock next week, starting Tuesday July 5th - looks like the near future could be chaotic - or not

Or maybe Elon / Tesla's Twitter team is just hot testing Twitter's algos - so a Twitter induced TSLA shock?

The EMSK Elon show goes on 24x7

TSLA.Crider.July.1.jpg


And the beat show goes on

 
Last edited:
Whole heartedly disagree. Love the interiors. Tesla has done an admirable job of improving them. The difference between our ‘17 S&X and our ‘22 S&Y is astounding. Sparse is minimalist. And that is elegance. Making things simple is hard.
Don't get me wrong, I'm with you. I own a 'luxury' hybrid lexus, and a model S. I much prefer the model S, despite the fact that the materials used are cheaper. I also ordered a model Y, which is even more spartan. Its a great look IMHO. My comment was more about market perception.
I'm a tech guy, so I want simplicity and functionality. But I know a LOT of people my age or older, in the market for new cars, who are turned off by what they perceive as a 'cheap' interior.

Decades of very expensive and wasteful marketing have persuaded people that a car with lots of leather with neat stitching and dozens of polished chrome switches and levers are what we should think of as 'premium'. That sort of marketing doesn't go away overnight.

I don't think tesla need to change their interior now at all, but when a certain scale is reached, and when demand starts to match supply, I'd be sympathetic to them introducing a higher priced trim level for the 3 and Y that caters more to the 'chrome and leather' market. Whether we like it or not, people ARE buying porsche taycans and audi e-trons, and it sure isn't for the charging network, or the performance or technology!
 
This recession will be gone as soon as all the ports are working, China stop randomly shut down, chips are of abundance, and the entire world(minus Ukraine) goes back to full production. GDP is a measure of goods produced and services for a country. Key word here is GOODS PRODUCED. Supply chain disruption is the culprit for a decrease in goods produced and high inflation. Once this is restored, GDP will shoot up.

I don't know why people are freaking out as if the supply chain will never be restored back to pre disruptive levels. If this is your belief then inflation is not transitory and we are in the beginning of a major recession. I'm on the other side believing this will all blow over in a few months(and things already massively improved). Third quarter GDP will correlate what Tesla Q3 production will be. If you think it's flat or negative compared to Q2, then recession continues.


There is debate about how quickly the supply chain issues will be resolved. Large companies are spending big money on related projects at major consultancies which suggests no easy or quick fix.
 
UPDATE: something big (?) will be announced by Johnna Crider after July 1st - and it isn't in the interview

Someone please recap all events likely to possibly shock next week, starting Tuesday July 5th - looks like the near future could be chaotic - or not

Or maybe Elon / Tesla's Twitter team is just hot testing Twitter's algos

View attachment 822578

She says "(not the Elon interview)"
versus
(not in the Elon interview)

May have nothing to do with Tesla/ Elon.
 
Tesla's IR dept compiles the Analyst Consensus for Q2 deliveries. There were 20 analysts that confirmed they adjusted for China COVID lockdown and their consensus is 256.7k vehicles. If you add in the 5 analysts who did not confirm that they adjusted for China Covid lockdown you get 25 analysts with consensus of 266.8k. As Gary points out, you can calculate that the 5 analysts have an average of 307.2k.

I feel certain that the media will go with the 266.8k expectation.

 
Whenever Elon or Drew mentions the difficulty in scaling battery production they always seem to be focused on raw materials.

For 4680s, Panasonic, LG and Samsung are also making cells.

For Model 3/Y Tesla can use 2170s and LFP packs from CATL and possibly BYD.

We don't know what type of pack the CT will have, but IMO it is very unlikely that the Semi will have a structural pack.

Whatever issues Tesla may or may not have with 4860 production and structural packs, they don't seem to be show stoppers. Some 4680s and structural packs are being made.

It is likely that the issues are yield, linespeed, reliability, problems that can be overcome with some time and effort.

Even assuming the worst case that the problems linger on beyond Q1 2023, there are other sources for cells and other ways of making battery packs.

If the problem is yeild and machinery reliability, Tesla has solved similar problems in the past.
I don't think it is a trivial change for CT and Semi to use 2170's. First of all Elon said the design is locked for the CT. If they need to change the calls that would be a significant re-design as well as a ton of retesting and certification. It would surely mean no CT by mid next year. I assume the same is for the Semi, though they will ship fewer of those - they might be able to get by with the reduced output there.

For the MY - yeah, they just don't ship any more MY SR until supply is ready.

The other suppliers are a long way off. Pano, LG and Samsung won't have sells until 2024 at the earliest.
 
New video from Tobi Lindh filmed yesterday - there are lots of trucks with Model Y (performance judging from the wheels) leaving the Giga Berlin factory, so whatever is stopping some deliveries does not seem to halt production at least:



So either they fixed the problem already, or it is fixable at the delivery site later once they receive new parts. If the latter is true, they must have visibility that the parts will come soon. So does not seem a major/longer lasting problem IMHO.
 
What if 4680's and single structural pack were super ambitious (but unnecessary) step for MY, but absolutely necessary for CT, Semi?
Could GF Austin - revert to just 2170's for MY and scale MY that way? From the get go, reverting to 2170s was the backup plan it seems.
With demand and pricing power, (& inflation) there seems no longer be a need to try to bring down price of MY?

the failure in production output for 4680 to scale for MY, might still be a great step in starting the CT, Semi programs early?
The issues with the 4680 might further delay CT and Semi but they will figure this out. I doubt they will go backwards on the MY as they really are doing the right thing by having options in Austin and going slow with the 4680 launch. 2170 was the backup plan to de-risk the launch of the MY in Austin. The risk being the car factory has more capacity/capability than the battery factory due to a slower than expected launch of the 4680.

They need the MY launch with 4680 to volume to then de-risk the launch of CT and Semi which will not have a backup plan to 2170. There will be plenty of other non-battery risks for CT and semi for which they need to plan.

Honestly seeing this play out is giving me more confidence they have thought this out and know the risks. You will never completely eliminate the risks but if you aware of them and make a solid plan which incrementally scales new technology you have the best chance at success.
 
Last edited:
Tesla's IR dept compiles the Analyst Consensus for Q2 deliveries. There were 20 analysts that confirmed they adjusted for China COVID lockdown and their consensus is 256.7k vehicles. If you add in the 5 analysts who did not confirm that they adjusted for China Covid lockdown you get 25 analysts with consensus of 266.8k. As Gary points out, you can calculate that the 5 analysts have an average of 307.2k.

I feel certain that the media will go with the 266.8k expectation.

Note that the Covid Shutdown section has higher yearly total numbers that the combined estimates.
Even 2022 was bumped up showing Tesla improvements countered the shutdown and enhance production going forward.
 
I agree, people are being idiots but you are also discounting the impacts of the Ukraine/Russia conflict which could continue to result in disruptions in food and natural resources for years to come. Will China change the Covid policy? Declare victory and stop?

If China doesn’t change its COVID policy there will almost certainly be more shutdowns next winter/spring.