Search results

Acc. to Joe T's best guestimate/speculation, DIE is an acronym meaning 'Dynamic Ion Exchange' Timestamp 09:00

ChatGPT is getting a lot of use and attention. For many it is an iPhone moment wrt what AI can do. I wonder how it will affect the way people think about what tasks are possible for an AI to do. The next step is generalizing from the text and picture domain to other tasks: So, if an AI can...

Very likely. Lot of buzz re. Tesla selling inventory vehicles across many countries/locations with great success. My WAG is delivery of 470K+ cars. Hopefully this will be a 1-2 punch: First, the better than expected amount of delivery. Then af lot of FUD re. Tesla 'giving' cars away for...

Fair point. Implicit in my WAG is the assumption that EM prefers to max out existing locations. My WAG is that the Austin alone can eventually do ~6M and Berlin and Shanghai ~4M each when fully ramped. These number may seem wild, but I assume that the 3D 'packing' of the GFs and the speed...

My WAG is 8.5M vehicles at $20K gross in 2027 => 170 B (cont. production optimization and improvement, cont. extreme demand and pricing power, cont. ramp of current GFs plus one extra) Assume 50 B for fixed costs and taxes - still 120 B left. Given such extreme growth a 100 forward multiple...

I think robotaxi adoption will vary widely based on mostly age. So yes, most people above 50 will take a 'wait-and-see' approach. My hunch is that the 20-50 group will be hard to predict - some will be early adopters, but most won't. All the TMC anecdotes of Tesla-loving kids going back many...

EM and Tesla should just go all-in and do the power move: A 1:100 split. Then, in 3-5 years, given massive growth, and investors waking up to Teslas dominance, that clown might be right.

On the flip side: Should bots become hugely succesful and the cost of effective and cheap labour therefore be lowered by a lot quickly, then logically, shortly after most of everything that exists should probably also be valued way less, given that the labour component is the key ingredient in...

Disagree. Elon is the CEO of Tesla. He has very recently and very clearly stated that the bot is the most important product. It makes a lot of sense to discuss the bot in this main thread.

The Bot App Store. Yes. Logical. Bound to happen. (Perhaps "Skill Store" is a better name) One difference comes to mind. Given how neural nets works there is also a value in elaborating on a 'loaded' practice. So, in most cases, the user might have to accept a limited 'upload' feature to ensure...

I get your point, but personally I like many of Dave's Tesla videos. Even the speculative ones. In fact many of the recent ones I find highly interesting. But ... to each his own. Also the spoken word can sometimes convey a message better than text - it varies. I find your sweeping criticism...

I'll give it a go: TESLA => tiger => STRONG! S and E and X and Y (Red lines: upper left, upper right, lower right, lower left) [Artful Dodger beat me to it. Well - it only strengthens my belief :)] Interpretation: Tesla: Strong and Sexy. Tesla: Fast like a Tiger! New Year is Tesla Year...

Quote: "When those with TA know how can predict price action with almost 100% success rate" Hm. I have never understood TA. But if the big players are truly somewhat coordination or colliding in their actions, there are definite patterns. If these patterns can be analyzed, then sure - TA might...

Very good point: That QA/warranty scenario will actually be very valuable! I was beginning to doubt that it was worth the trouble and hassle to bootstrap robots in a real-life production line, given that the first handful of iterations are bound to be somewhat lacking in both dexterity and...

Or bring earth with us, in 'carry on' fashion: IMDB

FTFY

Re. Bot: Elon said it was very important. We better believe him. The bot is being developed right now. (And perhaps have been, in stealth mode, for some time) Hard to know when it is a 'product' ready for sale. Or perhaps better: a worker or workforce ready for hire. 4-5 years sound plausible -...

In the long-term, optimistic, non-dystopian scenario, Bots will greatly aid in solving global warming. By lowering the cost of labour, whatever technology is developed to increase the uptake of renewables and/or carbon capture will all become cheaper. But isn't that true for all labour? Maybe...

Hm. My guess it that the Tesla FSD-team will have to do an abstraction layer or a set of hyper-parameters for the route planner. Something like: "Drive me from A to be, using whatever legal roads, using these two GPS locations. Assume that the map is sufficiently true to be useful for planning a...

Yeah. So, given that legacy auto are not very good at vertical integration and/or chip hardware and software, enabling a future Tesla OEM FSD package might not work from a practical standpoint. AFAIK, FSD is designed specifically for Tesla, where the 'bedrock' is a lot of instrumentation and...

Hm. Perhaps RT will over a couple of years become so cheap that it will steal customers from public transportation. And generate more demand. Also the standard pattern now is: Drive to work - park. Drive home - park. The cars are parked a lot of the time. New pattern will be: Drive customer -...

Cars will 'park' while driving to the next customer. Parking space on sideways and in dedicated buildings will become obsolete. Long term, the new standard form factor of the FSD-native car will in all likelyhood become a two-seater. Or even 1+ (parent plus child). Also FSD 'transporters'...

Also, consider the backup plan: What if in 2-3 years FSD is very good, and much safer than the average driver. BUT - still gets confused and stuck from time to time. And it still might make odd choices once i a blue moon. It is not flawless - there will be accidents and metal gets bent from...

In the new TAAS paradigm/pivot, FSD/robotaxi is the crucial thing. Cheap cars are nice, but the ~40K M3 is plenty cheap enough. (Subtract 30% if Tesla has their own fleet) So, a) solving FSD and b) ramping production to 2M, 4M, 8M and beyound is important. Cost saving on a new car is not. Off...

Old roadmap: Lots of car models. New roadmap: TAAS!

There are around 1 billion cars in the world currently, correct? So even at a 1-5 ratio, that takes 200 million robotaxis to replace. Perhaps a lot more, since above is steady-state. But, given that, if Elon is right, the future price of robotaxi miles are really cheap, then demand will...

If we take robotaxis *really* seriously, they are worth a lot: 'A big number'. They robotaxi lifetime value and price (subscription) dominates the cost of the car. It doesn't matter anymore if the car costs 25K or 50K - FSD makes the car worth 'a nutty amount' over its lifetime. It is the same...

FTFY (Might have to wait for 2030 though)

Almost true. But, if you can provide extra backup, in the form of remote human drivers, you can roll out before FSD is 99.9999999 percent perfect. Less will do. How much less? That is the trillion dollar question. If something weird occurs, the remote safety drivers can take over. Off course...

Yeah, that struck me as odd as well. For all the reasons you could be against robotaxis arguing entrenched marketplace - weird. Still very bullish considering that despite discounting robotaxi he had no qualms about proclaiming a) Tesla will be most valuable company and b) very soon. Weird...

Tesla and Elon has been under attack, scutiny and extreme pressure for more a decade and a half. They been pressed into diamonds. Tesla got weaned off steak and pudding, and learned to make bread out of stones and execute tall ideas and impossible plans with a shoe-string and a box-cutter...

Looking forward to Energy Day.

Agree. Also, just for fun: TSLA doubters, detractors and shorters parody: jan 2022: "WTF! 35% GM - that is crazy high and unsustainable! And irresponsible - what about all my clever analysis claiming something else!?" jan 2023: 2 million car sold - and 40% GM. "That's is madness! Competition...

They firms providing analysis (and analysts) do get paid, but by whom to achieve what - that is the question...

I am beginning to grow into the firm belief that like a lot of traditional media, the real customers for analyst are actually the corporations - and we, the public, is the product they sell. Now, the analysts often disagree. That has not so much to do with reality of the companies they are...

Thanks for the great contributions! Is there any chance you would hazard a guess going a couple of years forward? (i.e continuation of above table) I have some guesses regarding production of vehicles (which are crazy bullish) for the next 4-5 years, but I find it very hard to model Teslas...

So, asuming The Accountant is mostly correct, having the operating income a) increase 50% quarter over quarter and b) also just the mere fact of having 18% operating income while being a manufacturing company... Who does either? And ... who achieves both? Any previous examples of manufacturing...

OK - got it. But still - what is to stop Ford or other car companies from simply raise the price of the cars the sell to the dealers? I mean, they can clearly see that the demand is there in spades when people are willing to pay huge mark-ups. I understand that the dealers are making hay - the...

Re. Shanghai: Above run rate is without the new battery factory just now being erected (see WUWA video a few post back). Also without the around 3-4 times larger area where the parking lot/staging area for transport currently is. This is scheduled to become yet another factory building soon...

Norway is an EEA country so (with some exceptions) they agree to abide by all EU laws and regulations. And in Norway, like Germany, cars drive in the right side of the street so no steering wheel adjustment are needed. So no (or very few) differences should apply. Not sure about the color of the...

This might be OT but I am trying to understand Teslas advantage here. I don't get it ... what stops the traditional auto companies from doing a monthly or quarterly price adjustment re. the cars they sell to the dealers? Is such a thing prescribed by US law? I recall the many problems Tesla had...

Hm... Perhaps they are doing cold water testing in Norway? Come to think of it ... many parts of Europe is cold currently - perhaps a couple of thousand vehicles to do a proper test? Joking aside, is Elon really trying to pre-empt or second-guess the starting gun? That is gutsy - and risky...

My background is IT not law, but quite certain. Logically, a temporary permit only makes sense temporarily so it must be replaced by a permanent/final permit at some point. The germans may be inclined toward caution and bureacracy, but logic and reason is also a big thing.

Thanks for replying. I consider aquired companies to be in house. Do you have details on what parts are made in house versus by other companies?

No, not illegal. Current work is being done under a temporary permit, fully legal, where Tesla bears all cost for undoing that should the final permit for some reason be denied. That scenario is very unlikely. But dragging things out far longer than necessary, fair, or warranted - yes, that...

Great find by Joe. What are the chances that a lot of the pieces of equipment we see are made in-house? This is by the way primarily possible due the open and generous nature of Elon. Where else would drones have an almost free rein when an important and ground-breaking new factory is being...

Saw it and liked it! But a lot bigger than 2 cubic meters, no? The default capacity of the personal transport device is one person, so a lot smaller, think motorized wheel-chair on steroids, wrapped in bubble of some kind of composite material. Crazy, I know. Too crazy for right now, but maybe...

If (when) AI for real world driving become sufficiently good, the price of a transportation device can become almost artitrarily low. Why? Well, reasoning from first principles, why do we need all that metal? To protect in accidents. Well, what if, say in in 30 years or so, accidents with...

Short term there will be FUD. But longer term (~5 years) there might be actual, positive government cooperation. Why! Two reasons: Economic downturn caused by ICE having problems, going under or needing massive gov help AND also increasing transportaion costs adding to inflation and reducing...

Let Elon Elon. Yeah, CT may get delayed. But - that could actually be part of the overall plan. Making over-the-top tech is a valid excuse for the cybertruck to get delayed, all the while the real hardcore effort is all about ramping Model Y at Texas and Berlin, turning them into great cars and...