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.
My biggest takeaway from that was when Rob mentioned that Jim had previously spoken before about market manipulation and how it's done and asked him to talk a little about that, Jim misunderstood and immediately went into all the short sellers claims of how Tesla bulls manipulate the share price higher using hype. In other words, Jim thinks the "manipulation" story centers around a high stock price, not a low stock price.

Jim's a smart guy but this video brings out just how much he doesn't know about Tesla as well as how much he misunderstands. Rob did a good job of letting Jim speak his mind but I wish he would have educated Jim a little better. Maybe he can talk with Jim off the record and fill him in on some Tesla specifics. For example, I wish he would have explained to Jim why the Cybertruck looks like it does (exoskeleton) and how this makes for a superior truck while simultaneously lowering cost and reducing weight. I wish he had cited the payload and towing figures and mentioned ALL the accessories Cybertruck buyers get as part of the purchase: motorized tonneau cover that you can walk on (and that locks your stuff inside the "vault"), built-in air compressor, fully adjustable ride height front and rear (variable ground clearance for off-roading and ease of loading/unloading), built-in loading ramp strong enough for a battery-electric four-wheeler, and how the cab is roomier for big people than even a Ford Crew cab. He did mention the built-in inverter for mobile power and it looked like that alone blew Jim's mind!

These accessories would cost many thousands of dollars if you had to add them on but they are included with every model version of Cybertruck.

Yes, there were many more features and details about CyberTruck that Rob could have brought up. But, I think he rather wanted Cramer to talk than be talking himself. It was a fascinating discussion and Rob can learn a thing or two from Jim Cramer as well, in terms of getting his points across. Jim is a very seasoned commentator and a master in boiling things down to a simple argument and communicating that in an entertaining way so that everyone immediately gets it.
 
So they are planning on selling 90% ICE vehicles a lot of that time even if EVs achieve price parity with ICE.
If EVs are available from anyone (including Chinese car makers) for around the same price as ICE, many customers will go with the EV.
Plans change all the time, especially when they need to change.

Remember this?

Tesla%2BBad%2BPress%2B2009.jpg


Took them like what? 10 years to finally deliver this turd of a vehicle?

Even when a manufacturer actually has made a serious attempt and really wants to deliver a lot of EVs, such as with the Porsche Taycan and VW ID.3, traditional OEMs have shown that they don't deliver on their promises. The Taycan's numbers appear to be lackluster so far at ~4,500 in H1'20, and the ID.3 is WAY behind schedule.

Mass manufacturing EVs is not the same thing as mass manufacturing ICEVs, and most OEMs still don't seem to take EVs seriously even though it's already 2020. Many automotive executives, such as Mary Barra, still seem to be of the opinion that EVs won't take hold until after 2040.

Going by what OEMs have shown over the past decade, I think it's at most a single digit percentage likelihood that Tesla will have as little as 20% EV market share in 2030. I think it'll probably be around 40% market share, but I'd say there's a decent ~30% chance that Tesla will have a near 50% or 50%+ market share in EVs in 2030.
 
IMO the main point of Battery Day is Tesla about being to scale in house battery production quickly.

If we remember the original leak about Tesla having ordered Battery Formation Equipment that was for Fremont, Nevada, Shanghai and Berlin.

Berlin making cells for Model Y production is more or less confirmed...

No process scales instantly, the key is to scale cell production in sync with vehicle production.

I also expect Battery Day to say plenty about Energy Storage Batteries, but again production will not scale instantly.

They are also getting more cells from Panasonic, CATL and LG... fair chance these are long term supply deals...

While (I hope) they can scale quickly, it takes, capital, staff and raw materials to scale... there is no point in scaling too fast, that is worse than being too slow.
Scaling fast has zero downside if they are leaving money on the table from storage. Also it's still holding semi back. If 20gwh of batteries fell out of the sky Tesla would blow off the lid on revenue this q and next q.
 
Going by what OEMs have shown over the past decade, I think it's at most a single digit percentage likelihood that Tesla will have as little as 20% EV market share in 2030.

While all of that is true, we can take investment in Battery Factories as a leading indicator, more cell volume typically means someone is more serious.

At that is also the clear path for Tesla to accelerate the mission, make more cells, hence more energy storage batteries and vehicles, Anyone that wants to defend their market share has to play along.

IMO 1-2 determined race leaders can definitely set the pace, anyone who does respond risks falling further behind and becoming less relevant.

To determine who is serious we would need to look at the battery factories being built then trace those cells back to a car program...
Even when they have the batteries sorted, there is still getting everything else to work well. but batteries are the cap on their level of ambition.
 
This is 43 great minutes.

Rob really shines. He is such a valuable member of the blogging community. After this discussion, Kramer is very glad he added Rob to TheStreet.


I really enjoyed the discussion between Rob and JIm. It just shows that even while Jim has turned bullish there's still so much he doesn't know or really understand about Tesla.

One thing that really stood out for me was Jims expectations for battery day being heavily focussed on the million mile battery. He seemed to view a million mile battery as enabling an EV to drive for 9 hours straight with massive range. Thankfully Rob elightened him about what a million mile battery means and that Tesla probably already has this in their stroage products. The interview with Dan Ives on Yahoo a couple of days ago was similar with his 2 big items for battery day being a million mile battery and Tesla becoming a 3rd party supplier of batteries. This shows a relatively shallow focus and poor technical knowledge, research and understanding of a lot of the financial community.

For me this raises the prospect that what Wall Street is expecting from Battery Day is at a mismatch with what many of us expect to be the key takeaways. Some Wall Street analysts could come away initially disappointed that their pet theories aren't central and miss much of the significance of the day. At worst this could lead to a similar response to autonomy day. However I expect the importance and significance of what Tesla presents won't be fully lost on them, even if it takes a few days for everything to fully sink in.
 
If the Tesla chip is faster, uses less electricity and can run the same native code as Tesla cars, they’re going to use it. If Nvidia was so great, why are they waiting to build Dojo? If Dojo comes after the first v4 chips, it seems rational the v4 chip will be used for Dojo.

Tesla also does software in the loop (SIL) and hardware in the loop (HIL) based simulation and testing. For HIL they would use their own chips of course. The simulation data would be generated using NVIDIA GPUs. I'm convinced Tesla will be using NVIDIA compute technology in their data center for a long time.
 
I don’t think that is the case, if it’s just NV GPUs, why would they simulate it using FPGAs?
I am guess he is referring to the current system which would be replaced by Dojo 1.0 as Dojo 0.1?

Supercomputer - Wikipedia

As the price, performance and energy efficiency of general purpose graphic processors (GPGPUs) have improved,[38] a number of petaFLOPS supercomputers such as Tianhe-I and Nebulae have started to rely on them.[39]

....
A number of "special-purpose" systems have been designed, dedicated to a single problem. This allows the use of specially programmed FPGA chips or even custom ASICs, allowing better price/performance ratios by sacrificing generality.

We don't know which route they are going with Dojo, but like HW3, this is a case where sacrificing generality to get a really good solution to a particular problem might make sense..

It is also the case that Dojo might have multiple building blocks of different types...

For example HW4 might be use as pre-processor to process video, merge camera feeds and cache labelled video examples... or this kind of speculation could be (and probably is) far wide of the mark
 
Last edited:
  • Helpful
Reactions: UkNorthampton
Supercomputer - Wikipedia



We don't know which route they are going with Dojo, but like HW3, this is a case where sacrificing generality to get a really good solution to a particular problem might make sense..

It is also the case that Dojo might have multiple building blocks of different types...

For example HW4 might be use as pre-processor to process video, merge camera feeds and cache labelled video examples... or this kind of speculation could be (and probably is) far wide of the mark
OT:
My doubts and argument stands. In general, when you say use FPGA to simulate something, it usually means you are building a custom chip. Else you won’t use the word “simulate”, you would call your FPGA an “accelerator” as Microsoft and some others did in the past.

And if Tesla is building a custom training chip, it won’t be a supplement to GPUs, I doubt it would even need an x86 CPU, they would likely have some ARM cores in it and be very self-contained.

I know this won’t be a good news for any Nvidia investors, hence they would try to muddy the water before its clear what Tesla is doing.
 
Last edited:
To determine who is serious we would need to look at the battery factories being built then trace those cells back to a car program...
Even when they have the batteries sorted, there is still getting everything else to work well. but batteries are the cap on their level of ambition.

This is definitely useful to track, but the ex-Audi executive in one of @avoigt 's recent interviews blamed the lack of e-tron sales on the supply chain, which I don't buy at all. I think they had plenty more capacity in the supply chain, but almost nobody wants that car. Otherwise why has it been discounted so much...
 
May I recommend a mini-split air conditioning system?

The indoor units install easily on a wall or ceiling with a small condensing unit on the outside of the house (ground or wall mounted). They’re extremely quiet and offer better energy efficiency than the best central units. Multiple zones allow you to condition the rooms you want, saving even more money and reducing the impact on the grid. Their low electrical requirements make them ideal for houses with solar and could be powered by Tesla Powerwalls. They’re available in sizes from 9,000 BTU (0.75 ton) to 60,000 BTU (5 tons) and up to five zones per system with energy efficiency ratings up to 22 SEER.

Stay safe out their. Friends in the Bay Area have been sharing pictures of the smoke and it looks quite unpleasant and downright dangerous for the folks in the immediate vicinity of the fires.
Mitsubishi and Fujitsu 30.5 and 33 SEER last I checked!