TSLA Market Action: 2018 Investor Roundtable

[Gerardf]

Firstly, let me defend why I'm posting this here: the AI Chip is a huge deal to Tesla's valuation and I don't think it has been understood by the market yet, let alone priced into $TSLA:

• Waymo has a current valuation of around 175 billion dollars. Tesla is ahead of them both with the AI chip and by having hundreds of thousands of cars on the road providing feedback.
• The 'autonomy market' is worth trillions of dollars:
  • "The study, prepared by Strategy Analytics, predicts autonomous vehicles will create a massive economic
    opportunity that will scale from $800 billion in 2035 (the base year of the study) to $7 trillion by 2050."

The troll is actively spreading disinformation with statements like this:



The troll's suggestion that Elon somehow does a drop-in replacement with some ARM chip is either fundamentally confused or intentionally confused: the replacement will be for Tesla's NVidia's GPU module, which is a self-contained computing node in a simple, modular form factor.

No-one truly working on self-driving is going to confuse ARM chips with GPU and TPU chips, these are drastically different chips that have drastically different size, power budget and thermal envelopes:

• Most ARM chips are general purpose CPUs with a couple of watts of power use typically, and much lower idle power usage. They are the Swiss army knifes of computing.
  
• GPUs are proprietary special-purpose vector computing CPUs with an effective processing capability of thousands of (very simple) CPUs.
• TPUs are even more special-purpose vector computing CPUs designed for convolutional neural network (CNN) processing.
• Tesla's AI chip is a new, grounds-up chip designed by two of the world's leading chip designers: Pete Bannon of Apple A5/A6 fame and Jim Keller of x86-64 fame. It's a new CPU, with functional prototypes probably manufactured as custom ASICs - but if Tesla wants to they could build their own chip manufacturing plant as well and make the chips themselves.
  
• Tesla's AI chip is, according to Elon, even more special-purpose, it was specifically designed for CNN AI processing:
  • Elon mentioned that the AI-Chip computation model is neural network centric: I read this as the Tesla AI-Chip using minifloats (very small size floating point data that fits into 16 bits), similar to Google TPU v3's bfloat format. This is a major advantage over GPUs, which typically do not support minifloats and waste a lot of RAM, computing bandwidth, chip real estate and power on calculating with 32-bit floats.
    
  • I also understood Elon's Q2 CC comments: "So, it's a huge number of very simple computations with the memory needed to store the results of those computations right next to the circuits that are doing the matrix calculations. And the net effect is an order of magnitude improvement in the frames per second." (note that I fixed the complications/computations mistake Elon made when he said this). I.e. the Tesla AI-chip's memory is merged/embedded with functional units on the design level, which avoids memory bus traffic entirely while a single matrix multiplication is ongoing (!). It's also possible that all memory is embedded in the Tesla AI Chip, an external DRAM interface is only used to initialize the networks, to feed input data and to communicate the results of the computation. This is possibly another big source of speedups.
  • Elon mentioned the following curious bit of information, which I think most people have missed: "whereas the current NVIDIA's hardware can do 200 frames a second, this is able to do over 2,000 frames a second and with full redundancy and fail-over. So, it's an amazing design and we're going to be looking to increase the size of our chip team and our investment in that as quickly as possible." (emphasis mine)
    
  • I.e. the AI chip is fully redundant and supports fail-over: Tesla extended redundancy to the functional units as well. If so then this is groundbreaking as well: competitors don't have redundancy of functional GPU or TPU units at all, they only have redundancy of RAM modules (ECC RAM). Neither NVidia nor Google TPUs have functional redundancy (fail-over) features, at all. (This might be an interesting scoop for @ZachShahan to look into?) I.e. the Tesla AI Chip will be robust against various types of hardware failure - it will in fact be pretty close to space rated and I'd not be surprised if SpaceX was interested in this as well.

I.e. the troll either doesn't know and is just a bullshitting SA author on the wrong side of a $TSLA trade, with free time to spread the chaos in his mind to other minds, or he knows it perfectly well and is lying intentionally about Tesla's AI chip to disrupt the flow of information on this forum.

Make no mistake: the Tesla AI chip is a ground-breaking design that took years to complete, build and test by a team of top CPU designers. No competitor comes even close to this design currently (not Google, not Nvidia, probably not Intel/MobilEye either) - and Tesla has working field units (!).

If Tesla wins the autonomy race, or just gets ahead of the pack and utilizes first mover advantage, then when correctly priced in $TSLA should be worth above $1,000 today.

IMHO an excellent post, but maybe better in the general thread than market action.

Actually, I feel it deserves a thread where it will not disappear in the noise when we are 2 weeks and 1000 posts further.
You worked hard on that one and personally I want to be able to find it back.

 

[ValueAnalyst]

$TSLA should be worth above $1,000 today.

I don't feel like a lone nut anymore.

 

[dondy]

Required by what entity?

None, of course, the certification is voluntary.
there are very few car companies which actually bothered to certify ECU, not talking about instrumentation clusters. (What the f%k is "digital cluster"?). and so far most of them have ASIL C. I didn't see any ADAS getting beyond ASIL B. This "Asil d ready" is nonsense made by people who don't understand letters they use.
Since this SA troll is banned I will just drop here to avoid miss understanding by general public.
Auto industry in-spite of anybody would think about is a consumer product industry and is governed by consumer laws.
Current version of ASIL-safety standard is applied only to consumer autos (up to 3500kg). There is it's adaptation for all auto types, but I believe it's not adopted and it's not going to. Because of obvious reasons.
There is industrial standard IEC 61508, which targets industry applications and which indeed has "system reliability" in it's text.
ASIL or in other words ISO 26262 is safety standard "blueprint" for auto-parts certification which quantifies how likely these part can cause somebody harm. The approach is based on three elephants:
1)severity of a malfunction,
2)probability of detecting this malfunction,
3)control ability of the malfunction.
From point of ASIL standard pulling car to the right and stopping in case of malfunction (emergency stop) and negating malfunction by redundancy are equivalent and are both successful actions.

And of course if to make a trivial search you get inevitable examples like
"Safety ≠ Reliability" (https://www.volpe.dot.gov/sites/volpe.dot.gov/files/docs/Assessment of the ISO 26262 Standard, “Road Vehicles – Functional Safety”.pdf)

or their german buddies in German Electrical and Electronic Manufacturer’s Association.
"The term "functional safety" should not be confused with or, worse still, equated to
product characteristics such as reliability, availability, and security"
https://www.nxp.com/docs/en/fact-sheet/FUNCTIONAL_SAFETY.pdf

Guys, please stay critical. Doing stupid things (like shorting Tesla stock) implies being stupid, but at some moment these guys would start to hire real professionals in information dissemination or even information warfare. Don't take any information presented with face value. there more hoaxes to come.

 

[SO16]

Does AI have the reasoning skill of a human?
Will it any time in the near future?
No? Then it needs better senses to compensate for its failings.

That said, I don't think LIDAR is key. I fully agree with Elon - if you're going to have a transmit-and-receive system, why would you double up on the same spectrum as your cameras, and face that spectrum's same limitations? Radar is the way to go. Sure, the radar spectrum has its own weaknesses, but they're different weaknesses from the optical spectrum.

That said, I don't see them yet using radar to its full potential. As far as anyone can tell, they're just using it as an object finder / distance to object calculator, like LIDAR. But radar has the potential to be so much more than that. I wrote an article on M3OC a while back about this, but the short version: have you ever seen a satellite radar map (generally made using SAR, but any radar can do it, incl. small vehicle-borne phased arrays)?

That's Baltis Vallis (Venus), the longest riverbed in the solar system. Here you're not focused on the timing of the echos, but rather brightness (signal intensity). What determines the brightness? It's partially the material and the angle, but beyond that, you're looking at the roughness of the surface on the scale of the wavelength. Where you see white, that's rough areas, while darkness is flat areas. By changing the wavelength, you can probe the surface roughness on different scales - anything from texture the size of grains of sand or smaller, to texture the size of potholes or larger.

Think of what this means applied to a road. With some smart software, you should be able to discern potholes, ice, bad shoulders, debris, rocks, pavement changes, etc before you get to them, and take corrective action in advance (maneuvering, reducing speed, etc), even when visibility is bad or the changes are subtle. Every type of road defect should have its own characteristic radar reflection. Even if there's a defect in how the road lines are painted that might possibly lead a less capable driver assist to drive off the road, the fact that you have a characteristic "pavement vs. not pavement" radar signature should allow your software to avoid that defect.

I imagine the radar they're using currently is locked into a single band. But even that could potentially be useful information. And there's always potential for more capable radars in the future. This sort of "information from senses that we don't have" gives autonomous vehicles the potential in the future to compensate for their lack of human-level reasoning skills.

Another example of how "information beyond our senses" can mitigate problems is the case of flooding. You don't want your car stopping on the freeway due to a puddle, but you also don't want it just driving into deep water and drowning your car (and potentially killing you). We as humans assess water depth by a lot of complicated reasoning which just isn't going to be easy to teach an AI. But, an AI can potentially get a leg up by other means: namely, historic driving data. If you build up a detailed profile of all the roads your fleet has driven on in the past - including heights - then you don't need to assess water depth, only height. Wherever the water reaches, you know how much the road drops down ahead of you (unless nobody's driven it before, wherein excess caution is due!), and thus how deep the water will get.

Current systems are a lot more primitive than this. And these sort of changes take time. So while I'm not much of an AP/FSD optimist in the short term, I do think there's a lot of potential in the longer term.

I’m all for having more sensors to “help” but all I’m saying is that vision is the only sense that should be “needed”. AI is increasing at an exponential rate. Sure, one can probably get to a higher level of autonomy a bit sooner with more sensors and less powerful AI in somewhat controlled situations but the long term needs to be strong AI without extra crutches.

 

[KarenRei]

AI is increasing at an exponential rate.

Not really. The field of AI has been missed prediction after missed prediction for much of a century. Pretty much every prediction that's been made has been too optimistic.

That doesn't mean it hasn't advanced - it most certainly has. There have been a couple nice leaps in the past decade, particularly with regards to image recognition. Just don't bet on human-comparable AI the day after tomorrow.

 

[dgodfrey]

IMHO an excellent post, but maybe better in the general thread than market action.

Actually, I feel it deserves a thread where it will not disappear in the noise when we are 2 weeks and 1000 posts further.
You worked hard on that one and personally I want to be able to find it back.

I think it's almost too late for that now. The toothpaste is out of the tube. There's really good counterpoints to the apparent FUD that you would have to pick out every post related to the discussion to provide fair analysis. It most likely will fall away in a few pages. (famous last words).

 

[SO16]

Not really. The field of AI has been missed prediction after missed prediction for much of a century. Pretty much every prediction that's been made has been too optimistic.

That doesn't mean it hasn't advanced - it most certainly has. There have been a couple nice leaps in the past decade, particularly with regards to image recognition. Just don't bet on human-comparable AI the day after tomorrow.

Oh I’m not thinking this will happen the day after tomorrow. It’s a few years out yet before self driving works on most roads in uncontrolled environments and routes.

 

[Beltsbear]

I can’t say too much on their architecture.

At least what I am working on. The general CPU is an ARM (I won’t disclose). GPU is still needed to process the raw image from camera and imaging radar. So i don’t buy Elon comment on a drop in ARM to replace the GPU. Because they serve different purpose.

The GPU is in 12/14 nm process. Hasn’t moved down to 7nm yet. With global foundry bailing out just couple days ago. GPU volume won’t be attractive to likes of TSMC and Samsung. So that move isn’t happening anytime soon.

The AP 2.0 box uses a mix version of drive PX2 platform (2 Parker and 1 pascal GPU). Which runs between 8 TOPS to 20 TOPS. Parker run around 1.5 Watts each. The GPU is what consumes most power. Tesla box only uses one GPU instead of 2. So it’s probably in range of 100W.

The nVidia Pegasus platform- which is the first gen true level 5 platform (which at least what Nvidia believes). Runs at 320 tops with total thermal power at 500 watts. This is close to 5 times more. 10 times may be an exaggeration on my part. But the problem is significantly different. . In order to not run Pegasus at full bandwidth as no one really know what urban will require.. A version of PX2 May be retained to perform level 2 and below function.

From the proposal I have seen- the power needed is close to 750 watt.

But then again Elon probably knows better than everyone . May be he can do it with less than 200 watt on AP 3.0. You never know.

You throw just enough knowledge to appear as a trustworthy expert but unfortunately you relay lots of false information.

Arm cores are cheap. Licensing cost is not an issue for Tesla. Yet you bring this up as a point of doubt.

Your comments on die process are laughable to those in the industry. Tesla will pay a little more for low volume, if you consider their demands low volume. But this would be a rounding error. They will go with whatever process they need to not exceed their thermal max.

Tesla is going for a more nueral chip design so most of your knowledge of gpu’s goes out the window anyhow. Done right it will process many times more data at a lower power usage.

 

[MartinAustin]

Engaging 'cave rescuer hero' who by the media is seen as a 'little man' in any negative, personal fashion isn't making any sort of point, even if the guy was a...

Just so we're clear, the guy who told Elon to shove the mini-sub up his rear end (and was the the recipient of Elon's slur) was not actually at the cave site, and not connected to the cave rescue at all. He is a retired diver from England who moved to Thailand in his retirement. (Not a lot of Brits retire to Thailand!) He was just an uninformed person opining on the events as they were unfolding, and said a little too much if you ask me.

Sorry for typing this in the stock movement thread.

TSLA support appears to be around $308. Let's see if that gets tested again today

 

[Gerardf]

I think it's almost too late for that now. The toothpaste is out of the tube. There's really good counterpoints to the apparent FUD that you would have to pick out every post related to the discussion to provide fair analysis. It most likely will fall away in a few pages. (famous last words).

Maybe, but we have to start somewhere, and alla dhere to it. Impossble for the mods alone.

Also these are good posts, but IMHO do not belong in market action but in General :
#31883 by @dondy, #31884 by @SO_S90D, #31885 by @KarenRei


Otherwise we need to create a new " The_REAL_Market_Action_Only thread."

People have no time to browse 100's of posts for a few market-action nuggets like : " What news broke that made the SP jump "

 

[davecolene0606]

None, of course, the certification is voluntary.
there are very few car companies which actually bothered to certify ECU, not talking about instrumentation clusters. (What the f%k is "digital cluster"?). and so far most of them have ASIL C. I didn't see any ADAS getting beyond ASIL B. This "Asil d ready" is nonsense made by people who don't understand letters they use.
Since this SA troll is banned I will just drop here to avoid miss understanding by general public.
Auto industry in-spite of anybody would think about is a consumer product industry and is governed by consumer laws.
Current version of ASIL-safety standard is applied only to consumer autos (up to 3500kg). There is it's adaptation for all auto types, but I believe it's not adopted and it's not going to. Because of obvious reasons.
There is industrial standard IEC 61508, which targets industry applications and which indeed has "system reliability" in it's text.
ASIL or in other words ISO 26262 is safety standard "blueprint" for auto-parts certification which quantifies how likely these part can cause somebody harm. The approach is based on three elephants:
1)severity of a malfunction,
2)probability of detecting this malfunction,
3)control ability of the malfunction.
From point of ASIL standard pulling car to the right and stopping in case of malfunction (emergency stop) and negating malfunction by redundancy are equivalent and are both successful actions.

And of course if to make a trivial search you get inevitable examples like
"Safety ≠ Reliability" (https://www.volpe.dot.gov/sites/volpe.dot.gov/files/docs/Assessment of the ISO 26262 Standard, “Road Vehicles – Functional Safety”.pdf)

or their german buddies in German Electrical and Electronic Manufacturer’s Association.
"The term "functional safety" should not be confused with or, worse still, equated to
product characteristics such as reliability, availability, and security"
https://www.nxp.com/docs/en/fact-sheet/FUNCTIONAL_SAFETY.pdf

Guys, please stay critical. Doing stupid things (like shorting Tesla stock) implies being stupid, but at some moment these guys would start to hire real professionals in information dissemination or even information warfare. Don't take any information presented with face value. there more hoaxes to come.

Appreciate the added color, read about it on Wikipedia which pretty much mirrored what is explained by you.

Just astonished (sorta) that it took me 30 sec on google to unwrap this misleading info.

Now if only some journalists could take the same 30 sec prior to publishing. (Or at least their editors)

Fire Away!

 

[davecolene0606]

Not really. The field of AI has been missed prediction after missed prediction for much of a century. Pretty much every prediction that's been made has been too optimistic.

That doesn't mean it hasn't advanced - it most certainly has. There have been a couple nice leaps in the past decade, particularly with regards to image recognition. Just don't bet on human-comparable AI the day after tomorrow.

Only thing I would add is that once GENUINE AI is released the leap will be immediate and more importantly, not necessarily understood. This is the challenge in controlling it.

The speaks directly to self driving and valuation of TSLA. Most of the time when all of the side cases are spoken of it’s referred to as some sort of data gathering exercise. If a true AI algo is embedded, the data will be secondary confirmation that the AI is working not required for the AI to work.

First to this finish line is the winner. Tesla’s sensor suite is more than adequate to support a true AI algo.

Fire Away!

 

[jeewee3000]

People have no time to browse 100's of posts for a few market-action nuggets like : " What news broke that made the SP jump "

That's what Papafox' 'Daily Trading Charts' thread is for! /s

 

[Fact Checking]

"Before the opening bell" update:

Quick early morning Tesla market sentiment report:

• NASDAQ futures are up +0.2% since TSLA post-market trading closed: NASDAQ futures up from ~7580-ish levels, bouncing on 7600 several times already. After yesterday's profit taking today could be another new all-time-high day, if there's no negative event or more profit taking. [Edit: just as I posted this a minor selloff started, so NASDAQ futures are flat now too. Obviously things could still change - 3+ hours left until market open.]
  
• DOW futures are flat with low volume, hovering within 0.15% of ATH levels. Nothing earth-shattering happened in the Asian or European markets so far. Trump didn't start a distractionary war tonight either, and the Brits are still full steam ahead for an epic demonstration of economic masochism come March 2019 and a hard brexit.
  
• TSLA pre-market trading is very light, only 200 shares traded so far, direction of the trade not worth mentioning. (Ok, before NASDAQ's web server gets overloaded: it was a BUY)
• Tesla FUD, illegal "Stock Bashing" (hello SEC?) and outright smear campaign against Tesla and Elon is still at ridiculous levels: the "40 Tesla Headlines On CNBC In 2 Days — 31 Negative, 2 Positive" article is quite an eye-opener. The press, shorts and a considerable portion of Wall Street only have 4-9 weeks left to break Tesla before Q3 falsifies 90%+ of the current FUD and exposes billions of dollars in trading losses that NY hedge funds were able to hide in 2018 through pre quarterly report TSLA price manipulation. (hello SEC, got some time beyond reading Elon tweets?) We need Pravduh to generate metrics like the ones in the CleanTechnica article and the ones in the Tesla FUD Tracker, like yesterday.

I.e. on a normal trading day, if things stayed like this, the environment would be conductive to modest, macro driven 1-2% gains for Tesla. But when was the last time Tesla had a normal trading day?

• NASDAQ futures are now red compared to where $TLSA traded at yesterday's post market close, there was a minor selloff, with an unclear catalyst (to me). Bounced back from yesterday's closing price, but still in the red.
  
• Dampened spillover into DOW futures, so the selloff was possibly technology related.
• There's some spillover into $TSLA pre market trading - within normal correlation patterns so far and resistance at $310.

Futures volume is still below what is normal during the open session, so any trends visible now could still be reversed - or continued.

 

[Gerardf]

That's what Papafox' 'Daily Trading Charts' thread is for! /s

That is a thread were mainly Papafox posts his daily analysis, with a few Q&A in between.
If we all go posting market action posts there (like the good post of Fact_Checking above this one), that thread will lose much of its value.

 

[Fact Checking]

I expect Tesla to have the option to issue shares, but instead choose to pay the converts off with cash.

Correct. This follows directly from what Elon said during the Q2 conference call:

"And from an operating plant standpoint, from Q3 onwards, I really want to emphasize our goal is to be profitable and cash-flow positive for every quarter, going forward. Now obviously, if there's a big recession or there's a severe force majeure event that interrupts the supply chain, that's not always possible, but we're confident that in, provided the economy is roughly where it is today or reasonably good and there's not a big force majeure event that we – I feel comfortable achieving a GAAP income positive and cash flow positive quarter every quarter from here on out. That's a – there may be occasional quarters, where we pay back a big loan or something, where there may be just because we paid back a big loan. But absent that, it would be cash flow positive."

(Emphasis added.)
​

He also said:

"We do not – we will not be raising any equity at any point, at least that's – I have no expectation of doing so, do not plan to do so. For China, I think, our default plan will be to use essentially a loan from the local banks in China and fund the Gigafactory in Shanghai with local debt, essentially. And we certainly could raise money, but I think we don't need to and we – yeah, I think, it's better to – it is better discipline not to."
​

The only loans they 'have to' pay back in large lump sums and cannot roll forward and pay off gradually are the convertible notes, $920m due next March.

The shares related to the 2019 notes are technically already registered, but actually issuing them would still be dilutive. So I think the second promise indicates that they don't want to pay back in shares - even if they have that option if the share price is beyond $360 by 2019.

So this is another TSLA valuation plus: new Gigafactories will not be paid with new shares issued, they will be paid through local loans, secured against the new factory itself.

No dilution to existing shareholders. Elon wants Tesla to use the Amazon path of self-financing through operations basically.

 

[Johan]

Even though I truly think we're at a very different stage now than before, most investors who have been with Tesla for a long while will vividly remember Elon saying they don't need to raise capital in the spring of 2016, then a few days later they went on to raise ~1,5 billion dollars through a follow-on offering. So let's not pretend like they couldn't raise capital again in the future. I chuckled back then, made med think of Bill Cliton itching his nose and saying "I did not have sexual relations with that woman, Ms. Lewinsky"...

 

[Fact Checking]

Even though I truly think we're at a very different stage now than before, most investors who have been with Tesla for a long while will vividly remember Elon saying they don't need to raise capital in the spring of 2016, then a few days later they went on to raise ~1,5 billion dollars through a follow-on offering.

I think there's a big difference between "being forced to raise capital" and "having the option to raise capital". In 2016 Elon referred to the fact that Tesla didn't need to raise capital and was thus negotiating from a position of strength. He probably tried to communicate this before the equity deal, and it's being held up against him.

In the recent Q2 conference call Elon not only said that Tesla wont be forced to raise capital, he also said that they have no desire to raise capital through equity either:

"We do not – we will not be raising any equity at any point, at least that's – I have no expectation of doing so, do not plan to do so. For China, I think, our default plan will be to use essentially a loan from the local banks in China and fund the Gigafactory in Shanghai with local debt, essentially. And we certainly could raise money, but I think we don't need to and we – yeah, I think, it's better to – it is better discipline not to."
​

I.e. Elon has learned this lesson from Bezos.

This is a much stronger statement than what Elon said back in 2016.

And yes, I don't think this promise from Elon of no shareholder dilution has been priced into the $TSLA stock price yet at $300-ish price levels.

 

[scaesare]

Interesting. …

[SNIP]

EDIT: I saw the mod note after writing this. Should I remove it? I felt like there were lots of replies from people who didn’t have as much knowledge about such things and that it might be useful in that regard, but if not, I’m happy to delete it.

For what it's worth, I started a thread (or tried to) HERE with my response to the issues in order to move off this thread.

 

[jeewee3000]

That is a thread were mainly Papafox posts his daily analysis, with a few Q&A in between.
If we all go posting market action posts there (like the good post of Fact_Checking above this one), that thread will lose much of its value.

I was joking, hence the "/s".

I fully agree that:
- the thread "Daily TSLA Trading Charts" is reserved for Papafox' input, only meaningful additions are allowed;
- the Market action thread is losing a lot of its value since it has become the new general thread.

Barking up the wrong tree, my friend.