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

[Nocturnal]

Semi-OT, but Tesla brand/valuation related.

The incident with the Tesla sentry mode catching the guys keying a car, that's excellent free advertising for Tesla. This video made the local news at least and had the anchors saying "what's Sentry mode etc.". Yes, it's a FREE thing we received after delivery. Cool right?

Also, anecdotal, but last night I was washing my 3 in my neighborhood (affluent part of KC suburbs). I had two neighbors stop to ask about the car. The first asked me how I liked it and mentioned she was considering one. The other literally said "Where did you get your car?". She had no idea what a Tesla was until recently and had no idea where to buy one.

People who think demand has peaked are very much wrong. If people still don't even know about the cars, or where to buy them, that's a huge pool of untapped customers.

 

[ReflexFunds]

Exclusive: Tesla's solar factory is exporting most of its cells - document

Hopefully, it’s not true but no reason to doubt it that I know of.

I expect this headline is probably true, but it doesn’t look like Reuters has any evidence to prove it.
It sounds like Panasonic has set up a free trade zone in GF2 and within this part of the factory most cells will be exported (obviously) to module manufacturers who will then sell the modules back into the US. They don’t have any information about what % of the factory’s production will be in the FTZ vs non-FTZ. It’s also possible Tesla is still buying some of the modules which are assembled abroad.
"It is fully anticipated that the majority of the cells to be produced" in the FTZ "will be exported," the company said in its application.

I’m sure Tesla will start to use more cells with their new lower solar pricing and after they ramp solar roof and their retrofit module production.

 

[neroden]

I've just been browsing the Model S and Model X delivery threads on TMC. It looks like with both vehicles people have been (impatiently and without much feedback from Tesla) waiting for their deliveries of the refresh versions of the vehicles.

Yep. The reason for lack of deliveries is that the refreshed versions flat out haven't been delivered yet. Production problem of some sort (might be homologation).

So the S/X deliveries will look bad in Q2 but it has nothing to do with demand -- it's supply, again. I'm setting myself up to not expect good news recognizable by the market until Q3.

In both vehicles, there's someone claiming to be ready to take delivery within the next couple of days. Let's see how quickly the deliveries get moving. We have half a quarter in front of us.

@dqd88, there are a fair number of people on this forum who own millions of dollars of TSLA. It's really not so uncommon. BTW, I'm not one of them (yet).

*cough* I believe I understand AcesDealt's perspective.

 

[neroden]

Exclusive: Tesla's solar factory is exporting most of its cells - document

Hopefully, it’s not true but no reason to doubt it that I know of.

Eh, reasonable short-term move. They are still debugging the Solar Roof.

 

[DurandalAI]

To me FSD is not determined by Elon if it's near or not. It's determined by the people who will put FSD to the test after it's released. Once Tesla decides to flip the switch, there will be more miles driven on FSD mode in the first 24 hours than Waymo's entire existence. The miles driven will be more varied than Waymo's and will be monitored by unpaid morons. This is actually the scary part, and also why Tesla is slowly testing the waters. I wouldn't be surprised if FSD gets delayed again due to this.

I'd argue that I'm paid pretty well for being a moron. Of course, my payments come not from sitting in the car, but selling options on the stock... Which then pays for the car.

 

[JRP3]

Now I'd add that he's very pessimistic on timescales and on Tesla in general, he just doesn't believe that battery packs have enough charge cycles to be viable and that Tesla are somehow gaming the system to show such good degradation over time.

He doesn't know what he's talking about.

To me Maxwell is a stepping stone to solid-state that is highly manufacturable.

It's not.

It adds density, cycle rates and the more stable materials allows for higher charge rates and less cooling.

Data supporting higher charging rates and less cooling? The Maxwell process does not remove flammable electrolyte from the cell, it removes a solvent during the construction process which does not remain in the cell after drying anyway.

It's solid-state light.

No, it has no relation to solid state technology.

People assume some type of NMC 811 would be required for 5k cycles but the energy density is too low and the batteries would be too heavy, thus the physics jokes by Daimler.

NMC being less volitile needs less protection and less cooling, so the cell level density differences may not translate to pack level.

Maxwell would allow a zero Cobalt NCA battery..

How does the Maxwell process eliminate Cobalt?

 

[humbaba]

Does it? lol!

Dan

I'm not sure how much sense your question really makes to ask. It sounds like you already know what you want to do but are looking for some confirmation. That said, I'll have a go:

- put money in $TSLA -- in seven years $10k might be worth $100k (or more), or maybe it will be worth $10k (or less). As you said, its volatile. Opinions here vary on how volatile it will be in the long term, but I think the safest prediction is that it will continue to be so. This is the higher risk gamble.

- put money in a $TSLA focused ETF (ARKK) -- in seven years $10k should be worth considerably more, but if $TSLA does take off like the optimist case from above then an ETF will necessarily not gain as much. In short, this is a lower risk gamble. Using an ETF that invests heavily in $TSLA keeps some of that high payoff option in play, but mitigates the risk.

- put money in a general ETF -- in seven years you should have about what the market improves by. While that could be down it is far less risky than $TSLA volatility. Limited reward for limited risk.

But I can't help thinking you knew that already.

If it were me it would depend entirely on the distribution of my investment. Currently I'm ~97% in on $TSLA and if I had a windfall would currently put it in $TSLA, but with the intention of pulling out on any substantial upward movement and tossing it into ARKK. My reasoning is simple: unlike the bears I think $TSLA is so undervalued at this point that there will be an unpredictable surge upward. If I bought at $230 I'd probably sell it off somewhere around $300, but definitely not above $350. The goal isn't to maximize gains, but to realize profit and diversify.

But it sounds like you aren't so heavily invested in $TSLA (smart!) so if $TSLA is only a small part of your portfolio then going long with the new money could make sense. Just keep $TSLA a reasonable fraction of your investment (no where near mine...).

Even better advice: don't listen to me, but someone like @tivoboy instead.

 

[gtrplyr1]

Just a quick comment from Sunny Southern CA.

It’s unbelievable to me how many Tesla’s I see here on a daily basis. If Tesla could sell cars around the rest of the US like they do here it would be fantastic. I just spent a month in Dallas last January and would spot one occasionally.... not sure what the rest of the country is like but the statistics I’ve seen show that a huge majority of their cars are sold here in CA.

Just a matter of time before it happens ... IMHO.

 

[Bobbyducati]

Just a quick comment from Sunny Southern CA.

It’s unbelievable to me how many Tesla’s I see here on a daily basis. If Tesla could sell cars around the rest of the US like they do here it would be fantastic. I just spent a month in Dallas last January and would spot one occasionally.... not sure what the rest of the country is like but the statistics I’ve seen show that a huge majority of their care end up in CA.

im in NOVA/DC area and i see easily 15+ everyday. theres typically 10 or so in my work garage by its self. they are all over. a healthy mix of S/3/X too, not just 3's

 

[Cobbler]

He doesn't know what he's talking about.


It's not.


Data supporting higher charging rates and less cooling? The Maxwell process does not remove flammable electrolyte from the cell, it removes a solvent during the construction process which does not remain in the cell after drying anyway.

No, it has no relation to solid state technology.


NMC being less volitile needs less protection and less cooling, so the cell level density differences may not translate to pack level.

How does the Maxwell process eliminate Cobalt?

Time for you to read the Maxwell presentation from Jan '19, slide 9-10 and be amazed!!
Solid state, no cobalt, higher density and more cycle lifetime

And afterwards you watch this video

 

[mongo]

OT

Still not sure as if a cell reached 0 volts the cell gets ruined. So the cells at monitored to ensure they do not reach 0 or rather do not go below 2.5 volts. There is no practical way to insert a good cell for a week one. At a minimum the may replace the brick.

With the cells in parallel, the voltage at any cell (steady state) is the voltage at all the cells. So one cell cannot hit 0 votls, and the group is prevented from hitting 2.5V by the BMS.
What does happen is that a week cell contributes less power/ current for the same voltage drop at the strong cells.
Like 5 containers of water that are all connected and the same height, but one is narrower than the others.All at the same level, but when the water is drained, the narrow container contributes less. Or think of a strong cell as two weak cells grouped together in one package.

I agree it is not practical to swap in (or remove) cells. One would be better off having as many cells active as possible to reduce the SOC swing of the group.

 

[humbaba]

I’ve heard the number of 1,500 workers in India needed to manually sort through about 1 billion frames to end up with 3 million labeled images. This is per year and not low-skilled workers so at a significant cost.

Interesting, thank you. Where are your numbers from? Is this relating to Tesla or a general rule of thumb you’ve heard in the industry?
I heard Tesla largely outsources data labelling, but I’d be interested to know where to and how many hours of manual labelling work per year they are paying for.

Your numbers imply c.400 images reviewed per person per hour with c.1.25 useful labelled images per hour.
I would have expected Tesla’s selective data collection/automatic filtering to result in much more than a 0.3% acceptance rate at the level of the human labellers.

The implied rate of work with Indian workers seems a bit high to me -- I would not think quality work could be expected at those rates unless there was an excellent workflow to support it. Interestingly, one of Karpathy's talks mentioned the need for an effective workflow and that Tesla had in-house developed a solution. I think it is doable to human process that many images with assistance but it would require a custom solution. Just tossing people at it would not scale well IMO.

One thing that I think is also relevant here is Karpathy's talk of autonomous labeling. He gave a specific example that Tesla had done. While autonomous labeling is necessary for labeling at scale, I don't think it can dispense with the need for competent human labelers as not everything is suitable for automatic labeling.

 

[humbaba]

Time for you to read the Maxwell presentation from Jan '19, slide 9-10 and be amazed!!
Solid state, no cobalt, higher density and more cycle lifetime

And afterwards you watch this video

That's talking about the electrodes. Given on 10 where it says "solid state" I can understand confusion, but the title is "Dry Battery Electrode".

Solid state may, or may not, pan out as useful for automotive cells. But Maxwell isn't it. Which, given that they are a capacitor company, isn't exactly surprising.

 

[JRP3]

Time for you to read the Maxwell presentation from Jan '19, slide 9-10 and be amazed!!
Solid state, no cobalt, higher density and more cycle lifetime

And afterwards you watch this video

Unless Maxwell has some unannounced method to remove the liquid electrolyte from the cell their dry electrode manufacturing does not make a solid state cell. I see no explanation how their dry electrode manufacturing removes cobalt from cells. I've seen Jack's video, I don't remember him talking about Maxwell allowing solid state cells or cobalt free cells, maybe you could point out the specific parts in the video?

 

[Alphacrux]

Just a quick comment from Sunny Southern CA.

It’s unbelievable to me how many Tesla’s I see here on a daily basis. If Tesla could sell cars around the rest of the US like they do here it would be fantastic. I just spent a month in Dallas last January and would spot one occasionally.... not sure what the rest of the country is like but the statistics I’ve seen show that a huge majority of their cars are sold here in CA.

Just a matter of time before it happens ... IMHO.

I agree. In visiting San Diego, Teslas seem to be as common as dirt. In Austin, I usually see two or three 3’s and an S or X in a short span of time. Yesterday, I nodded at an ICE driver on the highway who was staring at my car. As he passed, he gave me a thumbs up. I wasn’t looking too closely, so I’m just pretty sure it was a thumb and not a different finger.

 

[humbaba]

Just assume a worst case of 70% of rated range. Some people are reporting worse, but I frankly don't believe them.

Taken at face value, I don't think that's true. You've made statements before acknowledging that 50% range is certainly possible when making numerous short trips with cold soaks between them. Its the "I drove a long haul at speed and got <50% range" that I've seen you take issue with.

However, I believe this particular bit started with "OMG how will Canadians cope with an SR+ in the winter." The answer to that is, I think, pretty simple: the same way they will in warm weather or the same way they deal with the short range of bolts or leafs.

Sure, there should be a CCS adapter for Canada, I get that. But I'm not sure how many Canadians are planning on long drives in -40C blizzards in areas with poor super charging network. Certainly the indications of take rate suggest this isn't a pressing concern for a large number of buyers.

 

[Spacemanspliff]

Yep. The reason for lack of deliveries is that the refreshed versions flat out haven't been delivered yet. Production problem of some sort (might be homologation).

So the S/X deliveries will look bad in Q2 but it has nothing to do with demand -- it's supply, again. I'm setting myself up to not expect good news recognizable by the market until Q3.

I drove by the factory the other day and saw S&X's coming out (obviously unable to verify rate) but on the call they made it sound like most of this quarter would be spent ramping production and only at the end of the quarter would they achieve close to old production rates.

It makes sense to make international S&X's first to go abroad and target domestic S&X demand later in Q2 (Also anecdotal note all M3's I saw on carriers at this time were international spec, so one could surmise the S&X's I saw were also international spec).

I see anywhere from 15-20k S&X this quarter, I'd like to say 18-20k but when it comes to ramping production Tesla does not seem to get it right often.

 

[Spacemanspliff]

I agree. In visiting San Diego, Teslas seem to be as common as dirt. In Austin, I usually see two or three 3’s and an S or X in a short span of time. Yesterday, I nodded at an ICE driver on the highway who was staring at my car. As he passed, he gave me a thumbs up. I wasn’t looking too closely, so I’m just pretty sure it was a thumb and not a different finger.

Californian as well, same thing I noticed in New York and Boston. Of all models I probably could have counted 8 or so between the two cities and I was there for 10 days.

 

[ggies07]

Just a quick comment from Sunny Southern CA.

It’s unbelievable to me how many Tesla’s I see here on a daily basis. If Tesla could sell cars around the rest of the US like they do here it would be fantastic. I just spent a month in Dallas last January and would spot one occasionally.... not sure what the rest of the country is like but the statistics I’ve seen show that a huge majority of their cars are sold here in CA.

Just a matter of time before it happens ... IMHO.

What part of Dallas because they are every now (North Dallas and Southlake area have the most).....I see more Model 3s now than S/X.

 

[BornToFly]

I've been investing heavily in TSLA since 2013. I'm currently down an insane amount in paper losses since 420 tweet. I'm starting to lose faith on TSLA being more than a $100 Billion company anytime soon. We've been told that TE would grow exponentially and be larger than auto since 2016. With battery cell shortage, I don't see how TE can grow significantly at this point for another several years. I do expect auto sales to continue to grow 50% per year, but that isn't enough. Full self driving needs to become a reality, and I bought into the hype and ordered a new Model S P100DL (no VIN after almost 3 weeks), but I'm becoming skeptical that we will even see real traffic light functionality this year. I admire Elon, but his predictions/timelines are useless, and have caused tremendous financial harm to believers like myself that trade based on what he says.