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

[hobbes]

Model X goes airborne for 41 feet, then tumbles 172 feet down an embankment, flipping twice. All five occupants survive.

Charges filed in Maroon Creek crash involving 5 Aspen teens

“The police at the emergency room said they haven’t seen an accident like that when people walked away,” said Elana Royer, mother of 18-year-old Lily Royer. “They’re almost always fatal.”

The 18-year-old driver has been charged with two misdemeanors, including carless driving.

Color me confused !

 

[Off Shore]

The hope / speculation has been Dojo does end-to-end training. Basically you just input video - and the output tells how to drive. No object recognition or anything like that. Essentially it is unsupervised and may be not even labelled.

That ofcourse would need massive amount of training time and hardware. Karpathy basically talked about the hardware part. He didn't say anything about the end to end training part - though he mentioned fully automated training - "project vacation".

BTW, even with the massive pressure on autonomy team, its good to see them maintain a sense of humor. Shows good morale.

He specifically mentioned that the labelers will remain chained to their workstations.

 

[Fact Checking]

When people talk about reverse engineering, it's almost always to refer to the case where you don't have the source code. It's very time consuming, but it can be very exact. You get all the instructions that make up a program, which is usually enough to discover whatever you want about it, even if packed (which Tesla software is not). Certainly reverse engineers don't really care about source code. They're used to working with assembly and pseudo-C from hex-rays (and now Ghidra).

Yes, but this doesn't change what I said: Tesla's software is large and complex, it's easy to miss something if all you have are binaries, and SoylentBrown already went on the record claiming that verygreen's FSD instrumentation and tracing isn't complete.

​

 

[Fact Checking]

BTW, on a bit of an unrelated topic: recent discussions about Model S and overheating, the different motor types, etc has led to a realization: the new Roadster with the SpaceX options package should be able to regen directly to the COPV.

The power levels needed to refill the COPV in any remotely reasonable period of time are huge . It's going to need the power output of one or more of the main traction motors. This means a direct physical connection to the motors. Ideally in the rear, since the COPV is in the rear, and there's two motors back there that could be utilized rather than the one in the front. But if you have a direct physical linkage with the motors, then the regen torque coming in from the wheels can also go to the compressor. So you can regen directly to recompressing the COPV. Brake into a turn, rocket out of it

I agree, and this is my working hypothesis as well, but there's one big complication: the insane adiabatic heating compressed air undergoes as it gets compressed into the COPV.

The resulting compressed air will likely be several hundred degrees Celsius hot (!), and will thus be a significant fire hazard. Regular carbon fiber COPVs are not designed for high temperature storage - their usual design targets are cryogenic temperatures. All the high pressure ductwork from COPV to the 10 nozzles distributed across the car will have to be fireproof and well insulated as well.

It's all solvable I think, but a major headache - and I'm wondering whether I got this right. Could someone double check the numbers: how hot will 100-200 kg of room temperature air become if compressed into a 300 liter COPV to over 100 bar pressure adiabatically? (I.e. at least 1:100 volumetric compression.) Intuitively the temperatures get really high.

On the plus side this should be very efficient energy storage, and the exhaust will be room temperature (instead of chilled), which (IIRC) improves the specific impulse of the thrusters as well.

 

[clmason]

I must have missed the conversation about Elon tweeting and letter exchange with the short guy.

Posting my plea for Chairwoman Denholm to immediately nix this exchange. Nothing good will come from this. We finally recovered to $300+. Stock should never again move based on twitter chest thumping.

 

[2virgule5]

Fully Charged posted an interview with the Rivian engineers discussing their battery design. They allude to having better energy density compared to "competitors" due to better cooling methods. Any thoughts on this.

Tesla is continually improving anyway, but interesting to see if there is another company who can match them in this area.

relevant point starts at 13:00 mins

Words are cheap. Let’s see when they deliver an actual product that you can buy. Great if its true, beating e-tron does not seem hard, catching-up to Tesla in this regards with their Gen 1 platform would be surprising.

 

[2virgule5]

I agree, and this is my working hypothesis as well, but there's one big complication: the insane adiabatic heating compressed air undergoes as it gets compressed into the COPV.

The resulting compressed air will likely be several hundred degrees Celsius hot (!), and will thus be a significant fire hazard. Regular carbon fiber COPVs are not designed for high temperature storage - their usual design targets are cryogenic temperatures. All the high pressure ductwork from COPV to the 10 nozzles distributed across the car will have to be fireproof and well insulated as well.

It's all solvable I think, but a major headache - and I'm wondering whether I got this right. Could someone double check the numbers: how hot will 100-200 kg of room temperature air become if compressed into a 300 liter COPV to over 100 bar pressure adiabatically? (I.e. at least 1:100 volumetric compression.) Intuitively the temperatures get really high.

On the plus side this should be very efficient energy storage, and the exhaust will be room temperature (instead of chilled), which (IIRC) improves the specific impulse of the thrusters as well.

Thanks. You’ve just added one more option in my burn-the-short toolbox

 

[Prunesquallor]

I agree, and this is my working hypothesis as well, but there's one big complication: the insane adiabatic heating compressed air undergoes as it gets compressed into the COPV.

The resulting compressed air will likely be several hundred degrees Celsius hot (!), and will thus be a significant fire hazard. Regular carbon fiber COPVs are not designed for high temperature storage - their usual design targets are cryogenic temperatures. All the high pressure ductwork from COPV to the 10 nozzles distributed across the car will have to be fireproof and well insulated as well.

It's all solvable I think, but a major headache - and I'm wondering whether I got this right. Could someone double check the numbers: how hot will 100-200 kg of room temperature air become if compressed into a 300 liter COPV to over 100 bar pressure adiabatically? (I.e. at least 1:100 volumetric compression.) Intuitively the temperatures get really high.

On the plus side this should be very efficient energy storage, and the exhaust will be room temperature (instead of chilled), which (IIRC) improves the specific impulse of the thrusters as well.

SCUBA tanks are pressurized to 200 bar. Here’s an electric commercial compact (29”x15”x19”) compressor that’s can pressurize to that at 96 L/min. Looks like it just has a fairly small fan for cooling.

MCH6 Compact High Pressure Compressor

I’m sure Tesla could figure out how to put that into a smaller package

 

[Fact Checking]

A few have done more detailed analysis and found higher ratios. But they are short sellers with black hearts full of Musk-hate, so we know every word they write is a blatant lie.

If you read through that analysis you'll see that those TSLAQ types moved heaven and earth to identify gaps and encoding errors in vehicle fatality databases - which increases the number of Tesla fatalities 3-fold. Their research, focused on increasing the Tesla fatality rate, is literally several pages long.

Then they compare the resulting Tesla fatality rate to BMW and Audi, performing the following amount of due diligence:

"Remember that BMW and Audi combined had 9 driver fatalities in nearly 900,000 vehicle years?"​

Yep, that's all the fact checking they did. Despite BMW and Audi fatalities being similarly infrequent as Tesla fatalities, they give no thought whatsoever to the accuracy of those BMW and Audi statistics...

They, just like you, simply accept the IIHS numbers, which are based on the same federal databases they found so lacking for Tesla fatality reports...

The second methodological error both the article and your comparison is suffering from is the comparison of Tesla's "vehicle miles driven fatality rate" to the IIHS "vehicle years fatality rate".

The two metrics are fundamentally incompatible: if the average 2012-2016 Tesla, with unlimited Supercharging and a superior "car experience" is driven more than the typical Audi or BMW, then there can be a significant increase in the vehicle-years fatality rate, even if Teslas are safer in terms of fatality rate per miles driven.

Third, neither the article nor your comparison factors in driver age: older drivers suffer from higher fatality rates, due to higher fragility and lower regenerative abilities to survive life threatening injuries.

With the average Tesla price of ~$100k in the 2012-2016 timeframe, the average Tesla owner age was significantly higher (10 years more) than that of less expensive luxury ICE vehicles.

Fourth, their list of Tesla fatalities include two "driving off a cliff" fatalities and one "driving into a pool and drowning" fatality, which are unlikely to be fatalities related to car safety - and which kinds of accidents I suspect are not encoded in the federal databases for BMWs and Audis driving off cliffs either.

These are four serious methodological flaws.

Sorry, but this is perhaps one of the worst researched posts of yours I've read on this forum - delivered in an arrogantly superior tone.

 

[Fact Checking]

SCUBA tanks are pressurized to 200 bar. Here’s an electric commercial compact (29”x15”x19”) compressor that’s can pressurize to that at 96 L/min. Looks like it just has a fairly small fan for cooling.

MCH6 Compact High Pressure Compressor

I’m sure Tesla could figure out how to put that into a smaller package

But they are certainly not doing it in seconds, right?

I.e. that scuba tank compressor is largely isothermal (constant temperature) - while the COPV compressor will be adiabatic (constant energy).

Very different results.

 

[Krugerrand]

An interesting and funny inside story from the early Model S days.

Gilbert Passien (sp?)

 

[JRP3]

Sorry, but this is perhaps one of the worst researched posts of yours I've read on this forum - delivered in an arrogantly superior tone.

Doggy turning back into a bear.

 

[Unpilot]

But they are certainly not doing it in seconds, right?

I.e. that scuba tank compressor is largely isothermal (constant temperature) - while the COPV compressor will be adiabatic (constant energy).

Very different results.

It is the rate of change that will drive the higher temps.
Rapid compression=rapid rise in temps
Rapid expansion= rapid decrease in temps.

 

[Artful Dodger]

OT:

Just had a conversation with my sister, who is new to the Tesla ecosystem. She had her first ride in a Model 3 last Sunday, and says it was an eye-opening, mind-expanding experience. Her curiosity is fired up, and she's imagining future possiblilities for Tesla.

I just told her about the Tesla Cybertruck reveal coming on Nov 21. She asked what a 'Cybertruck' was, so I explained some of Elon's statements about the truck, its rumored features, and new projections from Jack Rickard / EV TV about potential reservation numbers (400-500K).

She had just one question: "Will it work on Mars?"

Lol, this from a woman in her 60s, who first heard the name "Elon Musk" at our local National Drive Electric Week event in September (8 weeks ago).

ISYN, this is a true story. The times, they are a'changin'

Cheers!

 

[Artful Dodger]

Musk : As an outside investor there are two things you should track
- FSD
- Battery

TMC :
BEEF APPLE FARTS​

FTFY.

 

[Unpilot]

She had just one question: "Will it work on Mars?"

Lol

I believe it would

 

[AudubonB]

It is the rate of change that will drive the higher temps.
Rapid compression=rapid rise in temps
Rapid expansion= rapid decrease in temps.

Seemingly, but not true.

PV = nRT is independent of time. If you use time as a factor, you are providing a period wherein the heat generated may dissipate (again, in the case of expansion it appears to be coolth but that would be because one is looking from the "wrong" end of the heat pump).

Gee, can I change my sig-line to something more like "I'd like to live in a universe where the Laws of Thermodynamics can be violated"?

 

[bhtooefr]

Color me confused !

Clearly the driver had heard about those newfangled driverless cars, and was even driving a Tesla, but still, somehow, got it backwards.

 

[AudubonB]

Aaaannndd...Done!

At the expense of Mild-mannerism. AAAaaaaawwww.....I really liked that reference!

 

[MarcusMaximus]

Seemingly, but not true.

PV = nRT is independent of time. If you use time as a factor, you are providing a period wherein the heat generated may dissipate (and again, it appears to be coolth but in the case of expansion you're just looking from the "wrong" end of the heat pump).

Gee, can I change my sig-line to something more like "I'd like to live in a universe where the Laws of Thermodynamics can be violated"?

You’re actually both correct. The time doesn’t change the amount of thermal energy coming out of(or going into) the system, but the lack of dissipation when it happens in a smaller amount of time leads to higher(or lower) temperatures within it.