Carl Raymond
Active Member
No clue what this means, but it sounds good.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable
- Thread starter AudubonB
- Start date
2daMoon
Mostly Harmless
Dang it. All I have saved are rolls of War Nickels, where they replaced the Nickel content with mere Silver 'cause they needed the Nickel for WWII.
I bet they won't be worth a dime.
jschwefel
Tesla fan/TSLA, Model Y and Cybertruck owner.
Sorry…I should have been more specific. 90% silver coins only included dimes, quarters, half dollars and dollars 1964 and earlier. Nickels were actually 75% copper and 25% nickel except from 1942 to 1945 when they went to 35% silver in order to use the nickel for the war effort.
Heh, I was just making a joke about the value of nickel going up. Copper should see a nice boost too. Silver will undoubtedly keep on climbing. Metals in general are a good investment.
Didn't we just establish that the value of the coin for its metal content stayed about the same factoring inflation? So it's better than sitting in a checking account, but compared to, say, TSLA how can we say it's a "good investment?"
I'm saying that many metals are a good investment from this point in time forward. Tesla stock may have better growth potential, but most smart investors like to diversify to some degree.
CLK350
Member
Anyone knows: wouldn't it make sense then for GigaShanghai to do a US designed Y production run and ship it to the US? Would accelerate sales - tho I understand revenues would have to be spent on Chinese goods, still ..
CLK350
Member
So rather buy another Tesla (the car) for its lithium content (and rent it out to someone you trust* ? /s tho it may turn out to actually be true
(*) DM me if interested, I'm tired of waiting for EmmaY my Oct Y ... which I'll finance anyway
jhm
Well-Known Member
Add to this picture the idea that new ICE vehicles are more fuel efficient than the ones they are replacing, and you'll get that peak fuel consumption happens several years before peak ICE fleet. I estimate that 2019 was the last year that enough new ICE cars were added to support fuel consumption levels. See the "Shorting Oil" thread for details.
I wonder if the global auto fleet has become more fuel efficient over the last 5 or so years. I'd like to think so, but larger SUV style vehicles have grown enormously in popularity, and diesel passenger car sales have plummeted. At least we're at the stage in the growth curve where EVs are beginning to make a big contribution to reducing fossil fuel needs.
CLK350
Member
How can you be sure your Bolt isn't subject to random fires? I thought it was a design issue, or was the issue limited to certain models only?
[OT] Some local color (New York City) - as the SP does its thing and we read
I finally tested the Volvo EV (C40 recharge) in NYC - it is in stock and available for delivery now, about 45K minus incentive (of course, now checking, they only have the more expensive 60K versions in stock) . Amusing detail, the frunk is inexistent, and they kept the central hump that used to hold the (front) engine to rear wheel transmission in place (maybe using it to store part of the battery).
The result is a rather stuffy interior, like standard ICE's - thick paneling (outer metal skin to interior surfaces. Plenty of buttons, and (good!) completely Google / map driven interface.
When I quizzed the salesman about why I would prefer this over the Model Y he had a stock answer to the effect that the Volvo was more of a luxury car or something to that effect. Seems to work for some .. a woman (truck driver) said she was ordering one (not interested in Tesla) - being on the heavy side etc I could tell the "plush" and familiar feel of ICE cars was enough to convince her. She also had no feel or interest for mechanical/ design notions.
We talked about this a few months ago, that a big factory will be built in Hungary, because of the large amount of money involved, I thought it was related to Tesla. It’s unofficial yet, but it’s likely to be a CATL factory.
www.portfolio.hu
Egyre több az árulkodó jel: hatalmas autóipari beruházás jöhet Magyarországra
Egyre több jel árulkodik arról, hogy Debrecenben épülhet meg a globális akkumulátorpiac legnagyobb szereplÅjének, a kÃnai CATL-nek a legújabb európai gyártóbázisa â Ãrja a vg.hu.
www.portfolio.hu
Ah yes, the luxury of having to waste 4hrs charging on a L2 because all the DC fast chargers were broken. Let's just say my friend can't stand going long distance in his Volvo that's a company car. Almost got stranded multiple times thanks to the pos unreliable charging network.
Last edited:
Isn’t that a picture of a Polestar 2?
I thought the Volvo C40 had the Volvo logo in front?
Incorrect. I worked in 1964 for $1 an hour. I could buy about 3 gallons of gas more or less. A cheap VW got about 25 mpg and I didn’t care what a coin was made of as long as it got me down the road.
Today for $10 to 15 an hour I can buy about the same amount of gas BUT I can travel much further on that gallon and be safer doing it. I get much more value for that hour of labor In terms of travel value.
This is arguably because the US love affair with our cars and our wonderful people with some ingenuity and values, advanced travel value greatly With better tires, air bags, safety glass, colapsing steering columns and air conditioning. This even ignores the last decade of EV advancement.
The value was advanced by us. We give our currency value through our great people and resources. Can we do better, sure but we have done well and will do better once we unite around a vision for a better future.
Today I hate gas and move everything to electricity which gets me even more value for my labor.
So now China is reducing tax on iCE car purchases.
Cheap Russian oil ?
China halves purchase tax on mainstream ICE vehicles for rest of the year
The purchase tax will be cut in half for passenger cars of 2.0 liters and below with a purchase date between June 1 and December 31 and with a price not exceeding RMB 300,000.
cnevpost.com
Imo this is pretty huge news:
I will summarize it, but James Douma can probably do a better job of explaining it. I am speculating, feel free to correct me.
Before Tesla was using a huge neural network:
Before: Input 8 camera images, neural network does some magic, output dense rasters which kind of looks like images in birds eye view. These images are then processed by c++ into lots of ”car 10m in front,10cm to the right” statements.
Now instead Tesla are adding GPTs(Generative Pretrained transformers) that start with the output of the previous neural network(or from earlier layers of the neural network) as input and then writes the output that c++ normally would have done.
Now: Input dense rasters of birds eye view images, neural network does its magic, output a string with list of objects like ”object:car x=10m,y=-0.1m,size=6x2m,rot=0rad. object:lanemarking, type:solid, b-spline=[…”.
I could be wrong and that they are not fully there yet. Maybe it just outputs how many lanes there are, where each lane goes ie ”1=left, 2=forward, 3=right”. But eventually I assume they want it to output more and more.
I assume they have to run the same transformer quite a few times to finish the output. But maybe then they don’t have to discretize and will have lower errors from discretization and they don’t have to waste cpu time in c++.
I really look forward to hearing how they did this at the AI Day #2.
I will summarize it, but James Douma can probably do a better job of explaining it. I am speculating, feel free to correct me.
Before Tesla was using a huge neural network:
Before: Input 8 camera images, neural network does some magic, output dense rasters which kind of looks like images in birds eye view. These images are then processed by c++ into lots of ”car 10m in front,10cm to the right” statements.
Now instead Tesla are adding GPTs(Generative Pretrained transformers) that start with the output of the previous neural network(or from earlier layers of the neural network) as input and then writes the output that c++ normally would have done.
Now: Input dense rasters of birds eye view images, neural network does its magic, output a string with list of objects like ”object:car x=10m,y=-0.1m,size=6x2m,rot=0rad. object:lanemarking, type:solid, b-spline=[…”.
I could be wrong and that they are not fully there yet. Maybe it just outputs how many lanes there are, where each lane goes ie ”1=left, 2=forward, 3=right”. But eventually I assume they want it to output more and more.
I assume they have to run the same transformer quite a few times to finish the output. But maybe then they don’t have to discretize and will have lower errors from discretization and they don’t have to waste cpu time in c++.
I really look forward to hearing how they did this at the AI Day #2.
Last edited:
Similar threads
- Locked
- Locked
- Locked
