-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
- Status
Two cars never arrive simultaneously at a four way stop for computers that detect and process changes at the millisecond level. This is strictly only a human issue due to how slow our minds work.
The shoe I can't see and am waiting for is larger scale and greater number of TE projects. Currently announced projects are nice " demonstration of use" projects, but, at most, equate to tens of millions of revenue for a $10billion company. If production is scaling and some of the Giga TE production projections are accurate for 2017, I would like to see larger scale "relationship" announcements with large utilities or even countries to absorb some of the capacity. It's going to be costly to sell all these potential power packs 1-2-10 at a time.
In the case that 4 autonomous cars that can communicate approach the "4 way stop" there wouldn't even be such a thing. 2 cars (arbitrary, but say east-west) wouldn't even slow down. North-south would completely stop and then proceed.
larmor
Active Member
I think price movement reflects bargain hunters in the tech sector. It is becoming more apparent that tesla is more a tech company vs. auto company, and a utility-energy company vs. auto company. The latter is important since new funding for infrastructure rebuilding is happening. The former is from a self driving car for the roads by april!-- of course with a driver still sitting in the car...
jhm
Well-Known Member
If the cars can communicate as they approach an intersection, none of them even need to stop. They coordinate speeds to preserve momentum. No need to stop at all.
No offense, but I have to say, I always thought the US 4-way stop thing was one of the stupidest traffic rules ever invented. Although, I must admit, the solution (whoever got there first has the right of way) is very "American", or should I say "Darwinian".
Still, if anything, machines should be better than humans at monitoring which vehicle got there first, down to the millisecond. All your car needs to know is which car came before itself and it should not be confused by the multitasking of looking at all 3 directions to determine that plus everything else going on.
Krugerrand
Meow
Everyone sees unicorns at some point in time.
It's publicized mostly because unlike everyone else they have an actual product in use by actual customers. Remember reviews of M-B version of autopilot? Yea, the only reason that got to make some rounds is because how bad it sucked compared to AP1 of a year ago.
Maybe I don't know something you do, but I just don't see a realistic competitive path. Tesla can iterate a hell of a lot faster on their system because they get to do "testing in production" like any good software service development team would.
People claim that there's so much unknowns and corner cases that it'll take a long time to figure them all out. Well, yes there are and let me ask you how would anyone find out about those corner cases? Isn't that obvious that Tesla at this point is more aware than anyone else about the kinds of corner cases that are out there? Wouldn't it also not be much of a stretch to say that of anyone they're the ones who are in the best position to estimate the amount of work it'd take to solve them? They've been doing that exact thing for a few years now, solving the corner cases.
jhm
Well-Known Member
SebastianR
Active Member
Would now be a good time to do another capital raise and start building GF2? (And maybe GF3/4, too?)
Without getting into politics but I see a "sell everything and the kitchen sink as long as you can make a buck" period coming from the fossil fuel industry. Granted, this may not last more than a couple (maybe 8?) years but still - wouldn't now be the time to intensify the race and give the carbon industry a mighty kick into the "free market"? And with $TSLA being where it is now - why not make good use of the time?
Without getting into politics but I see a "sell everything and the kitchen sink as long as you can make a buck" period coming from the fossil fuel industry. Granted, this may not last more than a couple (maybe 8?) years but still - wouldn't now be the time to intensify the race and give the carbon industry a mighty kick into the "free market"? And with $TSLA being where it is now - why not make good use of the time?
I agree, but I think it's a matter of time. There's significant resources already devoted to existing energy storage/peaker plants etc. I think the larger players will want to wait until there's solid data out there showing proof of concept, demonstrated cost savings etc before committing.
If the savings are as blindingly obvious as Elon says they are, I agree with him that TE will go ballistic. I think these small and medium sized installations will go a long way in proving to the bigger players that a lot of money can be saved by moving to a greener solution (which companies love to crow about for PR purposes). Once the big players step in, the economies of scale and unmatched pricing of Tesla will either crush the competition or drag them down to unsustainable margins - assuming Tesla can meet demand.
Oh yea. I am not sure if I maybe was too oblique there. When I mentioned lip-reading, that was a reference to it being an example of a hard problem. Guess what, AI is doing that better than humans now.
Jesse,
Sorry, I am too lazy to search for your earlier post from today where you basically said there may be a pullback after the Q4 ER due to the miss. I read that post during work but did not have time to ask this...
I always thought that one of the reasons the quarterly delivery/sales announcements about 1 month ahead of ER were also useful was, that they should have a bit of a stabilizing effect for the SP. Hit or miss, this gives analysts and investors a month to re-do their projections, so the ER should not be such a shock anymore. With basically 2k cars missed and an ASP of $100k-ish (depending on S/X mix), wouldn't you expect that by now everyone has a good idea on what Q4 financials were? So should we really expect a shock?
PS: I know, I know lots of unknown moving parts, SG&A and other stuff which we can just guess from earlier comments from the company, but still... ballpark should not be a surprise.
This is a good example of what the crux of the argument is. One side is saying this requires some kind of sacred knowledge on the part of the machine to resolve this. The other side simply says "hey that's parking lot speeds, just start driving and because AI is so much better than humans in tactical aspects, it'll avoid collision even if the other car does something stupid".
larmor
Active Member
Missed you myusername...
I have not watched NCIS, so i don't know what that means. Take data as recorded by the car sensors, and GPS data, and run this through multiple simulations on supercomputers. Learn where mistakes were made and teach the machine...
Sam Harris: Can we build AI without losing control over it? | TED Talk | TED.com
jesselivenomore
Member
Again I have to reiterate that the gist of my posts today revolve around the 13 month analog setup that points to potential major breakout between now and 3-4 months from now(latter looks more likely). A small part of this analog also portends to a small pullback before this breakout, but requires SP to follow the analog to the T, which rarely happens. This pullback does fit with our current short term overextention and upcoming earnings used as an excuse. I know delivery numbers are already priced in, but since SP have rallied so much, market may use earnings to pullback anyways. But to emphasize I don't really suggest taking action on this since it is too micro, too much a gamble whether it will happen or not. Perhaps instead they preannounce blowout Q1 deliveries during earnings and we breakout instead, who knows. It is better off figuring out how to accumulate in anticipation of the 13 month breakout than to try to time a pullback that may or may not happen, IMO.
RobStark
Well-Known Member
Trump has used the word repeatedly. Not just twice in the debate.
"Bigly" is an archaic adverb.
Still in Websters dictionary.
So heads up in ridiculing Trump in "mixed company."
Somehow I doubt Trump picked up the word reading Chaucer.
- Status
Similar threads
- Sticky
