Fortunately Tesla has/had some very smart people working on it. IIRC they figured they needed about 5-6 Billion miles driven/recorded for it to be "2-10x" safer than a person, according to youtube videos by Sterling Anderson, which is a great way to learn about how Tesla is doing autonomous. Anyone remember if those miles recorded included cars with the early sensors or is it just from the cars with the newer sensor systems?
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
- Status
I noticed a potential loophole in TipRank's algorithm. Some analysts are taking advantage of it to improve their rank. There is an analyst who issued 104 "Buy" ratings on Apple in a few years. So his TipRank ranking is very high. But based on his bearish stance on Tesla, and a few calls on other companies, I think he is way overrated.
My point is when you see a high ranking analyst, always take it with a grain of salt. Look at how he achieved the high ranking. Look at his reasoning behind each call, you probably tell if he is intelligent and honest.
My point is when you see a high ranking analyst, always take it with a grain of salt. Look at how he achieved the high ranking. Look at his reasoning behind each call, you probably tell if he is intelligent and honest.
racer26
Active Member
What? I can't boil trustworthiness down into a single metric to judge people by?
Morgan Stanley Predicts 10 Million Tesla Vehicles On The Road In 10 Years
Jonas- note to clients:
“With the launch of the Model 3, we forecast the Tesla car population to multiply three times by the end of 2019. It has been generations since the investment community witnessed such a high growth rate in the population of a single auto firm.”
Jonas- note to clients:
“With the launch of the Model 3, we forecast the Tesla car population to multiply three times by the end of 2019. It has been generations since the investment community witnessed such a high growth rate in the population of a single auto firm.”
Sterling Anderson had a speech about his autonomous research before he joined Tesla. His approach was that the self driving car keeps running in this imaginary tunnel, only react when it's interrupted. I think that's a completely wrong approach to solve autonomous driving. I can't believe Elon hired him. The vehicle should pay attention to the environment, react before it's too late. Just like how a good human driver drives.
Depending on the correlation between the idea and the expression of it, that might not be such a bad approach (haven't heard the speech, so ignore this if I'm way off). The core algorithm could be operating the driving with a tunnel view while other processes watch the environment and track objects. If an object's trajectory will intersect the tunnel, the item/ event is passed along to the main driving routine based on the intersection point /time. Then the main routine reacts as needed. Pedestrian a mile ahead, no issue. Pedestrian 200 ft head, brake.
RobStark
Well-Known Member
Just think of it! Those two, Musk/Bezos or Bezos/Musk, both are such easy-going people, tolerant to a fault. Sure, and they'll grow to be collaborative with each other.
Were such a thing to happen the entire world would be revolutionized, but I'm really not at all sure just how.
Can anyone with actual auto manufacturing experience comment on this please? Does it look like the guy is speaking truth?
The related article and comment source can be found here:
S3X Appeal
Add: My key point is if this guy is speaking truth (which I don't know) with respect to Dodge Ram assembly then he is saying the line is producing more than 420K/annum back in 90's with so many variations of the vehicle.
Then even when Tesla achieves the 500K rate in Fremont, that doesn't feel like a significant achievement.
Elon claimed that in the long run (automated) manufacturing capability is what will be the key competitive advantage to Tesla. That proposition looks dicey if the Dodge Ram assembly info is actually truthful.
The related article and comment source can be found here:
S3X Appeal
Add: My key point is if this guy is speaking truth (which I don't know) with respect to Dodge Ram assembly then he is saying the line is producing more than 420K/annum back in 90's with so many variations of the vehicle.
Then even when Tesla achieves the 500K rate in Fremont, that doesn't feel like a significant achievement.
Elon claimed that in the long run (automated) manufacturing capability is what will be the key competitive advantage to Tesla. That proposition looks dicey if the Dodge Ram assembly info is actually truthful.
Last edited:
erthquake
Active Member
And how long did it take, from when it was first built, for that line to achieve the one vehicle per minute rate?
After the nuclear war, insects shall rule even more and they are so cooperative, although still tribal.
From the article:
So the question is why, with so much pressure to produce cars to meet presumed unlimited demand, is output so limited?
The author keeps asking why Tesla's factory output is limited to 25K / quarter if demand is "unlimited", utterly confusing the demand for $100K vehicles (S/X) with Model 3's demand (455K+ reservations). I don't get it at all. The whole premise is faulty.
It seems like such a fundamental mistake that I must be misreading it...
Umm ... which world are we talking now?Just think of it! Those two, Musk/Bezos or Bezos/Musk, both are such easy-going people, tolerant to a fault. Sure, and they'll grow to be collaborative with each other.
Were such a thing to happen the entire world would be revolutionized, but I'm really not at all sure just how.
I thought (from reading this forum) the limiting factor is the CA/ EPA cert for the paint booth.
My key point is if this guy is speaking truth (which I don't know) with respect to Dodge Ram assembly then he is saying the line is producing more than 420K/annum back in 90's with so many variations of the vehicle.
Then even when Tesla achieves the 500K rate in Fremont, that doesn't feel like a significant achievement.
Elon claimed that in the long run (automated) manufacturing capability is what will be the key competitive advantage to Tesla. That proposition looks dicey if the Dodge Ram assembly info is actually truthful.
Ford model T ran at that rate. Elon speaks to the base knowledge of his audience. Jobs did the same. To resounding applause. Almost like a flashlight and a... cat.
The differentiation is in labor content, step and repeat capability of the entire factory, battery costs and most importantly a development process that produces cars that people want to buy.
larmor
Active Member
I'm sure I'll be able to buy a bridge on Amazon Prime too..
Not sure what speech you are referring to, but his talk at EM Tech in May 2016 seems like a very smart approach, they are basically turning the whole fleet into recording devices and doing machine learning. I believe that the 5-6 Billion miles driven served two purposes. First of course it's a ton of data to learn from. 2nd, it's probably got a lot of redundant data built in for liability so that safety can be proven beyond a reasonable doubt in court, the legal ramifications of this are pretty huge, so it has to be bulletproof before it rolls out to the public. In other words it seems like they probably have already pretty much got most of self-driving down, maybe that's why Anderson left, hence West coast to East coast drive scheduled for later this year. The extra 3 Billion miles or whatever is probably just to prove that the safety numbers are legit.
Interesting thing...if they haven't already hit 6 Billion miles, it should be reached in the next 6 months for sure.
Would highly recommend watching
Warren Assembly (WA) is only part of the process. From: Dodge City: Chrysler’s Warren Truck Assembly Plant
WA gets the parts from Warren Stamping and engines from the Mound Road Engine plant (until it closed and they switched to Mack Avenue). Axles come from Detroit Axle down the road. Frame currently comes from Mexico (not sure on in the 90's).
So the truck was the results of five+ factories. Tesla is getting motors and packs from Sparks, so they are at two factories.
Assembly process is also different:
WA runs two paints lines, and it also runs two assembly lines, one for the cab and one for the chassis.
Post 1996, the doors were removed post painting and assembled before rejoining the cab.
Tesla is unibody, so this parallelism is not possible (well, could take doors off), but chassis, stamping, and cabin are all in house and fairly linear.
If you double the production lines then you get to 1 million vehicles a year. Or, increase automation and drop tack time on a single line.
https://www.allpar.com/corporate/factories/warren-truck.html
per: Warren Truck Assembly Plant | Hendon Publishing
2013 production was 291,500 RAM 1500 pickups
2015 production was projected at 320,000.
Moving this to the General thread ....
Some of this may be semantics, but I like to separate out two distinct things -- (1) the elements of a classic short squeeze where shorts are forced to scramble to find shares and short more and more on the way up versus (2) the effects of relieving the downward pressure from excessive levels of short selling when shorts decide to cover.
In 2013 IMO there were elements of both of these. Shareholders who saw the rapid price moves as a classic short squeeze might have been tempted to sell when the share price doubled, tripled or quadrupled from 30 to 60, 90 or 120 because they expected the price to drop when the "squeeze" was over. But as it turned out, the more important force at play IMO was a relief of downward pressure from an extreme level of short selling that had caused the share price to be kept artificially low even after the Model S was delivered and was receiving glowing reviews from all corners.
To put this in context of the Model 3 era, if Tesla delivers on its Model 3 plans, we could see rapid price moves upward driven in part by short covering reducing the supply of available shares (and possibly reduced FUD), but the stock could still be a bargain based on fundamentals. I don't consider this a short squeeze.
- Status
Similar threads
- Sticky
