Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

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

This site may earn commission on affiliate links.
Musk’s penchant to turn everything into a scalable process is a huge competitive advantage. While Electrify America outsources the design of the chargers and orders a few at a time giving their suppliers little incentive to reduce prices or optimize installation. Tesla is not only chasing bringing down production costs, but figuring out how to make them and install them 6 at a time to reduce installation costs.

IMHO, the smartest thing Tesla did with Superchargers was making the chargers as dumb as possible and putting all the smarts into the car. This gives them a huge advantage over other charging networks, particularly in reliability of individual charging stalls.

If they open their network to other cars, they will be printing money and the investment will pay for itself many times over. Why go to Electrify America station and take a chance that many (most?) of the units will be broken when you can go to a Supercharger and rarely encounter a broken one.
 
Musk’s penchant to turn everything into a scalable process is a huge competitive advantage. While Electrify America outsources the design of the chargers and orders a few at a time giving their suppliers little incentive to reduce prices or optimize installation. Tesla is not only chasing bringing down production costs, but figuring out how to make them and install them 6 at a time to reduce installation costs.
That's true. Tesla has a massive built-in advantage.

Electrify America has four different suppliers for their fast chargers.

I believe they did this because no single supplier could deliver them fast enough. Electrify America wanted to build out the network as quickly as possible. And they did a decent job of slapping together a coast-to-coast network that serves all models of EV in just a couple of years.

The "all models" part was particularly difficult as every car model has its own charging quirks and it took a long time to gain some semblance of reliability. I wonder if Tesla is having similar issues as they start to accept other brands at their Superchargers in Europe.
 
And when did "charging" become "recharging"? The author using recharge and recharging exclusively in this article. I kept asking myself if recharging was something different than charging...
I emailed the author on the "recharging" thing and he fixed the article already. I sent a link to Tom Moloughney's you tube channel "State of Charge" so he had some of the terminology from an expert.

Main stream media.....at least he listened.
 
IMHO, the smartest thing Tesla did with Superchargers was making the chargers as dumb as possible and putting all the smarts into the car. This gives them a huge advantage over other charging networks, particularly in reliability of individual charging stalls.
I hadn't really thought about this, but it makes a ton of sense. Explains in a lot of ways why the Superchargers are so much more reliable than the EA ones.

The "all models" part was particularly difficult as every car model has its own charging quirks and it took a long time to gain some semblance of reliability. I wonder if Tesla is having similar issues as they start to accept other brands at their Superchargers in Europe.
If you have 4 suppliers that means your repair techs have 4 systems they have to learn and it means any cars which present problems charging you have to solve the issue 4 times for each model.

Since Tesla owns 70% of the US EV market, I don't think they will worry overmuch about whether a particular competitors car works with their network. That's the car-maker's problem to solve isn't it? The bigger problem is dealing with the idiots in Porsches who think they are entitled to use any Supercharger even though the charge port is in the middle of the car and won't work with back in stalls at all.
 
I hadn't really thought about this, but it makes a ton of sense. Explains in a lot of ways why the Superchargers are so much more reliable than the EA ones.


If you have 4 suppliers that means your repair techs have 4 systems they have to learn and it means any cars which present problems charging you have to solve the issue 4 times for each model.

Since Tesla owns 70% of the US EV market, I don't think they will worry overmuch about whether a particular competitors car works with their network. That's the car-maker's problem to solve isn't it? The bigger problem is dealing with the idiots in Porsches who think they are entitled to use any Supercharger even though the charge port is in the middle of the car and won't work with back in stalls at all.

I do think Tesla will try to make things work with all the other EVs. That's what Elon seems to have promised.

But I am wondering how it's going to work. Maybe Tesla will sell a Tesla-to-CCS adapter at cost. The adapter could have a couple of feet of cable that acts as an extension cord to accommodate cars with ports that are hard to reach.
 
I do think Tesla will try to make things work with all the other EVs. That's what Elon seems to have promised.

But I am wondering how it's going to work. Maybe Tesla will sell a Tesla-to-CCS adapter at cost. The adapter could have a couple of feet of cable that acts as an extension cord to accommodate cars with ports that are hard to reach.
In one of the conference calls Elon mentioned something about having adaptors at the stations. Not sure how this would work. My guess is the simplest solution would be to add the second CCS cable to the US stations like in the EU. Tesla did this fairly quickly when they launched the model 3 in the EU with CCS.
 
The points you make vary in their usefulness, relevance and (spectrum here): importance—silliness. Here is the important one, to which I respond:

But twitter, covid, political stances that have DIRECT and indirect influence on share price, forget it. Why?

Why: Because, unfortunately, personal opinions, axes to grind, rebuttals of varying degrees of relevance ALL not only veer off simultaneously in all directions at once, so to speak, rendering this thread not a salad but an utterly impossible-to-navigate hairball. And this is because all of you have intelligent yet magpie minds - full of ideas, life experiences and deductive powers that you love to share.
Just want to point out the bold.
 
  • Like
Reactions: PeterJA
FSD Beta did this for me as I monitored my MS while it executed the commands ... I recently had this experience on the Garden State Parkway(NJ) and navigated around an accident ... it was impressive ...i did intervene a few times but I could see single stack executing this flawlessly

There is no doubt, FSD beta v10.12.2 is now routinely executing complex routes w/o intervention, and demonstrating very safe decision making:

Tesla's Robotaxi Future is Getting Closer | FSD Beta TESLA CHALLENGE #28 | "Dirty Tesla" on Youtube


When "single stack" is deployed (possibly coming with FSD beta 10.13), the source of training data will span the entire fleet of FSD-capable cars. That's a subtle point that I think most people don't yet grasp: Standard Autopilot will run Single Stack, and ALL such cars will collect training data, whether the owner has licenced FSD or not. Wide deployment of FSD (and revenue recognition) will soon follow as edge cases are exposed quickly in this avalanche of data.

FSD is like the cheese rolling downhill. Robotaxi will arrive sooner than many think, and deployed in the busy, congested cities that need it most. HODL. :)

Cheers!
 
There is no doubt, FSD beta v10.12.2 is now routinely executing complex routes w/o intervention, and demonstrating very safe decision making:

Tesla's Robotaxi Future is Getting Closer | FSD Beta TESLA CHALLENGE #28 | "Dirty Tesla" on Youtube


When "single stack" is deployed (possibly coming with FSD beta 10.13), the source of training data will span the entire fleet of FSD-capable cars. That's a subtle point that I think most people don't yet grasp: Standard Autopilot will run Single Stack, and ALL such cars will collect training data, whether the owner has licenced FSD or not. Wide deployment of FSD (and revenue recognition) will soon follow as edge cases are exposed quickly in this avalanche of data.

FSD is like the cheese rolling downhill. Robotaxi will arrive sooner than many think, and deployed in the busy, congested cities that need it most. HODL. :)

Cheers!
“Cheese” ?
 
Meh, chill pill is in order for many. This is not a mandatory contract to be here and post here. This is an internet forum.

I feel for the mods and I feel for us, the little people. We've lost a few great posters but this was their decision. It is easy to point out fingers at the mods for some of the departures but overall the show must go on.

For those that have been here for a while, self-moderation is really helpful. Thre have been very many political posts, abortion posts and it has not been easy for the mods. Many have attached their personal feelings to those posts and this is never good in such a large and diverse group.

The good I see - it is that overall this forum has been able to very quickly weed out the pest posters that show up for a bit and write idiotic stuff.

Keep your political and religious beliefs at home folks. There are many very heavy topics that don't belong here. Keep it rolling, don't behave like 12 year olds. It will be good.

Go Tesla.
 
Meh, chill pill is in order for many. This is not a mandatory contract to be here and post here. This is an internet forum.

I feel for the mods and I feel for us, the little people. We've lost a few great posters but this was their decision. It is easy to point out fingers at the mods for some of the departures but overall the show must go on.

For those that have been here for a while, self-moderation is really helpful. Thre have been very many political posts, abortion posts and it has not been easy for the mods. Many have attached their personal feelings to those posts and this is never good in such a large and diverse group.

The good I see - it is that overall this forum has been able to very quickly weed out the pest posters that show up for a bit and write idiotic stuff.

Keep your political and religious beliefs at home folks. There are many very heavy topics that don't belong here. Keep it rolling, don't behave like 12 year olds. It will be good.

Go Tesla.
 
No FSD step changes since last year? Really?

I don't test in Bay Area suburbia or anything similar because that type of FSD driving is already well covered. I test on a mountain road where one small mistake can lead to death. Cliffs or near cliffs on both uphill and downhill sides of the road, narrow and twisty with inconsistent corners and cambers, grades abruptly changing, lane markings that most often don't exist at all, no centerline or shoulders and almost every corner is 100% blind with no good clue as to what is coming next. Trees and rocks block vision everywhere. Because the road was built in the 1920's, the corners are dictated by the terrain and drainage needs which is why there are decreasing radius and off-camber corners, inconsistent grade changes etc. There are also no guardrails at all, the road edges are often a direct route down the mountain. In short, a nightmare for any system with pretenses of being autonomous. Humans tend to drive it either like Mario Andretti (because they know the road) or a complete beginner peeing their pants.

For the first 3 years I would typically end my testing at the beginning of the steep climb, the last 9 miles to the summit because it would fail multiple times, every time, and often in ways that were very deadly if intervention was not immediate. One of its favorite failure modes was to think the old-growth forest backdrop of a hairpin corner was a long straight-away. I suspect it was aiming for a large trunk as it would try to accelerate off the road at high speed into large trees, I can tell you, that would not end well. I would slam on the brake at the last possible moment and come to a stop at the outside shoulder of the hairpin turn.

Still, every few months I would let it have a go at the last section of road when traffic was non-existent, but it couldn't consistently and properly position itself to enter blind corners. I would stay hyper-alert for the first sign of on-coming traffic while letting it do its thing. Even letting it have its dangerous way, it could not hope to complete this section of road without multiple human interventions to avoid certain disaster. It was completely hopeless, and the most discouraging part was that it seemed to have no idea how to position itself to safely enter narrow and blind corners and it didn't help that there are no lane markings whatsoever on many of these corners. Just blacktop patches or chip seal blurring invisibly into road sand, forest duff, trees and rocks.

About 4 weeks ago, after much of the winter snow had melted, I gave it another go. I didn't have much hope other than my M3P had a new verion of FSD. To my surprise and delight, it drove the entire road without a single intervention, and, more importantly, it had learned how to set itself up to safely enter blind corners! This was a huge step change of the highest order! It was not a gradual improvement.

The entire 9 miles the car drove very much like a human, save for one 300-degree corner near the top where it briefly hesitated by slowing to 15 mph instead of driving smoothly around at 20 mph. In all other cases, the car slowed appropriately for each corner and accelerated smoothly and confidently out of each one (unless there was another sharp corner immediately in which case it just continued around, as a human would).

In the last couple of miles there was a wide pile of snow avalanche debris comprised of heavy, wet, melted snowpack completely covering the downhill lane. It was shaped like a tongue and was about 20 feet wide and 4 feet deep on the uphill edge of the road and tapering to zero exactly at the centerline, which is marked on this section of road. This would cause considerable damage to any car that drove into it at 30 mph. On the uphill leg, FSD simply drove by the avalanche debris in the downhill lane at 30 mph as if it didn't exist. On the downhill leg I was concerned because there was a blind bend in the road only a little more than 100 feet beyond the avalanche debris. If FSD exhibited uncertainty and poked along at slow speed in the on-coming lane, it would be subject to a head on collision if another motorist coming around the blind corner didn't react quickly enough.

Descending the mountain, I felt like we were approaching the wall of snow awfully quickly at 30 mph and FSD was showing no signs that it recognized the threat. I did not look at the center display because I had my full attention on "driving" and was preparing to take over because the uphill lane had no shoulder and no guardrail and any car venturing off the narrow pavement would accelerate as they bounded hundreds of feet down a 60-degree slope. Since my attention was outside the vehicle. I don't know when or if the center display pictured the avalanche debris, but the car smoothly changed into the on-coming lane and went immediately back into the downhill lane, as smoothly as the most professional driver, before gently applying the brakes (regen) and taking the corner at 25 mph. As if it were nothing at all. I could not have done it smoother or more appropriately myself! Going slower would have been more dangerous because it could have left us with nowhere to go in the event a fast-moving vehicle had rounded the blind corner at the inopportune moment.

The entire trip, up and down the mountain, positioning itself appropriately for the entrance to blind corners, slowing down and speeding up smoothly, very similar to the way a good human driver would react, avoiding the avalanche debris more safely than your average human would have done, all without intervention, I would call that a real step change in FSD. Remember, the FSD team does not work on all scenarios and driving conditions simultaneously and just because you, personally, are not driving in scenarios that have undergone massive transformation, that doesn't mean huge improvements have not been made. Just last year, FSD would try to kill me every time by accelerating off into old-growth trees or entering blind corners too wide.

The primary reason FSD is such a difficult problem is due to the huge variety of driving scenarios that must be solved for. I'm super impressed with the recent improvements on narrow twisty roads, often lacking any pavement marking whatsoever. People who think FSD is nothing more than a glorified "stay between the lines" are in for a surprising awakening!

I agree. I’ve put about 10,000 miles on my car since I got FSD last November and most of it has been with FSD enabled. I’ve covered 7 western states while avoiding interstates where possible, and lots of small towns. FSD has definitely gotten progressively better, with better speed control for curvature and visibility, smoother transitions between maneuvers, and better recognition of road geometry.

Here’s me going up and down Pikes Peak a couple of weeks ago (no safety-related interventions):

 
I agree. I’ve put about 10,000 miles on my car since I got FSD last November and most of it has been with FSD enabled. I’ve covered 7 western states while avoiding interstates where possible, and lots of small towns. FSD has definitely gotten progressively better, with better speed control for curvature and visibility, smoother transitions between maneuvers, and better recognition of road geometry.

Here’s me going up and down Pikes Peak a couple of weeks ago (no safety-related interventions):

Maybe Randy could have used FSD when the windows started fogging? 🤷‍♂️

That is a fun road on two wheels or four.
 
What doesn't make sense is the number of EVs on the road vs the quoted rates. it implies just as you said only about 208k EVs which is ridiculously low.

To find the rate of car fires by vehicle type, we collected the latest data on car fires from the NTSB and calculated the rate of fires from sales data from the BTS

Which either means they did something totally stupid like compare total fires in a year (of all cars on the road no matter the age) to sales in a year (of only new cars) which would be mixing units or they did something crazy complicated and reviewed all the fires to find which ones applied to one years cars in a precise fashion.

Occams razor suggests they did the stupid option.
 
There is no doubt, FSD beta v10.12.2 is now routinely executing complex routes w/o intervention, and demonstrating very safe decision making:
There is sure a lot of doubt in my mind. I got the FSD about 3 weeks ago, and I can scarcely drive a mile without an intervention.

Some of the problems may be local: most of the lane lines are still gone after winter, and I am probably the first FSD Tesla to (ever) drive down most of my routes. Many are situations that would annoy the drivers near me if I let them go on, and you can't say FSD is ready until it stops annoying so many others. Some are failures to recognize that a lane is closed for repairs, and would result in a crash if I didn't do a hard stop. Most 90 degree turns are jerky. If I took turns and curves as fast as FSD does, I would never have passed my safety monitoring. Navigation will get me there, yes, but the car goes into lanes that terminate too often. The car waits on railway tracks, duh.

So on the whole, trying FSD myself has given me a perspective on the reports from this forum, and I think on the whole youse guys are too sanguine.
 
There is sure a lot of doubt in my mind. I got the FSD about 3 weeks ago, and I can scarcely drive a mile without an intervention.

Some of the problems may be local: most of the lane lines are still gone after winter, and I am probably the first FSD Tesla to (ever) drive down most of my routes. Many are situations that would annoy the drivers near me if I let them go on, and you can't say FSD is ready until it stops annoying so many others. Some are failures to recognize that a lane is closed for repairs, and would result in a crash if I didn't do a hard stop. Most 90 degree turns are jerky. If I took turns and curves as fast as FSD does, I would never have passed my safety monitoring. Navigation will get me there, yes, but the car goes into lanes that terminate too often. The car waits on railway tracks, duh.

So on the whole, trying FSD myself has given me a perspective on the reports from this forum, and I think on the whole youse guys are too sanguine.
All I need is for FSD to be consistent. Like today, went on a 12 mile drive with 2 unprotected lefts across multiple lanes, the car did amazingly well and I had zero interventions getting to the restaurant. Going back home was a mess, it wouldn't turn when it's suppose to, wouldn't get into turning lane until the last minute which of course there's no possible way, and it almost ran a red light. My experience was night and day, it's almost like I had version 10.12 to the restaurant and version 1.2 on the way back.
 
Texas Senate Bill 8 went into effect on September 1st 2021 and yet somehow Tesla has not reported any major difficulty hiring and retaining employees for Giga Austin in the 9 months since then.

I’ll leave my personal opinions on this topic out because they don’t belong here. I just want this post to be factual and investment related and I’m tired of the arguing.

I’m just saying…wild claims that this is going to massively affect Giga Texas are radically overblown, and any employees or potential employees who feel that strongly about it probably weren’t going to accept a job in Texas anyway, meaning that recent events likely will have little net impact beyond what SB 8 already has. As others have already pointed out, in the rare circumstance where employees or their children/spouses/partners decide to access procedures in e.g. Colorado, they can do that on Tesla’s expense for the whole thing. It’s an inconvenient potential hassle for a couple of days but that’s about it. Making a major life and career decision based on wanting to avoid that is not something I expect many people to do, especially not among the cream of the crop some here are so worried we’re losing, because such people would know how to apply basic logic and prioritization to their decisions.

Additionally, consider that factories are multi-decade investments. I invite you to look up long-term demographic and election result trends in the state of Texas and the amount of donations the oil and gas industry makes to the current government leaders in Texas, and then assess the likelihood that the current policies in general will still be in place 10, 20 or 40 years from now. For comparison, the Fremont facility was built in 1962, during which time California was a reliable Republican stronghold. Ronald Reagan was the governor of California from 1967 to 1975. Since around 1992, it has been the opposite. Things change. Even the parties themselves change, including three major realignments in the history of the USA. The rapid growth of Austin and Houston and soon-to-be collapsing profits of the oil & gas industry point towards another one in the next 10-20 years as far as I can tell, and I have no idea what things would look like from then on.

Calm down, go for a walk, enjoy some 420 if you do that, and think some more before posting angry things everyone’s going to have to read.

I too am pretty disappointed we lost jbcarioca. I would like to let you know that if anything would make me want to stop posting here it would be the apparent inability of many members—who I would rather not put on ignore because of their good posts—to exercise the self-discipline to put their personal political comments in the politics thread. It is rather obvious what happens when those comments enter this one: the thread spirals out of control and nobody learns anything or changes their minds.
 
Last edited: