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

FSD Beta Videos (and questions for FSD Beta drivers)

This site may earn commission on affiliate links.
Tesla went from concept Model 3 in 2016 to early production in 2017.
Do you seriously believe Tesla fully engineered the model 3 in 18 months?
Both the X and 3 are built on existing platforms, with massive re-use. You think a legacy automaker can't build a "new" car on an existing platform in under 5 years?

Meanwhile whole new platforms from Tesla such as the original Roadster, S, Roadster 2, Cybertruck, Semi.... How many years did those take?
Do you even want to buy a car that was engineered in less than 5 years? Testing and tooling takes time to do right.
 
  • Disagree
Reactions: EVNow
I do too .... but they will have their hand full trying to stay alive.

Besides as a practical matter, Tesla will not be afraid of very low volume manufacturers, even if they miraculously produce "L4" in 3 to 5 years - except may be for image.

ps : I'd think Lucid is probably idea to have a tie-up with MobilEye for city driving. Rivian probably won't because Amazon has someone AV company ?

AFAIK, Lucid was going to use Mobileye but then ditched them to do their ADAS in-house. Personally, I think that was a mistake.

But I guess the question is who is Mobileye's secret "non-European automaker" who has agreed to deploy SuperVision? We do know Mobileye plans to deploy SuperVision on 9 car models from 6 brands by 2026. Mobileye expects SuperVision to be on 1.2M cars by 2026. Mobileye also has a deal in writing with a major brand to start production on L4 Mobileye Chauffeur in 2026. Obviously, plans can change or be delayed. I am just saying we should not get complacent. L4 will come to consumers cars.

But this is a discussion probably better suited for the Autonomous Driving Progress thread.
 
Do you seriously believe Tesla fully engineered the model 3 in 18 months?
Do you think Model 3 took 5 years to design & produce ?

From wiki ...

According to Tesla's CTO, JB Straubel, in October 2015, most Tesla engineers were working on the 3 rather than S or X.[92][93] Since electric cars have lower cooling needs than combustion cars, the Model 3 does not have nor need a front grille.[94] Musk intended for the final design to be released on June 30, 2016[95] but when the design was finished on July 27, it was not publicly released.​
They probably took 20 to 30 months.
 
AFAIK, Lucid was going to use Mobileye but then ditched them to do their ADAS in-house. Personally, I think that was a mistake.
Considering most US manufacturers are ditching MobilEye - I think they have a GTM problem as I was writing earlier. I find it strange that an Intel sub finds it difficult to get an agreement with an US OEM (let alone a major one).

But I guess the question is who is Mobileye's secret "non-European automaker" who has agreed to deploy SuperVision? If it is a high volume manufacturer, then it could be a threat to Tesla. But this is a discussion probably better suited for the Autonomous Driving Progress thread.
Yes - we should move the discussion. Could be an Asian one ... ?
 
They probably took 20 to 30 months.
In other words, not 18 months, and the idea that nobody was working on it until Oct 2015 is not supported. What JB said is "most engineers were working on it by Oct 2015"- that doesn't mean a smaller group wasn't working on it from some much earlier date.

But we're here laughing that "legacy" automakers take so long and Tesla is so fast, yet the Cybertruck and roadster aren't here, and the semi tool 5 years from announcement to shipping.

Chevrolet announced the C8 Corvette in April 2019 and began shipping it in February 2020.

Just not a lot of evidence here that "legacy" automakers are super slow while Tesla can reliably crush out a car in half the time. Especially in the last few years, given it's been 3 years since Tesla even began shipping a new model.
 
Chevrolet announced the C8 Corvette in April 2019 and began shipping it in February 2020.

Just not a lot of evidence here that "legacy" automakers are super slow while Tesla can reliably crush out a car in half the time. Especially in the last few years, given it's been 3 years since Tesla even began shipping a new model.
Announcement is not the same as beginning of design.

There is a LOT of evidence on how much time it takes for legacy OEMs. You just need to do a bit of search. There are tons of people who work/worked for legacy OEMs that have talked about it.

ps : Quick search ...

 
And also because a customer is never required to see it during purchase-- unlike the purchase page.




A customer literally has to see and sign the MVPA-- so yes.

You're trying to compare apples and guitars here my dude.
So are you saying that if a customer purchases FSD based on research of the autopilot/FSD features from Tesla’s OWN website, they don’t have a case? Imo post 3/19 have the same case. “No action required of driver” is STILL there on Tesla’s website.
 
Announcement is not the same as beginning of design.

There is a LOT of evidence on how much time it takes for legacy OEMs. You just need to do a bit of search. There are tons of people who work/worked for legacy OEMs that have talked about it.

ps : Quick search ...
I don't even understand what you are arguing. The model 3 was part of Tesla roadmap 16 years ago. Everyone takes 5 plus years to plan, design and manufacture, Tesla Semi and Roadster 2 are 7 years in the making, Cybertruck is 4 years in the making. The difference is legacy automakers have concurrent development cycles and release refreshes every year. Just because there is a short time between announcement and release does not indicate how long it has been in development. Model 3 certainly did not take 2 years to design and manufacture.
 
So are you saying that if a customer purchases FSD based on research of the autopilot/FSD features from Tesla’s OWN website, they don’t have a case?

They certainly have a vastly less strong one.

Since Tesla can show the only thing we can know, for sure, that you were shown during the actual purchase as a description of what you were buying.

And since 3/19 that list is entirely stuff that would be considered fully delivered to you once you have FSDb in wide release at L2.

(also the current post 3/19 language on the non-sales page has a disclaimer about "The future use of these features without supervision is dependent on achieving reliability far in excess of human drivers" in addition to regulators- which is a significantly stronger disclaimer than the one the pre 3/19 sales page had which only said it was dependent on "extensive software validation" plus regulators)


If you think the language change does NOT provide Tesla any legal cover at all, what possible reason could you give for why they made it?
 
Do you seriously believe Tesla fully engineered the model 3 in 18 months?
No, that's why I mentioned that was the concept date and it's unclear the exact date of when development started. But it would still seem to be well under 5 years.
Both the X and 3 are built on existing platforms, with massive re-use. You think a legacy automaker can't build a "new" car on an existing platform in under 5 years?
Legacy automakers reuse existing platforms in whole or in part almost all the time (even for "new" models). I'm just saying there are clear cases of Tesla developing a car under 5 years.
Meanwhile whole new platforms from Tesla such as the original Roadster, S, Roadster 2, Cybertruck, Semi.... How many years did those take?
Do you even want to buy a car that was engineered in less than 5 years? Testing and tooling takes time to do right.
Yes, those took longer, but just saying there are examples of ones that took less than 5 years.
 
I’ve been thinking about making some new ones and posting them here, but you probably wouldn’t like it.

Maybe soon. The number of safety disengagements would be insane, with the number of potholes around here right now. (No question that they qualify as safety DEs of course!)
No need. Until the potholes are repaired, using the system is unsafe for you, so best to drive manually for now.
 
I’ve been thinking about making some new ones and posting them here, but you probably wouldn’t like it.

Maybe soon. The number of safety disengagements would be insane, with the number of potholes around here right now. (No question that they qualify as safety DEs of course!)
I would like a more scientific analysis of the jerk issue. I need graphs!