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

Autonomous Car Progress

This site may earn commission on affiliate links.
I guess you are ignoring all the current complaints about driving Tesla pure vision, from hitting the garbage can to the stricter nanny system.
You are ignoring the fact that radar has never prevented the car from hitting static objects.

If you think somehow Tesla should have used ugly Lidar in 2016 in the cars - I can assure you exactly ZERO cars would have been sold and Tesla would be bankrupt by now.

What you don't seem to understand is that these hit pieces in NYT and others that question Tesla's pure vision idea are basically propaganda pieces by competitors.
 
  • Like
Reactions: mark95476
You are ignoring the fact that radar has never prevented the car from hitting static objects.
I am not. That is why sensor fusion, including Lidar is essential.
If you think somehow Tesla should have used ugly Lidar in 2016 in the cars - I can assure you exactly ZERO cars would have been sold and Tesla would be bankrupt by now.
So is it a choice between ugliness but alive or beauty but dead?

So is it a choice between wealthy but unsafe and bankrupt but safe?

What you don't seem to understand is that these hit pieces in NYT and others that question Tesla's pure vision idea are basically propaganda pieces by competitors.

Tesla could have disclosed all of this information instead of letting others do that for them.

It's insightful, informative, and factual. For this thread, it means Tesla still has a long way to go in terms of progress. And that is news.
 
Last edited:
It's a very good thing Elon has been running Tesla and not you.

I am not. That is why sensor fusion, including Lidar is essential.

So is it a choice between ugliness but alive or beauty but dead?

So is it a choice between wealthy but unsafe and bankrupt and safe?



Tesla could have disclosed all of this information instead of letting others do that for them.

It's insightful, informative, and factual. For this thread, it means Tesla still has a long way to go in terms of progress. And that is news.
 
So is it a choice between ugliness but alive or beauty but dead?
Who is dead ? Yes, 800k+ people because of Covid. Would you support a China like lockdown ?

BTW, you are speculating that vision doesn't work - purely based on what Tesla's rivals - who have a vested interest in portraying Tesla's vision as unsound - are saying. I guess you carry a lidar when you drive around too ?

So is it a choice between wealthy but unsafe and bankrupt but safe?
See above.

It's insightful, informative, and factual. For this thread, it means Tesla still has a long way to go in terms of progress. And that is news.
I'm sorry you can't distinguish between BS propaganda and news. Its actually easy to tell.

Writing about "recall" of software for 12k cars but not that an updated software went out in a day is a dead give-away.
 
...Writing about "recall" of software for 12k cars but not that an updated software went out in a day is a dead give-away...

So, your understanding is: Each time there's a recall, there needs to be a report of some good coming out of it or otherwise it's bad?

Prior to 8GB eMMC from the MCU recall, owners have been demanding from NHTSA to declare it as a "recall".

Apparently "recall" is something good to deal with something that's already bad. Thus, when there's a recall, it's automatically assumed that the defect is fixed.
 
So, your understanding is: Each time there's a recall, there needs to be a report of some good coming out of it or otherwise it's bad?

Prior to 8GB eMMC from the MCU recall, owners have been demanding from NHTSA to declare it as a "recall".

Apparently "recall" is something good to deal with something that's already bad. Thus, when there's a recall, it's automatically assumed that the defect is fixed.

Apparently according to you this is how to deal with issues.

What NHTSA is used to is
- People start reporting issues
- After months, NHTSA sends a letter to OEM
- OEM gives vague answers to NHTSA, after a month
- More people report issues to NHTSA
- OEM settles with some people in secret
- More people report issues to NHTSA
- NHTSA starts talking to OEM in earnest saying the issue can no longer be brushed under the carpet
- OEM and NHTSA negotiate a fix and recall schedule over next several months
- OEM sends letter to customers about the recall, asking them to expect a letter about when the customer can get the fix
- After several months, OEMs send a letter to the customers that they can schedule a fix at the dealer
- Customers call dealers and are given a date that is weeks or months away
- Customer forgets about the fix

NHTSA is just angry that Tesla is not following this process.

ps : This is getting quite OT here.
 
I use AP/FSDbeta every time I drive in Tesla-central (Silicon Valley). That's been about 18 months now.

“In the 2nd quarter, we recorded one crash for every 4.41 million miles driven in which drivers were using Autopilot technology (Autosteer and active safety features). For drivers who were not using Autopilot technology (no Autosteer and active safety features), we recorded one crash for every 1.2 million miles driven. By comparison, NHTSA’s most recent data shows that in the United States there is an automobile crash every 484,000 miles.”


I am a consumer, not an engineer. I rely on the information presented or hidden to be an informed consumer.
 
...one crash for every 4.41 million miles driven...

Companies with LIDAR have figured out how to avoid crashes for years like Google/Waymo since 2009.

Avoiding crashes is an easy and successful task for them but they are still lacking intelligence.

Waymo did not collide with a traffic cone as avoiding crashes is an easy task but it didn't know how to deal with the traffic cone that was out of place in its pre-mapping memory.

In the meantime, Tesla is still struggling with simple basic crash avoidance tasks such as curves of death:

 
I admit that you are correct about their non-sales.

Since the intelligence part is so difficult, Waymo is now partnering with trucking companies to take advantage of what machines like: repeatability, the unchanging environment of pre-set routes, and depots for trucks in L4.
Scaling and mass production are hard.

As the old joke goes, Soviet Union had no problem sending rockets but had trouble making good shoes.

ps :
Its like deja vu all over again. These two are different dimensions. Nobody is yet to crack both - only one or the other. Cracking one doesn't guarantee the other and you can't compare one dimension with the other. I've been saying this for a looong time.
- Features (i.e. how close someone is to robotaxi)
- Geographical scaling

pps :

Whether a car freezes when it rains or not !

The reason for endless, unresolvable debates is that Waymo and Tesla use very different approaches and are thus good and bad in different ways. We can't objectively compare them - because they are better/worse in different dimensions.

Actually ...

... which FSD solves
- most problems (scenarios)
- most reliable
- for most people
- in most locations
?

We'll see that no one system is the best in all the four dimensions.

We actually have 3 dimensions
- Scenario (or feature)
- Location
- Individual (i.e. people)

Reliability is not a dimension - its a "measure". For a given combination of the above 3 dimensions we get a particular reliability (or error probability) number. Ofcourse the reliability can be calculated for a group of individuals/scenarios/locations or any combinations of them.

Waymo has high reliability for a lot of features - but only in small # of locations and only for their own cars.

ps : Elevators work in a lot of locations but in extremely limited number of scenarios :D

From Chandler to ... another arid southern suburb ?

Currently they can't even handle different weather conditions - let alone different climates, ways people drive etc. They have not even attempted anything outside US. They have a VERY LONG way before you can claim they have "solved FSD".

They have not demonstrated ability to scale on any of the 3 dimensions we discussed earlier
- Features
- Locations
- # of cars

Only thing we can say is they are doing better on features than Tesla and Tesla is definitely doing better than Waymo on Locations & # of cars.

Insane Bus Ride in The Himalayas!-Getting To The Mountains Is Exciting As Climbing Them - YouTube
 
Last edited:
Not sure if it has been posted here or in other threads, but I'm seeing news today that Mercedes gained approval for a Level 3 system in Germany


Watching some footage of what the Mercedes EQS can do, that's functionality I would pay for
 
It's only approved for slow highway speed such as stop-and-go traffic. Tesla has done a very good job in that condition although it's not L3 due to its nanny system.
Yup I think they expect more to come, both in improvements to the system via OTA updates and regulatory approvals

Being L3 in any capacity is entirely the point. I personally wouldn't pay for a system that still requires hands on the wheel and eyes forward, but I will pay for a system that allows me to stop paying attention for long highway hauls
 
What do we call this? Being salacious or factitious? Or maybe just weak FUD?

Tam's "curves of death" = staying in lane, but very uncomfortable per the new user

"When using auto-steer it enters tight curves at full speed, it manages to stay in lane but makes a very uncomfortable ride."

In the meantime, Tesla is still struggling with simple basic crash avoidance tasks such as curves of death:

 
I think Tesla has the scaling and mass production problem in-hand. They've been dog-fooding their AP/FSD and have a super quick SW update cycle.

Scaling and mass production are hard.

As the old joke goes, Soviet Union had no problem sending rockets but had trouble making good shoes.

ps :
Its like deja vu all over again. These two are different dimensions. Nobody is yet to crack both - only one or the other. Cracking one doesn't guarantee the other and you can't compare one dimension with the other. I've been saying this for a looong time.
- Features (i.e. how close someone is to robotaxi)
- Geographical scaling

pps :
 
What do we call this? Being salacious or factitious? Or maybe just weak FUD?

Tam's "curves of death" = staying in lane, but very uncomfortable per the new user

"When using auto-steer it enters tight curves at full speed, it manages to stay in lane but makes a very uncomfortable ride."
Curves of death = curve where entering at the speed limit is entering too fast.
If stuff is flying around inside the car you're pushing it, and any sudden loss of traction could cause you to exit the road.

Responsible humans recognize the difficult conditions and slow down. They don't need sharp bend signs.