Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
In that case, vision is better as Tesla AEB has already been proven to be superior than any other system out there by the Euro NCAP

www.euroncap.com


That's a fail compilation, so it's going to be biased, but still some of the fails are surprising. The EQS and Lucid are loaded to the gills with sensors. Goes to show it's not about the sensors, but the software.
 
The EQS and Lucid are loaded to the gills with sensors. Goes to show it's not about the sensors, but the software.
That is what I have been stating for the last 3 years. It is not lack of data but the decisions that are still being developed. Even the 5mp HW3 cameras are capturing more data than we know what to do with it. Adding more sensors and higher pixels is not going to get us there any faster.
 
Those 6 lane changes in 1 minute. That has never happened to me, even on V10.2,
Sure, that was a little spastic.

They’ll just dial that down and ship it though. Just need to get back to missing turns and they’ll be fine.

This looks like fire. Guess I’ll have to take a close look at videos and see whether the 5x reduction is a lock.


It matters. If you believe it doesn’t matter, then you are not qualified to make any statements on that matter.
Lol, ok. Do explain now that I am disqualified. I fail to see how it matters whether a slow reaction is due to hardware or software. We’re not talking about fixing it!
 
While waiting for 12.4.1... I would like to try using an app on my iPhone to keep track of hard braking events. Something that uses the accelerometer and logs data with metadata (time, location etc) to refer to as I use FSD. Anyone have a recommendation? Thanks in advance.
 
In that case, vision is better as Tesla AEB has already been proven to be superior than any other system out there by the Euro NCAP

www.euroncap.com


AEB performance by most manufacturers is abysmally poor, no question. This has very little bearing on reaction time though, which is what we were trying to figure out. A slow-ish one-second reaction time is probably plenty good enough for these specific tests as shown; what they're testing is whether the cars properly react, not how quickly they react. Most of the cars clearly see the danger (and sound warning tones) well in advance, but fail to brake. So it's a control issue, not a sensor issue. A confounding factor is that AEB is quite different from autonomous driving; some AEB systems are deliberately designed not to completely avoid a collision, but merely to reduce the impact speed. But an autonomous system should be designed to completely avoid the collision, of course.

Ultimately, the question is how quickly the system can gain certainty that an unusual response (in this case emergency braking) is the correct course of action. With pure vision in a difficult environment (low light, rain, etc.), even if the occupancy network pops up a potential obstacle in 10ms, it may be difficult to distinguish a false-positive (phantom braking) from a true-positive (AEB required), so more processing and latency will be required to gain the necessary amount of certainty. This is where e.g. Lidar can be hugely helpful; in principle it can provide that certainty much faster than pure vision.
 
  • Like
Reactions: JB47394
That's a fail compilation, so it's going to be biased, but still some of the fails are surprising. The EQS and Lucid are loaded to the gills with sensors. Goes to show it's not about the sensors, but the software.
Lucid has a ton of sensors, but not all of them are activated. I was also surprised to learn (at a showroom recently) that most of their fleet actually doesn't have Lidar installed. I would have expected they would all have Lidar just to gather fleet data like Tesla does, but apparently not. Only cars with the highest-end DreamDrive package are getting physical Lidar, at least for now.

OTOH, once L4 autonomy is "solved", these early Lucids (back to 2021) should be able to support it, due to their extensive hardware suite. Whereas Teslas with HW3 or HW4 may never be able to, if the ultimate solution does turn out to require Lidar/radar. So it's about both sensors and software. EQS/Lucid currently have better sensors, Tesla has better software. But software is much more easily upgradeable/retrofittable than hardware.
 
  • Like
Reactions: edseloh
A study showed that Amazon customers were happiest in the period between the purchase and delivery. That is, while awaiting the arrival of their product.

I bet the same is true of Tesla owners awaiting the arrival of the next update.
Not me, not happy until I have 12.4, then after I get that won’t be happy until I get 12.5
 
I heard that if it’s been a long time since your vehicle has checked for an update it’s because Tesla is holding back checking your vehicle because they are sending you an update soon so there’s no need for your vehicle to check for an update before then

So the longer it’s been since your vehicle has checked for an update the sooner you will be getting the update GL all

Me and fasteddie have the longest time since last check up I believe
 
With pure vision in a difficult environment (low light, rain, etc.), even if the occupancy network pops up a potential obstacle in 10ms, it may be difficult to distinguish a false-positive (phantom braking) from a true-positive (AEB required),
That is why FSD speed is slowed down in bad vision weather just like how humans do.

That being said, even a car with LiDAR and RADAR will have to slow down in bad weather, not because it cannot detect objects but because it still needs to use vision to identify the road, the lanes, the signs everything else that is made for human vision.
 
IMG_6080.jpeg


What’s everyone checking to get? 14.9?
 
Try the service request for autopilot check. Didn't work for me, but may for you
Mine shows the software is up to date as of 8 May, doesn't change after reboot. This is strange so I tried the service request route and was told all is OK, I have the latest version. Funny how the date doesn't update as it always used to though.... 🤔

Edit: "Me and fasteddie have the longest time since last check up I believe" NOPE! I seem to hold the current record.... :cool:
 
Last edited: