FSD Beta Videos (and questions for FSD Beta drivers)

[S4WRXTTCS]

Uh, I'm not buying this idea that you have to drive at a crawl. I am now at 3,000 mile with a 99 score. Speed has nothing to do with the score.
And I am not even trying to get a high score. Just driving the way I always do. And on interstate 80, it is fast.

I have a 99 score as well, but I had to adjust my driving.

The Safety Score is both terribly thought out and terribly executed.

You're 100% correct in that speed doesn't have anything to do with the score, and you could drive 20-30mph over the speed limit with no impact to your score at all.

You can cut someone off and not get dinged., but if someone cuts you off you get double dinged.
You can run red lights and not get dinged, but if you stop for a yellow you get dinged for hard braking even if it wasn't really all that hard (6mph per second decrease isn't a whole lot).

If someone in the lane coming towards you decides to turn left in front of you with what is normally a mid nuisance distance you have to decide whether to brake to chance hard braking or not to brake that hard to chance a FCW.

The biggest problem with the Safety Score is it DOES change a persons driving. Things like doing braking sessions to reduce hard braking or going round and round to reduce hard cornering. Or they're resetting the computer while on the road to erase a FCW.

Elon said Tesla would always make fast cars, but now they're punishing anyone who actually drives a Tesla the way a Tesla was intended to be driven.

 

[qdeathstar]

I don't really have much of a point, other than to say that the document you posted is lacking in a lot of details. It's difficult to make conclusions based on it.

You're also reading the document incorrectly. You said that most fatal accidents occur at sub 35mph speed limits. But that's not what the graph shows at all. The graph only shows the proportion of fatal accidents (urban vs rural) within a speed limit range. All the bars within a speed limit range add to 100%! Please reassess!

View attachment 723060

I read the graph incorrectly

 

[timdoes]

 

[qdeathstar]

Out of curiously what is the difference between a disengagement and intervention?

 

[Phlier]

Out of curiously what is the difference between a disengagement and intervention?

I've used the term "disengagement" to refer to turning off FSD/AP by choice, whereas "intervention" means I had to turn it off to keep FSD/AP from doing something stupid. Just how I use 'em.

 

[powertoold]

I've used the term "disengagement" to refer to turning off FSD/AP by choice, whereas "intervention" means I had to turn it off to keep FSD/AP from doing something stupid. Just how I use 'em.

Nope, disengagement = fsd is turned off, intervention = accelerator, signalling, adjusting speed, etc. but fsd stays on

 

[aronth5]

I get the sense after reading this thread that few if any of the new FSD beta testers see a Robotaxis coming anytime soon to a Tesla near you. Certainly not in 2022. Anyone fell differently based strictly on FSD's technical capability to perform drive after drive with zero engagements?

I'm just glad I would never use my car as a robotaxi.

 

[EVNow]

Out of curiously what is the difference between a disengagement and intervention?

Disengagement is an extreme intervention.

 

[powertoold]

I get the sense after reading this thread that few if any of the new FSD beta testers see a Robotaxis coming anytime soon to a Tesla near you. Certainly not in 2022. Anyone fell differently based strictly on FSD's technical capability to perform drive after drive with zero engagements?

I'm just glad I would never use my car as a robotaxi.

I'm pretty confident FSD will be reliable and safe enough for geofenced robotaxis (easy routes, no construction, etc.) within the next 6 months. Whether or not Tesla will deploy them is another story.

 

[EVNow]

Out of curiously what is the difference between a disengagement and intervention?

Nope, disengagement = fsd is turned off, intervention = accelerator, signalling, adjusting speed, etc. but fsd stays on

To me disengagement remains an Intervention - though an extreme one.

 

[powertoold]

To me disengagement remains an Intervention - though an extreme one

From all the videos so far, disengagement is any time fsd is turned off (blue wheel graphic is off)

Intervention = wheel graphic stays blue

100% sure this is the definition. Not all disengagements are necessary. Some disengagements are simply to reroute.

 

[EVNow]

Intervention = wheel graphic stays blue

Isn't that how you are defining it i.e. that is not official Tesla definition ? To me anytime you do anything to change what FSD is doing, its an intervention. So, the extreme intervention is to turn it off.

 

[powertoold]

Isn't that how you are defining it i.e. that is not official Tesla definition ? To me anytime you do anything to change what FSD is doing, its an intervention. So, the extreme intervention is to turn it off.

Dunno if Tesla has an official definition, but all the prior beta testers were / are on the same page about disengagement vs intervention.

To your post, my analogy is a light bulb dimmer. Intervention = dimming the bulb, disengagement = turn bulb off. Most dimmers don't allow you to turn the bulb off by dimming all the way, lol.

 

[Terminator857]

I'm pretty confident FSD will be reliable and safe enough for geofenced robotaxis (easy routes, no construction, etc.) within the next 6 months. Whether or not Tesla will deploy them is another story.

How do you determine reliable and safe enough? Geofenced: in a senior community where max speed is 25 mph? When will it handle potholes?

 

[nvx1977]

I've been using disengagement to mean I did something and FSD disengaged. I've been using intervention to mean accelerator nudges. It's unclear whether other interventions, like adjusting speed with the scroll wheel counts as an intervention, meaning it might be something that FSD logs and sends back to the mothership. Same with manually signaling when FSD doesn't.

I doubt either term is "official", but it's become accepted vocabulary from pretty much all the original early access testers. Some people even have charts to track interventions vs disengagements.

 

[Daniel in SD]

I'm pretty confident FSD will be reliable and safe enough for geofenced robotaxis (easy routes, no construction, etc.) within the next 6 months. Whether or not Tesla will deploy them is another story.

I need to bookmark this one. haha.
How exactly will you determine that it's safe enough? Is this claim falsifiable?

 

[PhantomX]

I need to bookmark this one. haha.
How exactly will you determine that it's safe enough? Is this claim falsifiable?

Who is liable for any incidents incurred during robotaxi operations?

 

[Tam]

Who is liable for any incidents incurred during robotaxi operations?

From L2 and below, a driver is.

From L3 and above, technologically it's the machine while the system’s active. Legally, it's not defined in the US but some countries like Japan and the European Union, it's the car manufacturers.

 

[FloridaJohn]

I'm pretty confident FSD will be reliable and safe enough for geofenced robotaxis (easy routes, no construction, etc.) within the next 6 months. Whether or not Tesla will deploy them is another story.

You don't have a very good track record with these predictions. Remember this one?

Tesla.com - "Transitioning to Tesla Vision"

What is up with high beam nonsense talks? I use auto high beam on M3 all the time as well as my Model S. Never had any problems with it. If you overstress high beam use that is on you. If you get a ticket using auto HB then sue Tesla.

teslamotorsclub.com

From that thread, "I’d be surprised if V9 doesn’t match or exceed Waymo performance on the same routes."

I assume you are surprised.

Then there is this one, in which you predicted that by July 2021 interventions would be down to one every 150k miles. I think you mentioned a couple days ago that you are intervening once every mile with the current Beta version.

FSD rewrite will go out on Oct 20 to limited beta

Probably somewhere around an accident probability of every 100-200k miles. This would be a ridiculous achievement. I don't know what Tesla's internal goal is. That is three years away.

teslamotorsclub.com

(links go to the exact post even though the forum software is quoting the first post in the thread)

 

[DanCar]

You don't have a very good track record with these predictions. ...

Perhaps a result of his mind being blown with every release.