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.
Have you tried auto shift? Any possibility the user can make mistake and cause accident?
OT: Had Mobile Service just now and the tech showed up in a Plaid S.🤔
IMG_5066.jpeg
 
There is your answer. You can't let it laps AT ALL if you want to stay early access. Tesla probably uses this as a measuring stick as to how interested and commented to testing you are. Also maybe a little reward for paying the price too. So you will now heave to wait until 12.4.x is released as the standard FSDS.
My wife's car, bought new in Dec 2022 and which had subscribed to FSD 2 separate times within 2023 and recently had the free trial which ended, is still on 2024.3.25. So it seems like some non-early access cars are also being held back for 12.4.

edit: I thought about it some more, it could be that my car's status is also affecting my wife's car's status. Although in the past, my wife's car has usually been far ahead of my car on the main software branch.
 
My wife's car, bought new in Dec 2022 and which had subscribed to FSD 2 separate times within 2023 and recently had the free trial which ended, is still on 2024.3.25. So it seems like some non-early access cars are also being held back for 12.4.

edit: I thought about it some more, it could be that my car's status is also affecting my wife's car's status. Although in the past, my wife's car has usually been far ahead of my car on the main software branch.
The majority of people on 3.25 were because of the trial and we still make up around 50% of the entire N.A fleet according to TeslaFi data. Tesla is very slowly moving 3.25 people to 14.8 (about 350 in the last couple days on TeslaFi) but seems they are prioritizing getting the last 8.9 people to 14.8 for the free trial and getting the 14.6 and 14.7 people off of the buggy software.

If 14.8 turns out to be stable I think they will be accelerating most people off of 3.25 except for the people that usually are held back to get new FSD versions earliest.
 
Nice, my yoke replacement mobile tech came in a X.
OT: Yeah, an older X with Tesla Service is the standard Tesla service vehicle. My Tech had one of those up until a week or so ago. Said it had a couple of hundred K miles on it and the Battery was about shot and had to Super Charge every day. Said they offered him a choice of the same standard older X type service vehicle or the Plaid S, so easy choice. The Plaid is an early release and had some kind of "lemon" buy back problem so Tesla has been using as a loaner but gave it to him for service.
 
Back on topic. Where the hell is 24.9.6/12.4.1? I predicted it would be out by the first of this week and so far nada. Stop making me look bad Tesla. Or at least worse than I already do. 🤣

This 12.4.x release is starting to look boggy. We need to see some momentum in the employee testing.
Seems gottagofast is infectious.
 
Back on topic. Where the hell is 24.9.6/12.4.1? I predicted it would be out by the first of this week and so far nada. Stop making me look bad Tesla. Or at least worse than I already do. 🤣

This 12.4.x release is starting to look boggy. We need to see some momentum in the employee testing.
Slight delay it’s still releasing this week though
 
  • Funny
Reactions: aronth5
Several threads stated about it. Here is one you can rummage through.

Not much there except people complaining.

The main theory seems to be ‘safety’ because people got the modes confused. Quite honestly, this is bull$hit. I’ve been using TACC and FSD for the last 2 ½ years and never had an issue. It’s only a problem if you don’t bother to pay attention, and one could use their rationale to argue that TACC causes mode confusion because people don’t know when it’s enabled or not.
 
Not much there except people complaining.

The main theory seems to be ‘safety’ because people got the modes confused. Quite honestly, this is bull$hit. I’ve been using TACC and FSD for the last 2 ½ years and never had an issue. It’s only a problem if you don’t bother to pay attention, and one could use their rationale to argue that TACC causes mode confusion because people don’t know when it’s enabled or not.
There are hundreds (exaggeration, obviously) of posts of people disengaging FSD and TACC still being on and scaring them or FSD not activating and TACC activating instead.

I never used TACC and found it annoying that steering wheel disengagement kept TACC on.
 
The main theory seems to be ‘safety’ because people got the modes confused. Quite honestly, this is bull$hit. I’ve been using TACC and FSD for the last 2 ½ years and never had an issue.
It may not be a problem for you, but it certainly is for other people. (And sadly, changes often get made to protect the worst drivers, at the detriment of the best drivers.) From a recent NHTSA document, they show 25% of the crashes they looked at were caused by mode confusion, where people disabled Autosteer and didn't realize it: https://static.nhtsa.gov/odi/inv/2024/INIM-RQ24009-12199.pdf

1717023685363.png


And they make it sound like they are going to demand single-pull be the only option:
Notably, Item 1 of the recall (single pull activation of Autopilot) is not the default setting on vehicles that received the remedy in the field. For those vehicles, the consumer must navigate to a menu and enable the feature. Additionally, the feature can readily be enabled and disabled by the driver. VRTC testing also showed it was possible to make this change while driving.
Essentially, they don't think making a feature to reduce mode confusion optional meets demands of the recall.

1717024378301.png


Maybe Tesla can go back to a two-level activation process, but making a steering override still cancel both to meet NHTSA's demands. (Once you are in AP, that taking control of steering cancels both Autosteer and TACC.)
 
Last edited:
From a recent NHTSA document, they show 25% of the crashes they looked at were caused by mode confusion, where people disabled Autosteer and didn't realize it: https://static.nhtsa.gov/odi/inv/2024/INIM-RQ24009-12199.pdf

View attachment 1051731
Interesting that the remedy was to remove TACC fallback as opposed to keeping Autosteer active to address "Inadvertent Steering Override" where potentially there were cases the driver expected Autosteer to continue. I understand there's complications and different types of mode confusion in either solution, but potentially 12.x is much better suited for collaborative steering than basic Autopilot or even 11.x as end-to-end should be able to understand driver intent.

Pretty sure with 11.x, if I pushed the accelerator while waiting at an unprotected left turn with oncoming traffic, it would incorrectly think it should continue on its current desired left turn path; whereas with 12.x, it understands accelerating with oncoming traffic means to creep straight forwards. Similarly 12.x detecting manual steering could infer the driver is trying to make a lane change to complete the maneuver whereas an accidental steering towards the road edge is would likely result in correction back to the lane. There's also great potential for positive training data such as driver making slight adjustments to avoid potholes then letting end-to-end complete the maneuver.
 
  • Like
Reactions: sleepydoc
Interesting that the remedy was to remove TACC fallback as opposed to keeping Autosteer active to address "Inadvertent Steering Override" where potentially there were cases the driver expected Autosteer to continue. I understand there's complications and different types of mode confusion in either solution, but potentially 12.x is much better suited for collaborative steering than basic Autopilot or even 11.x as end-to-end should be able to understand driver intent.

Pretty sure with 11.x, if I pushed the accelerator while waiting at an unprotected left turn with oncoming traffic, it would incorrectly think it should continue on its current desired left turn path; whereas with 12.x, it understands accelerating with oncoming traffic means to creep straight forwards. Similarly 12.x detecting manual steering could infer the driver is trying to make a lane change to complete the maneuver whereas an accidental steering towards the road edge is would likely result in correction back to the lane. There's also great potential for positive training data such as driver making slight adjustments to avoid potholes then letting end-to-end complete the maneuver.
1. V12 will begin a maneuver (for me) if the pedal is pushed, regardless of safety. I've tested it on a left turn to see if it would, then hit the brake. I wanted to see if it would go closer to the stop line, but the car just started going.
2. Not sure what you mean by collaborative steering. When V12 is active your input on the wheel satisfies nag, that's it, until you disengage.
 
Turn arrows on the lane itself is one of the few things FSD visualizes. It doesn't do that for road signs other than speed limits and stop signs.

Whether it acts on that information is another matter.

For your particular example are you saying that even when the road is completely clear (such that turn arrows is visible) the car doesn't visualize it?

What I did observe just driving is the lane arrow marking will disappear if it is partially obscured by another car. That seems to indicate it is read in real time, not something that is based solely on a map. If also doesn't show up until it is in visual recognizable range.
Nice sunny Sunday morning. The intersection was just painted after the winter sand and salt scrubbed the paint. No one else in the intersection. Couldn't be more perfect. The car drove into the middle (straight only) lane (green light) and ignored the left turn only lane (green arrow) and attempted the left turn I had set navigation for. This is the same spot that would have been the left turn lane a year ago. It followed an out of date map.

Another example. Same intersection, nav set to go straight this time. It pulls into the right turn only lane (also with an arrow) and attempts to go straight.

It acts as as though there are only two lanes (which it did a year ago) instead of three.

I didn't give it too much thought when the lanes weren't clearly striped, but now they are.

I have 2024.8 map software. Date doesn't make it accurate.
 
Because not every sign or road marking is the same, and depending on traffic it may not even ever see the painted turn arrow on the ground. (That is part of why having a map input as a hint is helpful.)
Well, come on over to Idaho and try it! Guessing about my experience is like looking at a cookie. Until you've tasted it, what do you really know?

We'll have a beer and a laugh.