Ok, I agree to boycott FSD, at least tomorrow. In fact I won't drive my Tesla at all tomorrow and that will teach Elon a lesson.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
FSD v12.x (end to end AI)
- Thread starter Buckminster
- Start date
-
- Tags
- #fixphantombraking 12.4.x is coming autopilot chill is not chill elon is an ass elon rocks fix fsd extreme late deceleration fsd does not commit to go fsd does not maintain proper speed fsd doesn't read signs properly fsd hard and late braking fsd has decision wobble fsd jerky driving fsd late deceleration fsd sometimes may not detect train crossings fsd stops too early too quickly fsd stops too far behind line fsd stops when there is zero visibility fsd tailgating in city streets fsd very slow at stops fsd will never enable l3 with hw4
zoomer0056
Active Member
hey8you
Member
No only because FSD works mostly well for me plus it allows me to fire one out the window while passing student housing on the way home from work. Some of the new RNs here are hot!
Great news and very useful NO FSD info in the FSD thread.
We are also very concerned about your NO FSD tire pressure. Could you please post this info also.
stopcrazypp
Well-Known Member
I laughed at this. No it's not made up. It's literally in the UNECE regulation standard for ALKS (AKA L3):
"ALKS’s standard safety concept defines a 10 seconds transition period so that human driver must remain able to respond to a system request so that the human driver assume control of the vehicle when driving system do not do it anymore"
Also what Mercedes explicitly uses as I quoted up thread. That period was not randomly chosen, it was because the regulation that the EU follow a calls for it.
"the buttons in the steering wheel rim turn red, the vehicle requests the driver to retake control within ten seconds."
Will Mercedes jump to level 3 before Tesla? Looks like it.
Sleeping is not desired on 5 minute trips. But your design must still account for it (as well as for intermediate times like 30 minutes) if it explicitly allows sleeping. You can't rely on the "desire" of the owner. To me, a standby driver car is one that legally must leave its ODD while...
teslamotorsclub.com
As I noted multiple times upthread, SAE doesn't use 10 seconds, what they use is "at least several seconds" (J3016 3.12 Note 3; 3.17 Example 2)
In case you don't believe the source posted, I checked and it's right there in page 11 of the PDF document you posted!
"NOTE 3: At Level 3, an ADS is capable of continuing to perform the DDT for at least several seconds after providing the fallback-ready user with a request to intervene. The DDT fallback-ready user is then expected to resume manual vehicle operation, or to achieve a minimal risk condition if s/he determines it to be necessary. "
Last edited:
So anyway... has anyone on 12.3.6 with 2024.20.3 having any issues with turn signals either not being used or turning off only after a few seconds before the car even reaches the turn? My Model Y RWD is doing it almost all the time and even if I manually turn them back on, it will even turn them right back off.
I've tried rebooting the car (twice) and though I can't imagine the whole recalibrating the cameras thing would matter, I'm almost to that point.
I've tried rebooting the car (twice) and though I can't imagine the whole recalibrating the cameras thing would matter, I'm almost to that point.
Last edited:
Just noticed something with the new spring update. I have HW3 (Intel Atom), and when FSDS engages, I get the full screen visualizations (I have that enabled), and starting to enjoy it. The upper right has a small ghosted version of the map, with traffic information in it, as I'd expect. However, when I disengage FSDS, the ghosted map in the upper corner loses traffic info.
Anyone else experiencing this? I'm sure it's just a bug that'll get fixed, but wasn't sure if it was something to do with the older Intel Atom vs newer Ryzen MCUs.
Anyone else experiencing this? I'm sure it's just a bug that'll get fixed, but wasn't sure if it was something to do with the older Intel Atom vs newer Ryzen MCUs.
The only time I've experienced no turn signals is when I engage FSDS while in a turn lane. But if I engage outside the turn lane, I always get turn signals for upcoming turns.
gottagofast
2022 Model 3 Performance
Yeah this is just on regular two lane roads turning left or right onto another two lane road. Happens probably 80% of turns. It actually started doing this on my last week of the one month trial on 2024.14.9. I decided to subscribe for a month or two for the summer months, thinking that perhaps now that I was on 20.3 they might have slipped in some kind of fix. Nope.
stopcrazypp
Well-Known Member
I highly doubt this. Extensive studies have been done on this. Generally novice drivers take about 8 seconds to regain situational awareness, experienced drivers around 6 seconds. This is from being awake and doing activities where you still have peripheral vision.
That's probably why the UN standard settled at 10 seconds.
When asleep, the time should be longer, not shorter, because you need to add the time for you to perceive the alarm, open your eyes (maybe rub them to clear them up), eyes readjust to brightness, then gain some peripheral context. Unless you received special forces training or you are an exceptionally focused person, I doubt someone can respond/regain situational awareness in 10 seconds or less from sleep, certainly not an average person.
This is putting aside people may not want to wake up (how many people can claim they never hit the snooze button or let the alarm ring a bit longer).
Last edited:
Mine has become much more tolerant of it with 12.3.6, but she's still not super-happy about it. She especially complains about it missing potholes near certain corners, and the sometimes excessive acceleration taking off from stop signs in low-speed-limit areas. I still get a lot of "I wish you would just drive the car yourself", but she's allowing a lot more miles to happen while she's riding. With previous versions, it had got to the point where I would almost never engage FSDb with her in the car, because it wasn't worth the verbal assault.
EVNow
Well-Known Member
While this is a possibility - in this particular case I think Elon has a (atleast had a) naive view of AI training.
This is the reason he repeatedly talks about Alpha Go and how it took forever for it figure out basic moves but got very good very quickly after that. In other words, he does this AI progresses rapidly ... "exponentially".
In the case of Autonomous Driving we have so no evidence for that ... and yet, he continues to behave as if he truly believes in that. I think, even though its an act, at some basic level he does believe it to be true.
I mean who talks this way besides Alan?
What in the world does decades of support mean when people drive so differently?"This is objective, quantifiable, decades of support, well established limits for human comfort"
zoomer0056
Active Member
I think my favorite is when someone says the slowing/stopping is smooth for them, and they're told "no, it's not smooth, look at the regen bar". I just imagine someone in the car, the car is slowing/stopping and they think "wow, this is pretty smooth, I like it", then they look at the regen bar and change their mind "oh wait, it's not smooth, I clearly must have been wrong".I mean who talks this way besides Alan?
"This is objective, quantifiable, decades of support, well established limits for human comfort"
It's like someone walking into a room and saying "it's kinda warm in here", and then you point them to the thermostat and say "look at the gauge, it's perfectly comfortable". And they say "oh, you're right, it is perfectly comfortable - I'm wrong, it's not warm in here".
Daniel in SD
(supervised)
It’s definitely something that people study. Lots of references and citations here.I mean who talks this way besides Alan?
"This is objective, quantifiable, decades of support, well established limits for human comfort"
I’m
This was in a Mustang. No not the car, the P51 airplane.
This V12 was very smooth, and wouldn't disengage - unless shot down.
Not what you asked for, but here is a photo of a V12.
This was in a Mustang. No not the car, the P51 airplane.
This V12 was very smooth, and wouldn't disengage - unless shot down.
My SO gets spooked when I disengage by turning the steering wheel. That always causes at least at quick swerve - even if very brief, it is the jerk that startles her. So I always try to disengage with the stalk instead.
I think that FSD/AP should not disengage from torque, but rather should let torque overpower the AP. This would be an intervention, rather than a disengagement, like the accelerator does now. I think this would help with FSD comfort and acceptance, by allowing small corrections rather than abrupt jerks. But I'm not Elon, so my vote doesn't count.
sleepydoc
Well-Known Member
It's the minimum number before the system starts a "minimum risk maneuver" (i.e. turns on the hazards and pulls over or stops) They do not list a maximum so saying "you need to respond within 10 seconds" is not accurate.
