I’m just hoping to get months worth of the underlying features added to my car. It stinks being so far behind the base software stack due to being in FSDb.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
FSD Beta 10.69
- Thread starter Buckminster
- Start date
Zane Liu
Member
With V11 incoming it seems you may be in parallel for all future features soon enough!
willow_hiller
Well-Known Member
Can anyone who just received 10.69.3.1 confirm whether the feedback button is still there? Reports of it being gone on Reddit:
FSDtester#1
"Circle Jerk Old Guard" per cyborgLIS aka "CJOG"
10.69.3.1 rollout seems to have stalled for now. Teslafi is showing only 77 and not increasing at this time.
Looking through other threads around here, seems like nearly 20 people have posted about newly getting added to FSD Beta since last night. If FSD Beta population was around 160k, then I would estimate roughly 3000 vehicles were added last night and maybe half have installed the update so far.
No additions since last night's batch nor any upgrades directly from 10.69.2.x yet.
FSDtester#1
"Circle Jerk Old Guard" per cyborgLIS aka "CJOG"
Maybe we should all go and tap the advanced button a bunch of times... see if that works?
right - currently it looks like there are more new additions than there are coming from a previous release.
Following conventional definitions of words isn't really a strong point for EM
AlanSubie4Life
Efficiency Obsessed Member
Chuck with the jitters means another beer for me from @Daniel in SD !!! (Combining this with the earlier 6 attempts from 10.69.3 - which we counted as 6/6 even though it was actually 5/6 - you’ll recall my magnanimity.) 
I’ll come up with the final count in a bit and modify this post.
----- Modifications....
It was 6/8 or 5/8.
First attempt was a failure. It was definitely correct for Chuck to disengage (not jitters). Car's body language was completely wrong.
5th attempt was arguably a failure with a large gap (> 6 seconds) it should have gone on. We'll give it a pass, but note this is NOT setting a precedent for allowing such mediocrity.
6th attempt was a failure due to a massive gap (>7 seconds I think) that was missed, but also Chuck disengaged in the median (human/machine interface issues!).
Combine with 6/6 to give us 12/14 or 86%. So close!
Overall the behavior looks like it could work. Working on improving that communication between the driver and the machine is clearly needed. It's also not nearly as assertive or fast as it should be.
This was the version that had the better cross-traffic speed estimation (wasn't it?), and that seems to have led to some regression in its ability to shoot gaps.
As discussed before, it's essential that it be more aggressive in crossing the road. It will not be comfortable at all until it drives more assertively.
Looking forward to them overfitting a few more scenarios. My unprotected left turn is still a debacle of different outcomes. It can usually do it without disengagement, but it is very slow and clumsy and the path it takes (using the merge lane or not) is completely unpredictable. And there's a lot of jerking. It also doesn't get the speed right on the main road so I have to remember to manually dial up the speed to 45mph before it starts the turn, otherwise it turtles at 25mph on a road where people legally travel at 60mph (speed "limit" 45mph).
Definitely more analysis of this video to be done, but don't have the time at the moment.
I’ll come up with the final count in a bit and modify this post.
----- Modifications....
It was 6/8 or 5/8.
First attempt was a failure. It was definitely correct for Chuck to disengage (not jitters). Car's body language was completely wrong.
5th attempt was arguably a failure with a large gap (> 6 seconds) it should have gone on. We'll give it a pass, but note this is NOT setting a precedent for allowing such mediocrity.
6th attempt was a failure due to a massive gap (>7 seconds I think) that was missed, but also Chuck disengaged in the median (human/machine interface issues!).
Combine with 6/6 to give us 12/14 or 86%. So close!
Overall the behavior looks like it could work. Working on improving that communication between the driver and the machine is clearly needed. It's also not nearly as assertive or fast as it should be.
This was the version that had the better cross-traffic speed estimation (wasn't it?), and that seems to have led to some regression in its ability to shoot gaps.
As discussed before, it's essential that it be more aggressive in crossing the road. It will not be comfortable at all until it drives more assertively.
Looking forward to them overfitting a few more scenarios. My unprotected left turn is still a debacle of different outcomes. It can usually do it without disengagement, but it is very slow and clumsy and the path it takes (using the merge lane or not) is completely unpredictable. And there's a lot of jerking. It also doesn't get the speed right on the main road so I have to remember to manually dial up the speed to 45mph before it starts the turn, otherwise it turtles at 25mph on a road where people legally travel at 60mph (speed "limit" 45mph).
Definitely more analysis of this video to be done, but don't have the time at the moment.
Last edited:
Zane Liu
Member
FSDtester#1
"Circle Jerk Old Guard" per cyborgLIS aka "CJOG"
I hope you are an alcoholic with all these free drinksChuck with the jitters means another beer for me!!! (Combining this with the earlier 6 attempts from 10.69.3 - which we counted as 6/6 even though it was actually 5/6 - you’ll recall my magnanimity.)
I’ll come up with the final count in a bit and modify this post.
! You would appreciate them at least...Went out for a spin with it. I don't see the feedback button.
I love reading the release notes.
"Peripheral response time increased by 9.275% by adding turboencabulator side channel to the additional 20x photon training data input validation."
While the driver response is never "hmm, the car took that corner 6.5% better than last time"
Its always "wow, that was smooth"
"Peripheral response time increased by 9.275% by adding turboencabulator side channel to the additional 20x photon training data input validation."
While the driver response is never "hmm, the car took that corner 6.5% better than last time"
Its always "wow, that was smooth"
clydeiii
Member
We’re in the earliest hours of its rollout. Don’t give up!
Zane Liu
Member
I hope so!
You think the rollout will continue this weekend? Or are people discovering the update still after last night?
it looks like they messed up the release notes which may or may not trigger a delay, can't tell
FSD Beta 10.69.3.1 (2022.36.20) Official Tesla Release Notes - Software Updates
Tesla software update 2022.36.20 (FSD Beta 10.69.3.1) includes Full Self-Driving (Beta) Suspension, FSD Beta v10.69.3.1 Release Notes, Energy App, Alternate Routes, Cabin Overheat Protection, Blind Spot Camera, Tesla App, Theater Mode, Tesla Profiles, Status Bar, Driver Profiles, Traffic Along...
www.notateslaapp.com
Lol, true. But would TMC users accept a release note that said "It feels smoother"?
EVNow
Well-Known Member
When critics are grasping at straws you know you are winning
EVNow
Well-Known Member
I think currently it is just too slow to figure out whats happening. It can't be assertive when its uncertain.
I've a 35 mph unprotected left T junction - single lane each direction and yet it fails many times.
Similar threads
- Article
