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.
I'll forego my turkey this thanksgiving in place of a nice big V11 stack of single
I would like that too, but it seems a bit curious that Tesla is preparing to release FSD Beta 10.69.3.1 / 2022.36.20 today. They could start rolling out this 10.69 update this week then 11 wider next week, but it somewhat feels like Tesla is preparing this not-single-stack version as an interim to maybe help gather data to finish polishing single stack for wide release in December.
 
  • Informative
Reactions: Discoducky
Broke down the current sub-set of non-final V11 release notes...
Here's the other FSD Beta 11 release note items (except the first 2) if you want to do an itemized again:
  • Improved merging behavior by leveraging lane geometry and lane bounds, association with coarse map information and better gap selection algorithms, allowing for smoother and safer experience.
  • Added highway behavior to offset away from blocked lanes and generic obstacles like road debris while also adding a smooth hand-off between in-lane offsetting and lane changing.
  • Improved speed-based lane change decisions to better avoid slowing down traffic in fast lanes, and interfere less with navigation.
  • Reduced sensitivity for speed-based lane changes in CHILL mode.
  • Improved lane changes to allow higher jerk maneuvers if required to stay on-route or to move away from lane blockages.
  • Improved smoothness at highway lane splits by being less strict about centering between lane lines and allowing lower jerk maneuvers, where safe to do so.
  • Reduced latency of trajectory optimization by 20% on average, without sacrificing behavior, by leveraging numerical tricks for more efficient computations.
 
I go into detail on the main thread ( #378,459 ), but I think it's truly bizarre that a "wide release" is starting with less than 2% of the vehicles that have the FSD Beta. And so far none of them have been upgraded from 10.69.2.4.
And here we are over a day later. Still only 77 TeslaFi vehicles (65 installed, 12 pending) with 10.69.3.1. And still none updated from 10.69.2.4. There's nothing about this that looks like a "wide release" despite Elon's tweet
 
Here's my take on V11 (first single stack drop) release notes...

V11
Tesla Release notesLaymens terms and how it applies to the experience
- Enabled FSD Beta on highway. This unifies the vision and planning stack on and off-highway and replaces the legacy highway stack, which is over four years old. The legacy highway stack still relies on several single-camera and single-frame networks, and was setup to handle simple lane-specific maneuvers. FSD Beta's multi-camera video networks and next-gen planner, that allows for more complex agent interactions with less reliance on lanes, make way for adding more intelligent behaviors, smoother control and better decision making.Single stack! FSD on highway! Occupancy map everywhere! Demonstrates huge confidence in this NN as the speeds are much higher. Occupancy flow of high speed kinematics is also very interesting.
Improved Occupancy Network's recall for close by obstacles and precision in severe weather conditions with 4x increase in transformer spatial resolution, 20% increase in image featurizer capacity, improved side camera calibration, and 260k more video training clips (real-world and simulation).OccNet improved recall means less false negatives, meaning less times the car thinks it sees something that is NOT actually there.
Improved precision is less false positives, meaning less times the car thinks something IS NOT there, when in reality IS there.
The improvement in precision is for poor weather, meaning that FSD will operate in these conditions much better and it is therefore less likely to put the warning message about "reduced functionality due to poor weather detection". This is great for Seattle!
The OccNet 4x increase in transformer spatial resolution simply means that it can see 400% more volumetric detail, meaning that it should be much better at seeing smaller things that are close up and see things farther away. 20% increase in image featurizer capacity I think means they can support 20% more features in the OccNet, which means they can hopefully now recognize more road signs.
Improved merging behavior by leveraging lane geometry and lane bounds, association with coarse map information and better gap selection algorithms, allowing for smoother and safer experience.I'm guessing that lane geometry (curve radius, width dynamics, intersecting locations) and lane bounds (precise location in the OccNet) were previously NOT used in lane merging, nor coarse map information (this was talked about at AI Day and includes things like 'Is_Bus_lane' and other deterministic markings/signs), but was already a NN as it says 'better gap selection algos'. Seems that we could then call it a Lane Merge Net (LMN). I'm guessing that they've been working on this a while as the same LMN would need to be used everywhere, so they needed to train it on high speed and very slow speeds. This sounds very difficult, so lets hope it is a good experience out of the gate!
Added highway behavior to offset away from blocked lanes and generic obstacles like road debris while also adding a smooth hand-off between in-lane offsetting and lane changing.This seems simple, but as I read this it sounds like they have at least two NNs (in-lane offsetting - ILO NN and lane changing - LC NN). So, I read this as they have these input NNs into the neural planner. The neural planner then can seemingly plan better (with more confidence and NOT changing at the last second jerking the steering wheel). Will be interesting to experience what objects are avoided and how well it senses and adapts to blocked lanes. Hopefully construction zones and "Road Closed" signs trigger a route change and it is handled smoothly.
Improved speed-based lane change decisions to better avoid slowing down traffic in fast lanes, and interfere less with navigation.Pretty straightforward, will be interesting to see how humanlike this feels. Hopefully less missed exits and entrances which I'm assuming they mean by saying 'interfere less with navigation'.
Reduced sensitivity for speed-based lane changes in CHILL mode.Again, straightforward. Chilled out lane changes, not last minute, nice and smooth, get in the correct lane and stay in the lane.
Improved lane changes to allow higher jerk maneuvers if required to stay on-route or to move away from lane blockages.This sounds like allowing FSD to operate outside the lane boundaries on occasion when merging like humans do when it is safe.
Improved smoothness at highway lane splits by being less strict about centering between lane lines and allowing lower jerk maneuvers, where safe to do so.More humanlike highway interchange behavior like humans drive, hopefully less lane centering when going around a bowtie interchange. I think the lower gerk maneuvers could be in stop and go traffic, to smooth out the behavior. I find it nearly unusable in some stop and go traffic, so I hope it applies in those situations.
Reduced latency of trajectory optimization by 20% on average, without sacrificing behavior, by leveraging numerical tricks for more efficient computations.Faster reactions to up-coming road and object dynamics. It doesn't say specifically, but this could apply to both the FSD car and cars around it. It is great to see this get faster as there are times when you just want FSD to react to something and then it seemingly reacts but a second too late. Let's hope it reacts more like a human!
SUMMARY
This build is for single stack integration. Making FSD work at speeds higher than 55 MPH zones and hopefully acting more humanlike at higher speeds like anything over 45 MPH (usually the limit where there are crosswalks and predictable VRUs)
Things to test/'pay attention to' with this build
When cars make rights in front of ego, should be much more human-like and not cause someone behind to honk
When cars leave the lane of travel while braking, should be much more human-like and not cause someone behind to honk
All types of merging, lane changing should improve as well as blinker behavior
Try Chill mode and see if there is a noticeable difference from other modes
Try lots of highway stuff and see how it handles on and off ramps, bowtie interchanges and traffic.
Does ego make faster decisions overall?
Does ego make smother turns?
Are there less unnatural slowdowns overall?
 
And here we are over a day later. Still only 77 TeslaFi vehicles (65 installed, 12 pending) with 10.69.3.1. And still none updated from 10.69.2.4. There's nothing about this that looks like a "wide release" despite Elon's tweet
And now there's another 459 on TeslaFi added to the pending queue, so maybe *now* it's finally going a bit wider. And I know that some vehicles with 10.69.2.4 are getting it, because one of mine is.
 
OK, I just got the FSD beta (I drive like Mad Max, so not surprised they didn't give it to me for a long time).

As an investor . . . I like the progress compared to "original" FSD that the cars come with.
As a user . . . it's got a long way to go before I consider it "wife approved", which is my benchmark that when my wife engages it that she is not going to get into an accident.

Only had it one day, but my observations are:
1) far too aggressive in our little town square, but where speeds bump up not always "aggressive" enough. Things that would be intuitive to a human driver (stay in this lane, I'm turning from it in 1-2 miles) it doesn't do.
2) turns still have a good way to go. Some were buttery smooth, but on one unprotected left this AM the car almost turned right into a car that was stopped to take a left (was a street where both the oncoming car and I were taking lefts at intersections just past each other, but about 20ft). I had to take control to keep the car from turning into the stopped car.
3) map errors (vision errors?) - our neighborhood has a 15mph speed limit, and it is posted. The car continued to think it was 25mph, and wanted to do 28, which was way way too fast for all the parked cars, etc.


Someone in the main thread, I believe, said that the current FSD version drives like a drunken 13 year old. I would concur with that. I'm excited it can do that well, and it's come a long way in a short period of time, but there is a long way left to go.
 
Why do you say "blind"?
I just feel like it can’t see far enough to the left or right at times, without pulling too far forward. I can’t see what the B-pillar cameras can see so I have no way to validate how good the visibility is. I hope this isn’t a problem, but it’s a concern I have. Has the side view cameras being too far back been debunked already?
 

I don't want to downplay the significance of v11, but if I recall correctly, Elon later on stated that the performance jump from 10.69 to 11 will be less significant. The smaller updates from v10.69.x.x contain software features already from v11.

Biggest progress will be made on highway driving.
I assume that the same bugs/problems/.... with city driving on 10.69.x will still exist on v11.
Take my opinion with a bag of salt nonetheless 😉
 
I don't want to downplay the significance of v11, but if I recall correctly, Elon later on stated that the performance jump from 10.69 to 11 will be less significant. The smaller updates from v10.69.x.x contain software features already from v11.

Biggest progress will be made on highway driving.
I assume that the same bugs/problems/.... with city driving on 10.69.x will still exist on v11.
Take my opinion with a bag of salt nonetheless 😉
I really wonder about that. We really have little issues in highway driving. Wonder what issues they found and how it will be better.
 
  • Informative
Reactions: ElectricIAC
I really wonder about that. We really have little issues in highway driving. Wonder what issues they found and how it will be better.
Lane change at many exits when unnecessary. On divided highways with cross streets, if you are in the left lane and there are no paint lines for the left turn lane, the car will straddle both lanes and then violently jerk back to the left lane when it realizes that the car is about to hit the median. Very unsettling. Also 50 mph limit shown in a 65 mph zone. There are likely more, but these are the most common for me.
 
Moving to single stack wasn't going to happen until it functioned as well as the existing solution on the highway. So we can presume that now it does. But in reality, that just means it's likely just as good on average. It will, for all practical purposes, just be different. So we'll see about as many problems, they'll just be new ones. Old ones will have gone away.

At least that's what I'm expecting. Then, with any luck, it should improve at a quicker pace.
 
  • Like
Reactions: FSDtester#1