Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

FSD Beta 10.69

This site may earn commission on affiliate links.
Now we get stuff like this, where it seems a bit too assertive around pedestrians (IMO):
This is an excellent demonstration of the issue. You can see in the regen bar the brakes (regen and maybe friction) get slammed on. There was no need for this. The car just had to slow down in time, like a normal human. Plenty of time. The pedestrian was detected and a collision was easily avoidable. Slow down later, more gradually, and go more slowly past the pedestrian. These are not incompatible.

Obviously I would expect extreme caution around pedestrians, so I fault it less here for this specific incorrect behavior. But it is capable of doing this type of nonsense at any time.
 
  • Like
Reactions: nvx1977
I do not believe vehicles behave differently. You can look at every 10.69.1.1 video showing the regen bar and you will see this behavior (not for every stop, though). It is of course worse when coming to a stop from 40-50mph than it is when moving around at 25mph. But it is there (sometimes) in both cases.
As I said before this is pretty much at the bottom of my wish list.
 
Unfortunately, that’s not the way it is right now. Basically it is something that unless you are specifically testing it (a perfectly legitimate and fulfilling activity in its own right - and even fun as you watch the vehicle dive for a curb as it goes for a turn lane 😂 ), it is not useful. You have to immediately turn it off except in the most trivial driving situations (where it does moderately well - marked roads, no turns, no traffic, no pedestrians, no traffic control). In these situations, my wife will not notice it is being used - most of the time (and if it screws up once it has to be turned off).
"It is not useful". I disagree.
I generally find FSD useful for many drives especially compared to when I first started using it last October. A significant reason is of course FSD has improved considerably but another reason is I am far more forgivable now that I understand how FSD works (and doesn't work). It's a fine line but find I don't disengage as frequently simply because I know what FSD is going to do. Hands on the wheel at all times though since FSD still does some bizarre things. Just one example, recently I was driving in an area that I was unfamiliar with, it was drizzling and at night. I was definitely safer using FSD in that situation.
 
Last edited:
"It is not useful". I disagree.
It’s not correct to summarize that way, since I clearly indicated there are specific scenarios where it is actually useful, though additional utility above regular AP can be debated.

It's a fine line but find I don't disengage as frequently simply because I know what FSD is going to do. Hands on the wheel at all times though since FSD still does some bizarre things.
I didn’t say it was not usable - it is. I just said it was not useful - a characterization of its utility. I can drive it and not disengage as well, as long as no one is around, but don’t see the point.

We had this discussion a few pages back but I guess the question is “what is the utility” in these more complex scenarios which bring out the bad behaviors? What’s the value, taking into account the undesirable behaviors, and forgetting about any value testing might have?

My feeling is that it might actually be a bit useful in a broader set of scenarios and reduce some mental effort if it actually fixed some of these behaviors. I’m not certain, because that’s not what we have, but it seems possible.
 
"It is not useful". I disagree.
I generally find FSD useful for many drives especially compared to when I first started using it last October. A significant reason is of course FSD has improved considerably but another reason is I am far more forgivable now that I understand how FSD works (and doesn't work). It's a fine line but find I don't disengage as frequently simply because I know what FSD is going to do. Hands on the wheel at all times though since FSD still does some bizarre things.
I agree completely. I drove to and from my office, 98 miles round trip for 2 days.
50% Noa and 50% fsdb. 2 disengagements today and 1 yesterday. Yes, that includes hitting the go pedal. 3 disengagements in almost 200 miles.
Seems like progress to me. Beta 10.2 or 10.3, whatever us 100 % SS people got last October averaged 7 to 8 each way or about 15 round-trip. I think the car, the roads we all travel, the traffic, all add the exponential differences in all of our experiences.
For me right now, it is at 98% there. Since I have to complain about 1 thing. Don't slow down soo frigging quickly sometimes. There, now that's a good and bad review.
 
We can hope of course that “this time will be different.”
But you don’t know Elon like I do. This time will be different! /s

I haven’t gotten the update yet, and luckily I’m too busy with the iPhone news to be too concerned about Tesla updates right now haha.

Just today on 10.12.2, it did the worst possible merge onto the highway that resulted in one car preemptively evading me which had another car evade them by going onto the shoulder. I think they were inconsiderate and overly dramatic, but that’s the sort of thing FSD needs to better handle. They were never in danger because I went onto the other shoulder before anything was even close to happening, but they didn’t know that. In retrospect, instead of waiting to disengage right before the situation became unsafe, I should have disengaged before the situation could have seemed unsafe to anyone around me.

That and it almost ran over my neighbor while they were crossing the street, but that’s just classic FSDb—barely worth mentioning.
 
I suggest that 10.69.1.1 should not be released more widely. Today, it missing two turns while on navigation and it attempted to run two red lights. I tried rebooting after the missed turns, but the red light issue is a major safety hazard. The car does make most turns correctly and it stops for red lights, so there is definitely a bug that is affecting my car. It has never tried before on previous FSD beta versions to proceed through a stoplight that has been visibly red for an extended time.
My red light running attempt situation was a case of interpreting the wrong light. I was in a left turn lane with 2 lights with red arrows, to my right was a straight lane that had a green light. It initially seemed like it was slowing down and then it started accelerating and I pressed the brake to stop the car. I re-enabled FSDb after all the lights were red.
 
  • Informative
Reactions: FSDtester#1
It’s not correct to summarize that way, since I clearly indicated there are specific scenarios where it is actually useful, though additional utility above regular AP can be debated.


I didn’t say it was not usable - it is. I just said it was not useful - a characterization of its utility. I can drive it and not disengage as well, as long as no one is around, but don’t see the point.

We had this discussion a few pages back but I guess the question is “what is the utility” in these more complex scenarios which bring out the bad behaviors? What’s the value, taking into account the undesirable behaviors, and forgetting about any value testing might have?

My feeling is that it might actually be a bit useful in a broader set of scenarios and reduce some mental effort if it actually fixed some of these behaviors. I’m not certain, because that’s not what we have, but it seems possible.
You said it was not useful and @aronth5 disagreed. I also disagree - I use FSD on a regular basis. Yes, I have to babysit it, but that’s far from being useless as you argue. If you want to use it to take a nap while you drive through rush hour traffic then you’ll clearly be disappointed, but not being suitable for all uses is different from not being useful.
 
You said it was not useful and @aronth5 disagreed. I also disagree - I use FSD on a regular basis. Yes, I have to babysit it, but that’s far from being useless as you argue. If you want to use it to take a nap while you drive through rush hour traffic then you’ll clearly be disappointed, but not being suitable for all uses is different from not being useful.
I don’t know how I could have been more clear about the situations where it does moderately well (and I would call it useful there, as was clear).

I mean, we are discussing City Streets here…the other stuff is largely (not completely of course!) covered by NOA - part of EAP - and that middle ground…is where I described it as doing moderately well.
 
Last edited:
I don’t know how I could have been more clear about the situations where it does moderately well (and I would call it useful there, as was clear).
Unfortunately, that’s not the way it is right now. Basically it is something that unless you are specifically testing it (a perfectly legitimate and fulfilling activity in its own right - and even fun as you watch the vehicle dive for a curb as it goes for a turn lane 😂 ), it is not useful. You have to immediately turn it off except in the most trivial driving situations (where it does moderately well - marked roads, no turns, no traffic, no pedestrians, no traffic control). In these situations, my wife will not notice it is being used - most of the time (and if it screws up once it has to be turned off).
Edit - tha majority of my use is on city streets. FSDb, not NoA
 
Last edited:
  • Like
Reactions: FSDtester#1
well this is just peachy - the 12v in my Model 3 just decided to start to fail.
My car has nice messages like this
Schedule service to replace low voltage battery
Software will not update until battery is replaced

I'm presuming it's saying that because there is a software update pending. Oh well.
I got the same message yesterday. I contacted service and they came over this morning and replaced it. Still under warranty too so it was free.
 
Have any of you used 10.69.1.1 on highway (not talking about merging)? Is it improved vs 10.12?
Not sure if this is the consequence of any recent changes but one of my regular routes has a very short distance to merge twice to exit, and usually there are other cars in the way forcing NoA to brake hard to get behind them. Today was another such situation, but NoA actually sped up to get in front of the car to my right, before merging. The second lane change also felt more urgent than usual. But that could be because of the extra distance covered while passing the car.

Of course this was followed by another 2 necessary lane changes in FSDb mode, which were too timid and the cars behind me aggressively merged right and passed me in the meantime.
 
Not sure if this is the consequence of any recent changes but one of my regular routes has a very short distance to merge twice to exit, and usually there are other cars in the way forcing NoA to brake hard to get behind them. Today was another such situation, but NoA actually sped up to get in front of the car to my right, before merging. The second lane change also felt more urgent than usual. But that could be because of the extra distance covered while passing the car.
I've had this behavior for some time on NOA. However, I generally don't find the car speeding up to make exits because the other cars are rarely going slow enough for my car to be able to do this while remaining under the set speed. So, it usually resorts to slowing down to get behind traffic.
 
  • Like
Reactions: FSDtester#1
I've had this behavior for some time on NOA. However, I generally don't find the car speeding up to make exits because the other cars are rarely going slow enough for my car to be able to do this while remaining under the set speed. So, it usually resorts to slowing down to get behind traffic.

Yeah I’ve observed this behavior before but it’s been pretty consistent about this particular interchange. The speed differential was small (< 5mph) so I was surprised NoA was willing to risk the exit by trying to pass in the first place. Of course one data point isn’t much to draw inferences from especially with the beta software, so I’ll pay attention to it more next time.
 
  • Like
Reactions: FSDtester#1