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

FSD = Flaky Student Driver

This site may earn commission on affiliate links.
This morning my car received the 2021.36.5.2 firmware, and with it the latest FSD beta. Today I drove 38 miles, most of that with FSD engaged. My conclusion is that this software is extremely dangerous. I've seen lots of posts from people who have been eagerly awaiting it, noting how many thousands of dollars they paid for it. My conclusion: This ain't it! The software is extremely flaky. I've never taught a human being to drive, so I'm a little uncertain in the comparison, but it seems to me that it is, at best, at the level of a teenager who's driving a car for the first time. It's dangerously inconsistent -- taking turns at high speeds sometimes and too slowly at other times. It brakes hard to slow down to avoid colliding with a line of cars that had been clearly visible (stopped) far enough away for a much more sedate braking maneuver. It frequently can't decide which lane to be in, and sometimes weaves back and forth unnecessarily. It once planted itself firmly in a right-turn-only lane when it had to go straight, then jerked the wheel back and forth as if uncertain whether to turn or try to change lanes long after it would have been safe to do so. (On this occasion, I had to take control, and was rewarded with an FCW alert because of the car that was turning left from the other direction.)

It's best at highway speeds, but that's just plain old Autopilot. Make no mistake, though -- although both Autopilot and FSD are labeled as "beta," they're nowhere near the same level of beta. I suspect that Tesla called Autopilot "beta" as a CYA tactic, but that now creates the problem that, as something being tested outside of Tesla, FSD is technically beta -- but really it's at an alpha-test level of refinement. (I work in the computer industry, and the level of bugs I've seen in my 38 miles of driving far exceeds what I'd expect from beta software, particularly when the phyiscal danger of software driving a car is taken into consideration.)

I was also fairly impressed at how the car handled a low-speed road around a local park, which was packed with pedestrians. It kept a safe distance from the pedestrians and maneuvered with reasonable speed. This was a task that's similar to the Smart Summon feature, which has been tested extensively for a year or so.

To be clear, I'm sure that FSD will improve with time, and testing on real roads is required for this improvement to occur. I am not trying to be critical of Tesla or even of Tesla's FSD development strategy; I don't see a way around putting poor self-driving software in cars on real roads at some point, if we're to develop self-driving capabilities at all. I am, however, saying that using the software in its current form is an extra job. It is not fun to use, it does not reduce the stress of driving, and it is not worth the $X thousand dollars (or even $200/month) that we've paid for it. It is, at best, early beta-test software (really more like late alpha-quality), and it needs to be treated very cautiously. So if you want to beta-test it, and if you feel up to the task of monitoring a machine with less driving skill than the average teenager who is learning to drive, then do so -- but do so with caution and extra alertness. You're still legally responsible for the car, and if the software drives off a road, slams into another vehicle, or kills a pedestrian, YOU are responsible. Remember that every second you're supervising this Flaky Student Driver.
 
As many have stated, expectations especially dictate how one reacts to FSD beta. It’s not what we paid for (I paid $3k, but longer ago than those that paid more), but since I knew that going in (it’s “limited early access beta”) it’s been an amazing experience so far. I agree it’s more work and stressful compared to manual driving and especially compared to driving with production autopilot, but my expectations were perhaps more properly set by the Tesla’s verbiage and my experience with other beta testing programs.

Elon Musk has differentiated between FSD beta and production AP (which is still labeled beta). Just like so many others like Google’s gmail or 5-10 year long games and services in beta, production AP beta is available to all who pay for it. Purchasing FSD Capability or subscribing is for a product in the future that doesn’t require Safety Scores, etc.
 
  • Like
Reactions: legendsk
I hate to admit it, but after 1 full day of FSD beta, I have similar feelings. I am a huge Tesla fan, and am pretty happy with regular autopilot, as it is usually quite smooth and confident. My only grip with traditional AP is lane position in curves. It often reacts to the start of curves late, causing the car to drive toward the outside of the lane, away from the apex. This is especially concerning on curves to the right, as the late response often puts us much closer to oncoming traffic on the left. Other than that, I would rate it an 8/10 overall.

However FSD beta is nothing like AP. The mind of car view is amazing, perhaps leading to higher expectations of performance. However, I find FSD to be jerky and hesitant in actual execution. Its indecisiveness and unpredictability makes me extremely nervous and even a bit car sick. I have intervened several times, often with pretty dramatic steering or braking due to aggressive changes on course by FSD. That said, I will no longer be using this with passengers in the car, as it can be very unsettling. I understand this is in beta, but it feels much further off from full release than I had originally anticipated. I will occasionally use it by myself in order to monitor progress, but at this point, I will stick with manual driving for a while. I would rate FSD beta at 3/10 overall.
 
This is something I wish people who are obsessively clamoring for "single stack" Autopilot and FSD would understand.

I WOULD NOT want single stack Autopilot and FSD until FSD drives as predictably and consistently as Autopilot.

IMO they have refined Autopilot to a point where most of the time it drives better than me on highways. By this what I mean is the lane centering and follow distance is mathematical perfection and I don't worry about it suddenly swerving or trying to do something totally unexpected. But I remember a time Autopilot was not so elegant and look how good it is now and FSD will get to that point as well.

With FSD I'm so careful and vigilant ALL THE TIME. When there are cars around me, my hands are firmly on the steering wheel.

Put another way, Autopilot makes driving less stressful and FSD makes driving more "engaging." :)
 
As many have stated, expectations especially dictate how one reacts to FSD beta. It’s not what we paid for (I paid $3k, but longer ago than those that paid more), but since I knew that going in (it’s “limited early access beta”) it’s been an amazing experience so far.

Just like so many others like Google’s gmail or 5-10 year long games and services in beta, production AP beta is available to all who pay for it.

Autopilot is free to everyone now. Since you paid $5,000 for EAP you actually paid $8,000 for FSD beta. Still cheaper than what it is now. Judge FSD with that price tag. Will you get $8k worth out of it? My car is already 60k miles, but I am giving it to my kids so I assume it'll go over 200k miles. Hopefully FSD will work 100% of the time by then.

If I was to get a Tesla today, I would not buy FSD. I have it reserved with my Cybertruck, but will remove it when I place the order. I realized I like control too much. I even hate how NoAP make lane changes on the freeway. That feature has been in beta for a couple of years and only shown minor improvements.
 
Autopilot is free to everyone now. Since you paid $5,000 for EAP you actually paid $8,000 for FSD beta. Still cheaper than what it is now. Judge FSD with that price tag. Will you get $8k worth out of it? My car is already 60k miles, but I am giving it to my kids so I assume it'll go over 200k miles. Hopefully FSD will work 100% of the time by then.

If I was to get a Tesla today, I would not buy FSD. I have it reserved with my Cybertruck, but will remove it when I place the order. I realized I like control too much. I even hate how NoAP make lane changes on the freeway. That feature has been in beta for a couple of years and only shown minor improvements.

I wish they would make automatic lane changes optional/require permission in FSD like they do with NOA with an optional. Those who like it the way it is simply can not use that option.
 
  • Like
Reactions: ContrarianDC
I will no longer be using this with passengers in the car, as it can be very unsettling.
This brings up the question of how many miles of FSD testing Tesla will get out of each beta tester over time. That is, in the first week or two, I'm sure lots of beta testers will be trying out the feature; but if many of them become disillusioned, usage rates will likely drop, so Tesla may get fewer miles of FSD beta testing per car. Of course, if they need X hours of testing per week to feed to their neural net learning algorithms, they can compensate by entering more beta testers. I don't think this really directly affects us, except perhaps if the number of miles driven drops enough to affect how soon Tesla can come out with a production version.
Autopilot is free to everyone now. Since you paid $5,000 for EAP you actually paid $8,000 for FSD beta.
Comparisons like this are tricky to impossible, since Tesla has changed the price of its cars rather frequently. I don't recall every price change, of course, but I do recall one shortly after I bought my car (in March of 2019), in which AP became standard "equipment," but the base price of the car was raised by slightly less than what AP had cost before then. IMHO, the best approach to comparing prices across such changes is to look at the price of comparably-equipped cars -- but that's often impossible, since features come an go. There's no current vehicle to directly compare to my LR RWD Model 3, for instance -- at least, not in the US. (ISTR that Tesla is selling that configuration in China.)
 
Autopilot is free to everyone now. Since you paid $5,000 for EAP you actually paid $8,000 for FSD beta. Still cheaper than what it is now. Judge FSD with that price tag. Will you get $8k worth out of it? My car is already 60k miles, but I am giving it to my kids so I assume it'll go over 200k miles. Hopefully FSD will work 100% of the time by then.

If I was to get a Tesla today, I would not buy FSD. I have it reserved with my Cybertruck, but will remove it when I place the order. I realized I like control too much. I even hate how NoAP make lane changes on the freeway. That feature has been in beta for a couple of years and only shown minor improvements.
This is still the same misunderstanding. What we are getting with this early limited access FSD beta isn’t what we paid for. That’s why it’s not disappointing. We paid for not the existing features, but future features. When I paid $3k (the $5k EAP doesn‘t count unless you then consider us cheated for not being able to get AP for free like the buyers today, which I don’t), it had even less features. I knew I was paying for 3 things: 1) locking in the price to protect against price hikes, 2) helping to invest in Tesla’s future development of FSD (back in 2018 it wasn’t all that clear whether Tesla would make it as a company or ever be profitable), and 3) adding future features including full self driving (not full self driving beta). Some didn’t feel those 3 things were worth it, so they didn’t pay for it. Others, like myself, thought it was worth it/the risk. To each their own, but even now it’s the same: you aren’t paying for the current feature set. You can‘t even get the current FSD beta feature set just for paying the $10k. Tesla never says that you do. It’s only current production AP and NoA features, future features, and locking in the price.
 
It's a miracle that no one has gotten in accident yet...the freeway is great but just cant trust this beta
This is something I wish people who are obsessively clamoring for "single stack" Autopilot and FSD would understand.

I WOULD NOT want single stack Autopilot and FSD until FSD drives as predictably and consistently as Autopilot.

IMO they have refined Autopilot to a point where most of the time it drives better than me on highways. By this what I mean is the lane centering and follow distance is mathematical perfection and I don't worry about it suddenly swerving or trying to do something totally unexpected. But I remember a time Autopilot was not so elegant and look how good it is now and FSD will get to that point as well.

With FSD I'm so careful and vigilant ALL THE TIME. When there are cars around me, my hands are firmly on the steering wheel.

Put another way, Autopilot makes driving less stressful and FSD makes driving more "engaging." :)
it's a damn miracle with no accidents so far …I have had some close ones today I do not trust this build at all