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

FSD beta v9 release date: this Saturday?

This site may earn commission on affiliate links.
I recently watched one of the videos where it was on NOA in the city. It approached a stop sign and just ahead of the stop sign to the right, was a temp "Road Closed" barrier. The car approached the barrier, saw the sign, went around it to the proper lane that was open. It was an impressive display of the capabilities.

However..in that same video, its clear that a human driver would have handled it more "naturally", quicker, and would have never attempted to drive directly towards the "road closed" sign to begin with. Would have looked ahead, seen the sign, and simply..driven into the open lane adjacent to it at a quicker/more natural speed/process.

And that's what I personally think will occur in a lot of city driving miles: after the initial excitement wears off, after you show it off to friends and passengers to impress them? in real world use, drivers will tend to turn off the FSD in many city situations, and simply..drive themselves.
 
The vehicle I owned prior to our 2012 S was a Lexus LS. It was the first production vehicle with "self-parking" - which was pretty impressive at the time.

But...

To get the vehicle to park - it required moving a parking box around on the console display, placing the box on the rear camera image where you wanted the vehicle to go - and then letting it move around to get into that spot. Unfortunately, you could just drag the box to the right spot - you had to use directional controls to shift the box - which took much longer that the time required to do the parking.

So it was a fun feature to demonstrate to guests - but not very practical for actual use.

We've tried using Navigate on AutoPilot when driving highways - and have tended to go back to just lane keeping, since the FSD/AP software doesn't do the lane changes as well as a human driver - yet.

As the software gets more sophisticated through improved Tesla Vision and better rules (based on actual miles driven) - over time, I expect the FSD software to continue to improve - and at some point to be as good as or better than human drivers in typical driving situations.

But as Musk recently admitted - self driving is harder than he expected - and I also expect there will always be situation with our current FSD hardware when human drivers will do better - especially in unusual situations such as poor light conditions, rain/snow, ...

And that's OK - when we purchased FSD for our 2017 and 2018 vehicles - we didn't expect them to really self drive - but to provide much better driver assistance features - and to continue to improve during our ownership - unlike other vehicles that we could have purchased at the time - which would never get improved or new features...
 
The vehicle I owned prior to our 2012 S was a Lexus LS. It was the first production vehicle with "self-parking" - which was pretty impressive at the time.

But...

To get the vehicle to park - it required moving a parking box around on the console display, placing the box on the rear camera image where you wanted the vehicle to go - and then letting it move around to get into that spot. Unfortunately, you could just drag the box to the right spot - you had to use directional controls to shift the box - which took much longer that the time required to do the parking.

So it was a fun feature to demonstrate to guests - but not very practical for actual use.

We've tried using Navigate on AutoPilot when driving highways - and have tended to go back to just lane keeping, since the FSD/AP software doesn't do the lane changes as well as a human driver - yet.

As the software gets more sophisticated through improved Tesla Vision and better rules (based on actual miles driven) - over time, I expect the FSD software to continue to improve - and at some point to be as good as or better than human drivers in typical driving situations.

But as Musk recently admitted - self driving is harder than he expected - and I also expect there will always be situation with our current FSD hardware when human drivers will do better - especially in unusual situations such as poor light conditions, rain/snow, ...

And that's OK - when we purchased FSD for our 2017 and 2018 vehicles - we didn't expect them to really self drive - but to provide much better driver assistance features - and to continue to improve during our ownership - unlike other vehicles that we could have purchased at the time - which would never get improved or new features...
I get it/totally understand.

But after having FSD (paid $4k during a sale to upgrade from EAP (already on the car when I got it) to FSD) and experiencing it for over a year, combined with Elons' many tweeted promises ("NYC to LA with zero intervention by end of 2019", "the camera inside the 3 and Y is only going to be for Robotaxi service") that proved to be...less than accurate. Combined with the much hyped smart summon going up onto curbs during broad daylight at least 3 times for me, combined with "we will soon have self parking. Car will drop you off then go find a parking spot and park itself" promises from Elon.

Then add in that it really doesnt handle things in a "natural" manner, but more cautious/conservative/slow/hesitant or just the opposite (taking exits at excessive high speeds, then a fast hard braking mid corner). Or situations where its a 2 lane curvy road (broad daylight again) with a bike rider just ahead of me but close to right edge of road and the car would never pass. Just stay creeping behind at a safe distance. Until I turned off FSD and carefully went around him myself. Far too many times/situations with FSD where I ended up just..driving the car myself much of the time.

Exception being long highway trip using just the basic AP functions. Thus, I opted to not pay $10k/buy FSD for my pending order. Value is different for each person but for me? Not worth even remotely close to $10k
 
Last edited: