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

FSD rewrite will go out on Oct 20 to limited beta

This site may earn commission on affiliate links.
To be what possibly shows this is the rebuild / rewrite is that established functionality that is unreleased to the surface street turns have changed in performance; and in some cases somewhat backwards. This would be expected in a complete rebuilt that some items not be 100% to what they were before a rebuilt. Regardless city turns is what I’ve been happily waiting for. If this is the rebuild all the better even if it’s a temporary step back in some areas. At this point almost no one has the limited beta. By my estimates half dozen people but it could be 20, 100, 1000 when knows. But there are only about a half dozen talking and showing what they have.

Just (somewhat) patiently waiting until the beta or next release with city turns hits my FSD AWD LR.
 
I don’t know about you guys, but I’m in it for the long haul. Our Tesla’s have the potential to last quite a while without major repairs.

They seem to have nailed the 3D (or I guess 4D with time) perception problem with the cameras. I’m curious to see how that visualization looks like when it snows though.

So now the planning looks to be really jumpy/unstable at the worst times.
Elon said the re-write was putting the perception, planning and object-detection in software 2.0 during Part 2 of the Third row interview. (YouTube) I imagine he wasn’t outright lying about that.

It may take them 5 years and another complete re-write (end-to-end learning of the driving task & maybe even that MuTwo self-learning simulator idea George Hotz hinted at) but that’s ok for me.
As long as it’s getting better and rapidly it will bring value to our lives.
We can’t sleep in the car but monitoring is still way less tiring than continuously applying steering wheel micro torque adjustments.
IMO, one major benefit that doesn’t get talked about is how entertaining autopilot makes our drives.
It’s amazing to watch my stupid car try to drive, keeps me awake and alert since it cannot be trusted. I can also appreciate that there are parts of my drive I know it can handle 100% like slow traffic & being alone on the highway; my mind can relax at these times.
This brings such benefit to everyday life and watching it try to handle city will be even more incredible.

so are they going to add the 4D re-write to highways too? My car still brakes when there are aggressive dips on the highway and I suspect it’s because NoA assumes the road is flat(2D), we can see the predicted lines on the display squish together when the front dips down.
 
They've made a cheaper version of a useless thing. :p
Monitoring FSD seems like it will be just as much work as driving. It will be interesting to see how many people actually use it and how safe it turns out to be. Personally I'd rather have the machine monitor me while driving and catch my mistakes than the other way around.
Mobileye does not use radar on their vision based AV prototypes.

I have to respectfully disagree with your second sentence. Monitoring a system is not anywhere near as much work as doing it all yourself. I spent thousands of hours early in my career hand flying commuter aircraft, shooting approaches to minimums in all kinds of weather, no autopilot whatsoever. When I was able to upgrade to an aircraft with an autopilot, my workload was greatly reduced, although I still had to monitor the autopilot, you now had more time to widen your view and give more attention to other important matters of the aircraft systems, weather, traffic, etc. etc., because the autopilot would not let you stray away from heading and altitude when not paying direct attention to that. And bust an altitude or heading assigned can get you in real trouble with ATC, if not lead to an accident. Using autopilots is a much safer, and easier proposition, and less stressful, but it still requires FULL oversight, it’s a machine that can fail suddenly, or do something unexpected.

I recently took my first long drive in my Tesla from Florida to NY and back. Used FSD the entire way. Compared to other vehicles on long drives, I found driving a Tesla was far less stressful than any car I have ever owned. I drove twenty hours over two days, and was not nearly as tired as in my ICE car over the same drive. It’s much safer, much less tiring. Using FSD properly is nowhere near as much work as hand driving. The only time I would turn off FSD was when I had no car to follow, and somebody decided to get right on my tail, (phantom braking, which has happened more than once to me). These are workload reduction systems, and they do reduce your workload substantially, if only in the lane keeping assist, and TACC. And get into a two mile long stop and go traffic jam, well that’s a piece of cake in a Tesla.

As for in town FSD, yes, you will still have to monitor it closely, and be ready to intervene, but I believe overall it will be quite a bit less work than actually hand driving.

Having said that, I agree with your idea of the car monitoring me for safety’s sake. The combination of automation and human oversight is the safest and least stressful in my view.
 
Last edited:
So now that the FSD "rewrite" costs is $10k, which makes it official (L2, L3, L4... ?), when does the FSD rewrite will be available?

If I was paying an extra $10k option for a new car today, and possibly getting my car next week, I would expect having the new FSD version (11 ?).
Or should I have to wait for another week(s), month(s), year(s)...???.

I felt that this is a one way contract where the customer paid first, and the delivery content and date are both 'whatever'...

This is a good question, and I think a lot of people would like a clear answer.

I think Tesla didn’t give a clear answer because they want to hit a specific level of maturity/stability/safety before wide rollout, and they don’t know how long that will take.

I think Elon said something about a month or so, but I haven’t found that again. A couple articles suggested the Beta group would get wider within a couple weeks but still be a limited set - not sure where that came from.

By the end of the year has been a common promise for a while now, which might have more to it than past end of the year promises this time.
 
  • Like
Reactions: Watts_Up
Not an expert on anything, but I’d like to add a comment. I’m 80 years old and just helped an 80 year old friend drive her Mercedes 900 miles over two days. We were both exhausted, which surprised me as I has just driven my Tesla on Autopilot 720 miles by myself in one day. I finished the drive not tired at all. Autopilot, right now, is a huge boon to me. It is very relaxing to just sit there looking for the occasional problem down the road or put the car to following a truck through a busy city. I would happily take what they have now if they would give it to me.
 
And in April at the earnings call, Elon said end of year 2020. Tesla confirms self-driving as a subscription service but brace yourself for the price - Electrek

So I think it's fair to assume that when a target date shifts from end of 2020 to 2021, it's probably shifted to early 2021.
I am still hoping for end of 2020 for a simple reason that it is still on the Model Y order page :)
upload_2020-10-30_10-37-35.png
 
I feel like this is going the wrong direction. 10k is just too much for the average person to spend on an accessory when buying a car.

If I was buying new right now looking at the price go from 50 to 60 is a huge nope.

Tesla needs to have a black friday sale. FSD 6k order now. They'd sell like crazy now that everyone is afraid it's out of reach.
 
  • Like
Reactions: Watts_Up and DanCar
I feel like this is going the wrong direction. 10k is just too much for the average person to spend on an accessory when buying a car.

If I was buying new right now looking at the price go from 50 to 60 is a huge nope.

Tesla needs to have a black friday sale. FSD 6k order now. They'd sell like crazy now that everyone is afraid it's out of reach.
I think the idea is to compare the FSD price with the salary of a private driver. From that perspective, it's very affordable. I don't think they look at it as an accessory.
 
People are talking about FSD as if it'll be on full time. I don't envision many people using it full time until it becomes a true set-it and forget-it technology, or they drive in locations that it can easily handle. Instead, they'll be more inclined to use it where they know it works well, until it improves. My city layout is simple, so it will most likely take me from my driveway to the parking garage at work without much fuss (~11 miles). I could see maybe 1 disengagement where traffic merges because of construction. But even so, I don't see myself using it everyday.

As for monitoring, regardless of the actual capability of the system, people aren't going to trust it outright at first. So, having to monitor it as it improves isn't going to be the deal breaker some thinks it is. How long it takes that trust level to rise will depend on how quickly the system improves.
 
  • Like
Reactions: Snerruc
10k for true hands off FSD where i can have the car come pick me up and drive me places. worth it
10k for current FSD which is still only to a handful of beta testers and still requires constant supervision/nags. not worth it

But that is just my opinion. I'm optimistic about FSD, but 10k is a big pill to swallow for new features that I can't even get after paying 10k.
 
I have to respectfully disagree with your second sentence. Monitoring a system is not anywhere near as much work as doing it all yourself. I spent thousands of hours early in my career hand flying commuter aircraft, shooting approaches to minimums in all kinds of weather, no autopilot whatsoever. When I was able to upgrade to an aircraft with an autopilot, my workload was greatly reduced, although I still had to monitor the autopilot, you now had more time to widen your view and give more attention to other important matters of the aircraft systems, weather, traffic, etc. etc., because the autopilot would not let you stray away from heading and altitude when not paying direct attention to that. And bust an altitude or heading assigned can get you in real trouble with ATC, if not lead to an accident. Using autopilots is a much safer, and easier proposition, and less stressful, but it still requires FULL oversight, it’s a machine that can fail suddenly, or do something unexpected.

I recently took my first long drive in my Tesla from Florida to NY and back. Used FSD the entire way. Compared to other vehicles on long drives, I found driving a Tesla was far less stressful than any car I have ever owned. I drove twenty hours over two days, and was not nearly as tired as in my ICE car over the same drive. It’s much safer, much less tiring. Using FSD properly is nowhere near as much work as hand driving. The only time I would turn off FSD was when I had no car to follow, and somebody decided to get right on my tail, (phantom braking, which has happened more than once to me). These are workload reduction systems, and they do reduce your workload substantially, if only in the lane keeping assist, and TACC. And get into a two mile long stop and go traffic jam, well that’s a piece of cake in a Tesla.

As for in town FSD, yes, you will still have to monitor it closely, and be ready to intervene, but I believe overall it will be quite a bit less work than actually hand driving.

Having said that, I agree with your idea of the car monitoring me for safety’s sake. The combination of automation and human oversight is the safest and least stressful in my view.
In my opinion beta city FSD is like using autopilot with 1% of human skill to fly a Blue Angels show. It probably wouldn't make it easier even it had 50% human skill. I don't think it's analogous to airplane autopilot like Autosteer and TACC (and even that analogy has limitations since there are way more thing to run into on the ground than in the air).
 
  • Like
Reactions: P3dStealth
10k is a big pill to swallow for new features that I can't even get after paying 10k.

It's the most capable feature available in any consumer car... Super Cruise costs an extra $6500 on the middle trim, and extra $2500 on the top trim, and I think there's also a monthly fee at some point. You also can't buy Super Cruise on the base trim Cadillacs.

And Super Cruise has no future potential.

Cadillac Super Cruise Pricing Starts At $2,500 On 2021 Escalade
 
It's the most capable feature available in any consumer car... Super Cruise costs an extra $6500 on the middle trim, and extra $2500 on the top trim, and I think there's also a monthly fee at some point. You also can't buy Super Cruise on the base trim Cadillacs.

And Super Cruise has no future potential.

Cadillac Super Cruise Pricing Starts At $2,500 On 2021 Escalade

I already have Autopilot that I got for free which is obviously better than supercruise.

Again, I'm not hating on FSD. It just doesn't make sense for me personally to pay 10k for it.
 
10k for true hands off FSD where i can have the car come pick me up and drive me places. worth it
10k for current FSD which is still only to a handful of beta testers and still requires constant supervision/nags. not worth it

But that is just my opinion. I'm optimistic about FSD, but 10k is a big pill to swallow for new features that I can't even get after paying 10k.

Unfortunately, it turns into a bit of a gamble. By the time FSD is actually capable of hands free driving, it'll surely cost more than $10k. Will hands free robo-chauffer be worth $15k or $20k? Or is it better to bet it'll actually happen, and pay $10k now. Personally, I made that bet in 2018, when it was $3k over EAP (which was a $5k no brainer)*. I suppose if you're confident FSD will be a thing while you still own your car, it's something to consider.

* Of course if I put that $3k in TSLA instead...
 
Last edited: