I think this is exactly what it is currently and elon jumped the gun on the price increase. Technically this "rewrite beta" means nothing currently, zero improvements to any random new buyers yet they are expected to pay $10k for it
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
FSD rewrite will go out on Oct 20 to limited beta
- Thread starter diplomat33
- Start date
I think this is exactly what it is currently and elon jumped the gun on the price increase. Technically this "rewrite beta" means nothing currently, zero improvements to any random new buyers yet they are expected to pay $10k for it
It's almost pointless trying to estimate/guess such a thing. Tesla themselves probably dont know, since this is new ground for everyone. However, software/AI systems like this tend to increase in reliability asymptotically, with rapid early progress that tails off as major problems are solved leaving only the edge-cases that may be much harder to solve. Right now I would say we are at the start of that curve and will make fast progress, if the beta videos are anything to go by.
The basic question, then, is how far along that curve do we need to be to meet and then beat the average human driver? My guess, given the poor quality of the "average" driver, is not very far. If that is the case, we may (repeat, *may*) see FSD exceeding average human performance in as little as 6-12 months. Technically, the issue has been: can the AI system create an accurate enough representation of the situation around the car for it to make safe and effective driving decisions? Many of the bets/posts on this forum (and others) have been skeptical (and I confess to some of that myself), but Tesla seem very close to cracking the vision issues necessary.
Sure, the car will probably remain stumped by some things, like complex road diversions, or missing road markings. But if it disengages and/or stops safely then I think FSD will be shown to be safe sooner rather than later.
Remember that agencies like the NHTSA are primarily interested in safety, not accuracy. They dont really care if the car goes round and round in circles, takes wrong turns, or just sulks in a parking spot .. it just has to do these things safely. It should also be remembered that right now, the car is probably better than humans at some things. I bet it's reaction time to stop for a pedestrian that runs into the road is far better than a humans.
This means that, while the car may lag behind humans in some intricate situations, its ability to just drive reliably mile after mile may allow it to beat out a human on average sooner rather than later. This is why so many posts here that focus on "the car wont be able to do X special case" miss the point .. the point is, can the car do better at things humans are bad at? And we are bad at stuff like running red lights, or forgetting to look both ways at a stop, or not seeing a car in our blind spots. Precisely the things that cause the majority of accidents, and precisely the things the car is going to be very very good at very soon now.
Last edited:
Tesla has been intentionally encrypting and obfuscating their software in increasing degrees so I think it would require an insider to really compare their neural network approach to Mobileye or any other competitor. I don't think that they're still labeling all their secrets and methods in plaintext for verygreen to tweet about. For all we know the hydranet could have been replaced with plaidnet, or rainbows_and_unicorns_net at this point without changing the name in a way we can see.
Ahhh... but WHEN can we get regulatory approve to remove side mirror and use cameras and the display; additionally replace the rear view mirror with a display display instead of say one that switches between. Or just remove it as well since it’s also on the display. BTW cameras are capable of offering a better view than just mirrors themselves.
It does seem quite likely Tesla's "private beta" terminology is quite intentional to avoid NHTSA scrutiny of a wide release even as a "Level 2" driver assistance feature. Unclear how far Tesla is allowed to push that boundary of how many "private" testers is too many for NHTSA then to investigate if it's putting the driver and more importantly others on the road at risk. From Elon's tweets, it looks like at least Tesla is tracking "intervention probability" before a wide release "dependent on proving high safety," so if early access testers are how they're collecting data to prove high safety, will Tesla be limited by the number of testers and will they need to cycle people through the program to get more variety of miles? And what's the thresholds anyway…
I think it's mostly development bandwidth .. once FSD is stable they will rework both Smart Summon and NoA to use the new vision stack and thus seamlessly integrate with City Streets (in fact, City Streets and NoA will become one feature at that point, from a technical standpoint).
powertoold
Active Member
I think so too. We've already seen multiple videos of the FSD beta driving into parking lots (like Target), where we can see all the cars and islands detected in the visualization.
Not much a difference whether or not you have FSD enabled in these situations. If you have FSD on and a toddler jumps out of the bushes, if you are paying attention you still have the same amount of time to react whether or not FSD is on.
DanCar
Active Member
Wrong. Please study your incorrect interpretation of English. Availability is definitely part of technology. As I already stated, if it is not available it is about as technology advanced as a rock. There are reasons Waymo only has 4 cars shuttling people around in a very small geographical region. Those decisions are part the technology. When Waymo expands next, will they have 8 cars running in a new location? Do we care about so few very expensive cars? Doesn't compare to a technology that in my estimation will be available to a much wider audience. And a big part of the reason is technology. The reason isn't because it is oh no just not available at the moment, it is a continual problem with foundational issues.
Last edited:
Mo City
Active Member
If they significantly increase the number of FSD beta testers, I doubt many (if any) of them will be public. There are enough videos being regularly posted to generate positive buzz on improvements. But at some point as the edge cases per tester decline, the number of NDA testers must increase. Perhaps proportionally.
Doing this would give Tesla the best of all worlds: Additional data and videos showing an improving product. I also believe they have (or will have) NDA testers spread all over the place.
There is so much at stake. I think Tesla's goal is reaching the leveling-off point near the top of the S-curve before blasting FSD out to everyone. Doubtful that is by EOY.
Not going to happen any time soon. FSD beta will be less than 40 cars for the next ~18 mos.
DanCar
Active Member
What makes you say that?
That probably already isn't true already. There are ~20 people with the FSD beta not covered by a NDA. There are surely at least that many covered by a NDA, and then there are the Tesla employees testing FSD. We have no idea how many cars are participating in the beta. My guess is that it is in the hundreds, if not thousands. (Unlikely to be thousands at this point.)
Just want to add that there are some things that accelerate. The number of miles driven per day autonomously will grow exponentially with the reliability, the amount of data will grow exponentially with that, the amount of money available for research and development will grow exponentially with it also etc. So expect some diminishing returns and some accelerated returns making it very complex and hard to predict.I imagine it will look like this.
The rate of improvement starts off fast but then slows down as you get closer to 100%.
Many a true thing said in jest.....
Since Tesla need to PROOVE that FSD is safer than human by a decent margin, you could easily see discounts applied the more miles you have FSD engaged.
Slightly difficult for Tesla, might need to distinguish between highway and city performance of FSD. Millions of freeway miles without disengagements doesn't represent city performance and kinda annoying to have to focus specifically on humans still being safer around town.
I've lost track of what disengagements get reported. Are they already catagorized?
Scary stuff.
Once it gets down to 1 intervention a month how much attention do you think the crash test dummies, I mean beta testers will pay? One of them is going straight into the back of a truck at 70 MPH.
And updates every week? So they get used to it safely navigating a section of road, stop paying attention at that point and then one day the updated code decides to drive into oncoming traffic. Better hope it's not you coming the other way.
True. I agree Tesla will likely be first at a widely deployed FSD. The question is how advanced will it be. IMO, it is less impressive to deploy FSD wide that requires driver interventions every 50 miles than a wide deployment of FSD that is true driverless for example. For me personally, I am more impressed with limited deployment of driverless FSD than I am of wide deployment of FSD that requires constant driver supervision. For me, driverless FSD is the true prize, not how many cars you put FSD on.
They are deep questions.
Do you not already have electronic mirrors? There are several / many implementations from various manufacturers in Europe.
Remember 'your entertainment system failing (MCU1) not a safety issue' except if it effects demisting, wiper control, light control, backup camera etc etc?
Personally I think there is a good case for retaining old fashioned reflective mirrors for as long as there could be a human driver. The rear view mirrors with integrated monitor that I have seen do have some advantages, but also drawbacks that on balance leave the old fashioned mirror with plenty of value.
Of course no harm at all adding birds eye and other cool display / camera features, but with cameras still prone to problems like b-pillar cam condensation, I don't want to drive a car that relies on them.
Similar threads
- Question
