Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
The next big milestone for FSD is 11. It is a significant upgrade and fundamental changes to several parts of the FSD stack including totally new way to train the perception NN.

From AI day and Lex Fridman interview we have a good sense of what might be included.

- Object permanence both temporal and spatial
- Moving from “bag of points” to objects in NN
- Creating a 3D vector representation of the environment all in NN
- Planner optimization using NN / Monte Carlo Tree Search (MCTS)
- Change from processed images to “photon count” / raw image
- Change from single image perception to surround video
- Merging of city, highway and parking lot stacks a.k.a. Single Stack

Lex Fridman Interview of Elon. Starting with FSD related topics.


Here is a detailed explanation of Beta 11 in "layman's language" by James Douma, interview done after Lex Podcast.


Here is the AI Day explanation by in 4 parts.


screenshot-teslamotorsclub.com-2022.01.26-21_30_17.png


Here is a useful blog post asking a few questions to Tesla about AI day. The useful part comes in comparison of Tesla's methods with Waymo and others (detailed papers linked).

 
Last edited:
Some of the items discussed in the past three pages of this thread are many years old, and remain unresolved. No passes here. We do get a regular allotment of new games, however. Priorities are twisted, it seems.
I've said this before - the resources to create games and AI are not the same.

The basic mistake you are making is - assuming that if only they put in the effort they can easily and quickly solve it. May not be true ...
 
  • Disagree
  • Like
Reactions: Yelobird and jebinc
Good thing medicine doesn’t act like technology. “We’ve got this new cancer drug that uses a different approach. It’s not as good as the old drug but we’re pretty confident with some development. It’ll eventually be at least as good. Some day. “
There are actually a lot of experimental drugs for cures not yet available ... or doing it differently but less accurately.

Good example is blood glucose sensor. Yes, you can prick your finger and get (somewhat accurate) numbers. But its a major pain ... ;)

So, we get all kinds of sensors you put on and get continuous readings with no pain / effort. But they are all less accurate ... should we wait until the sensors get as accurate as finger prick ? Why not ?

Similarly for simple reading of temperature. A mercury thermometer under the tongue may be the most accurate - but a IR scan of the forehead is so much easier, even if less accurate.

You have to only think a little to find such examples in the medical field.
 
  • Disagree
Reactions: jebinc
I've said this before - the resources to create games and AI are not the same.

The basic mistake you are making is - assuming that if only they put in the effort they can easily and quickly solve it. May not be true ...
Well, then they shouldn’t implement (sh)it until it’s better tested (FSD aside). Implementing TeslaVision and pulling Ultrasonic sensors is yet another good example of this never-ending BS. Clearly Telsa priorized cost savings over finishing/testing the replacement. We agree to disagree - fine. To each. 🥂
 
There are actually a lot of experimental drugs for cures not yet available ... or doing it differently but less accurately.

Good example is blood glucose sensor. Yes, you can prick your finger and get (somewhat accurate) numbers. But its a major pain ... ;)

So, we get all kinds of sensors you put on and get continuous readings with no pain / effort. But they are all less accurate ... should we wait until the sensors get as accurate as finger prick ? Why not ?

Similarly for simple reading of temperature. A mercury thermometer under the tongue may be the most accurate - but a IR scan of the forehead is so much easier, even if less accurate.

You have to only think a little to find such examples in the medical field.
This is categorically false. My son is a type 1 diabetic. Finger prick testing is/was the gold standard for years. Today’s CGM sensors are as good, and don’t even require finger prick calibration. Thinking Dexcom and Libra 3 here. You can take that to the bank. I feel much safer with these sensors than I ever will with current Tesla technology implementation.
 
This is categorically false. My son is a type 1 diabetic. Finger prick testing is/was the gold standard for years. Today’s CGM sensors are as good, and don’t even require finger prick calibration. Thinking Dexcom and Libra 3 here. You can take that to the bank. I feel much safer with these sensors than I ever will with current Tesla technology implementation.
Libre 3 user here.

The ones that don't fall out (~65%) are so inconsistent as to be outright dangerous. Reading 40 points low prompting me to eat sugar pills, when if fact I'm already too high. Worse than using no meter at all.
 
Well, then they shouldn’t implement (sh)it until it’s better tested (FSD aside). Implementing TeslaVision and pulling Ultrasonic sensors is yet another good example of this never-ending BS. Clearly Telsa priorized cost savings over finishing/testing the replacement. We agree to disagree - fine. To each. 🥂
Again you are making the mistake of thinking I agree with what Tesla did. I'm just giving an explanation of "why" ...

If you go back and see my posts on ultrasonic (and aradar) removal you will see I put that down to cost cutting / parts shortage.
 
Libre 3 user here.

The ones that don't fall out (~65%) are so inconsistent as to be outright dangerous. Reading 40 points low prompting me to eat sugar pills, when if fact I'm already too high. Worse than using no meter at all.
Not our experience. Three users of them in this household, fall out aside. 65% fall out rate and that inaccuracy amount - something else was be going on as that is not the normal, but rather the exception.
 
This is categorically false. My son is a type 1 diabetic. Finger prick testing is/was the gold standard for years. Today’s CGM sensors are as good, and don’t even require finger prick calibration. Thinking Dexcom and Libra 3 here. You can take that to the bank. I feel much safer with these sensors than I ever will with current Tesla technology implementation.
As a T2D I've been using Libre for a number of years and I've done a TON of research on this. You simply haven't done the research. Even finger pricks are about 10% off .... Libre 3 too. I've also compared Libre with the lab tests ...

In anycase, I think even you get the point when you says *today* CGM sensors are as good - you accept they were not good when they came out. BTW, Libre asks you to check with finger prick to confirm (say low blood glucose).

Both the Dexcom and Libra 3 are FDA certified and have +/- accuracy of x, within a given typical range of blood sugar values.
Yes - but do readup what they say is their error rate. I'm happy FDA has certified - otherwise I'd not be able to get them and I absolutely refuse to finger prick daily.
 
If you go back and see my posts on ultrasonic (and aradar) removal you will see I put that down to cost cutting / parts shortage.

Precisely. Around the time Tesla removed the USS, there were Fords piling up in parking lots because they couldn't adapt to the chip shortage: https://www.motortrend.com/news/2022-ford-pickup-trucks-f150-chip-shortage-kentucky-speedway/

When the alternative is ceasing selling vehicles entirely, removing the USS looks like a smart decision.
 
As a T2D I've been using Libre for a number of years and I've done a TON of research on this. You simply haven't done the research. Even finger pricks are about 10% off .... Libre 3 too. I've also compared Libre with the lab tests ...

In anycase, I think even you get the point when you says *today* CGM sensors are as good - you accept they were not good when they came out. BTW, Libre asks you to check with finger prick to confirm (say low blood glucose).


Yes - but do readup what they say is their error rate. I'm happy FDA has certified - otherwise I'd not be able to get them and I absolutely refuse to finger prick daily.
Yes, I agree early CGM sensors sucked more than Tesla auto wipers!
 
I do not believe that’s why Telsa removed USS (Covid supply chain reasons), or even radar for that matter. I think it was a pure profit padding play. Sure, they may have claimed Covid, as doing so was “in Vogue” at the time - but reality is likely more to cover for TeslaVision - which was absolute crap at that time. Today, TeslaVision has improved some, to “polished turd” levels, IMHO.
 
I do not believe that’s why Telsa removed USS (Covid supply chain reasons), or even radar for that matter. I think it was a pure profit padding play. Sure, they may have claimed Covid, as doing so was “in Vogue” at the time - but reality is likely more to cover for TeslaVision - which was absolute crap at that time. Today, TeslaVision has improved some, to “polished turd” levels, IMHO.

How does removing USS "cover" for Tesla vision? Surely if they wanted to make up for the faults in Tesla vision, they would install as many additional sensors as possible.
 
  • Disagree
Reactions: jebinc
How does removing USS "cover" for Tesla vision? Surely if they wanted to make up for the faults in Tesla vision, they would install as many additional sensors as possible.
Let me try to make it more clear for ya…

Premise:

1. Tesla removes USS and implements TeslaVision

2. Because TeslaVision was incomplete (and sucked) they blame the removal of USS (and Radar) on COVID - Using COVID as an excuse for pulling USS (prematurely) and implementing TeslaVision that was widely criticized as being way undercooked, but served anyway.
 
Let me try to make it more clear for ya…

Premise:

1. Tesla removes USS and implements TeslaVision

2. Because TeslaVision was incomplete (and sucked) they blame the removal of USS (and Radar) on COVID - Using COVID as an excuse for pulling USS (prematurely) and implementing TeslaVision that was widely criticized as being way undercooked, but served anyway.

You're not explaining why Tesla removed the USS in the first place. We're saying Tesla did it to reduce part count during the chip shortage. You're saying they're falsely blaming the chip shortage on the removal. But you're not explaining why you think they removed them.

If they didn't need to remove the USS and just blamed COVID as a "cover", then why on earth did they remove them?
 
You're not explaining why Tesla removed the USS in the first place. We're saying Tesla did it to reduce part count during the chip shortage. You're saying they're falsely blaming the chip shortage on the removal. But you're not explaining why you think they removed them.

If they didn't need to remove the USS and just blamed COVID as a "cover", then why on earth did they remove them?
“Asked and answered”.

Please reread my OP. Not going through this “CJ” again with you as I did yesterday - just move on…
 
I got the opportunity to do my 120 mile round trip camping run again this weekend. The first with 11.4.8.1
Again it was some good, some not so good, some bad.
First was unexpected. Over both sides of the trip we had maybe three dry wipes, couldn't tell how it did in rain due to the lack thereof. I'll count that as an improvement.
Similarly to my commute, overall FSD felt confident and controlled everywhere except where it lost its mind.
Still loves being in the left lane on an empty road and still takes ages to get back into the right lane.
Still slows down for flashing yellow warning lights but now doesn't offer any explanation, it just does it.
Making dives for occasional short exit lanes was MUCH worse and much more aggressive and uncomfortable - and this with it on chill.
This was a definite regression from the previous versions.
It almost made an odd junction but freaked out half way. Going from left->right on Carl Parker Blvd SE on the exit to Rices Crossing Rd, then turning right onto S Main. It gets to 68mph on Carl Parker and slows downs gently before ending up aggressively braking when it figures out it has a sharp right/left to get onto Rices Crossing and craps out on the left-hander part. I took over there and re-engaged for it to handle the right onto S Main. Most uncomfortable and another regression.
Screenshot 2023-06-09 at 16.16.02.png
 
...and behaves the same.
ok, so we all agree the auto wipers dont work (though they USED to for me at least). Rather than complaining here, why dont we all get on Twitter X and tweet bomb Elon? As in, you know, try to do something rather than just complain to each other?

Suggested text: "@ElonMusk How about auto wipers that work? We know FSD is tricky but isnt it a bit embarrassing that auto wipers (that USED to work fine) are so broken right now? They come on during sunny days, and stay off during a rainstorm."