FSD Beta 10.11

[Kyrne]

Teslafi showed two batches of cars. First 230 and then 30 more. After that nothing.

You have to look at installed + pending, since some people install later.

Someone updated from a non-fsd version to an fsd version. Seems like a new enrollment.

 

[Sigma4Life]

Someone updated from a non-fsd version to an fsd version. Seems like a new enrollment.

Same version I’m on. Hopefully that’s a good sign that they are ready to expand the beta. Odd that it’s exactly 1 though. Just as easily could be an employee (or friend) who’s car was manually added to the beta.

 

[EVJAutoJunkie]

Teslafi showed two batches of cars. First 230 and then 30 more. After that nothing.

You have to look at installed + pending, since some people install later.

I'm specifically referencing installed updates which are identified by VIN #s. The VIN # history shows all previous updates versions for each car, therefore, it's easily to identify which cars were freshly enrolled in FSD Beta.

 

[mjolley9]

Someone updated from a non-fsd version to an fsd version. Seems like a new enrollment.

Yep - It shows the car was manufactured in the US but the region is "Russia"....super odd.

 

[generalmilk]

Yep - It shows the car was manufactured in the US but the region is "Russia"....super odd.

Omg, Elon weaponizing cars confirmed!

 

[fcastrillo86]

Omg, Elon weaponizing cars confirmed!

 

[EVNow]

I'm specifically referencing installed updates which are identified by VIN #s. The VIN # history shows all previous updates versions for each car, therefore, it's easily to identify which cars were freshly enrolled in FSD Beta.

You are talking about this .... I'd not go by this one example. Not sure what this is, given the region is Russia. BTW, Tesla has included some influencers as beta testers once in a while. Could be just some special invitation.

 

[EVNow]

New hypothesis :

Tesla had not finished Canada release of FSD Beta last week. That’s why Elon gave a later date.

Now with stalling of 10.11 rollout - Tesla is going to combine the fix with Canada release and both will come out together this weekend / next week.

ps :

https://twitter.com/x/status/1504601385886785543

 

[EVJAutoJunkie]

Would a Tesla FSD Beta programmer please stand up...​

We welcome you to spill a bit of tea about software release dates with us. Even if you happen to be a cousin of a cousin's dog walker who happens to have access to software project manager's meeting notes. We promise not to tell Elon.

 

[Dr Gez]

I had FSD on my last model S but now waiting 3 weeks on my new model S (2021.43.201) with 100 safety score. Would be nice if the FSD acceptance followed you into the new car, right?

Do you have the FSD Beta Button on 2021.43.201? I've been stuck on 2021.43.201 for 6 weeks since delivery & don't have the button so can't even do my safety score

 

[hamoneaster]

Getting Tesla to meet a date is like trying to nail jello to a tree

I wish someone would nail me. Jello out here getting nailed for no reason

 

[Ramphex]

I wish someone would nail me. Jello out here getting nailed for no reason

Makes me want to try nailing jello to a tree, I feel like I could do it

 

[FelixORiley]

New hypothesis :

Tesla had not finished Canada release of FSD Beta last week. That’s why Elon gave a later date.

Now with stalling of 10.11 rollout - Tesla is going to combine the fix with Canada release and both will come out together this weekend / next week.

ps :

https://twitter.com/x/status/1504601385886785543

I got super excited when I got a notification a few hours ago that I had an update ready…. and then it was 2022.4.5.17.

If beta 10.11 is still possible, I don’t want to update past 2022.4.5.15, but can a different update get pushed once my car downloaded one already?

 

[Sigma4Life]

I got super excited when I got a notification a few hours ago that I had an update ready…. and then it was 2022.4.5.17.

If beta 10.11 is still possible, I don’t want to update past 2022.4.5.15, but can a different update get pushed once my car downloaded one already?

That’s the scenario I’m concerned about. I don’t want an update that pushes my car betond the latest beta build. You might be ok though since I suspect 10.12 (or 10.11.1) is imminent and will be used for wide release.

What’s your safety score and how long have you had it?

 

[JulienW]

Makes me want to try nailing jello to a tree, I feel like I could do it

Simple, but you must use Sugar Free Strawberry for it to work.

 

[FelixORiley]

That’s the scenario I’m concerned about. I don’t want an update that pushes my car betond the latest beta build. You might be ok though since I suspect 10.12 (or 10.11.1) is imminent and will be used for wide release.

What’s your safety score and how long have you had it?

I’ve been 98+ since the end of Jan. Currently 99.

 

[aronth5]

And supercharging opening dates. Often by years, at least in and around my area.

At least the pace of installations has increased significantly. Tesla is averaging a completeted supercharger installation every 1.75 days in the US this year. 82 now in construction plus 182 permits. I remember when permits totaling over 20 was something to get excited about.

Source
supercharge.info

 

[aronth5]

Simple, but you must use Sugar Free Strawberry for it to work.

View attachment 782950

Are you sure that is just not a jello tree? Looks like harvest time to me.

 

[nvx1977]

My bottom line is the car cannot see as well to the left or right as I can when stopped at a blind intersection. The root cause has nothing to do with creeping since that can be corrected via software and I'm sure Tesla will improve it. Think physics "first principles" which Elon is well known for. What is the only thing with the current camera layout we know for sure. Camera specifics including viewing angle. Tesla's FSD engineers have to work within those constraints. Unfortunately I believe Tesla failed to use first principles correctly when locating cameras. Imagine how much creeping would be improved with cameras closer to the front. A lot.

I'm simply a better/safer driver in this particular instance. Tesla will get better but there is only so far you can creep into the intersection before you have to stop and make the decision to go or not. When I'm at a really blind intersection I will lean forward above the steering wheel which gives me about 30 inches of additional vision. I do this every time I leave my house when approaching a blind corner.

I tend to think it's that Tesla is not able to blend front and B-pillar camera data into a cohesive picture (so far), and it's possible that when the cameras individually don't agree with each other, for safety, the car doesn't go. For example, if in my example where I'm turning RIGHT at a T, and the car is already facing toward the right, the B-pillar data shows OK to go, but front camera is saying I can't see sh*t, the car doesn't go.

We've already seen that the car likes to approach intersections straight on even if the route shows a right turn. This to me indicates that the front camera is relied upon way more (if not exclusively) for decisions.

Not sure you can cite first principles failure for the placement of the cameras, as we don't have the full picture of what constraints there might have been for placing side cameras on the front of the car. What were the tradeoffs? I'm not disagreeing that cameras further up front provides better visibility for blind intersections. I'm just more cautious about being critical of their current placement because I don't know the original constraints up front.

 

[sleepydoc]

How'd a car in Moscow get 10.11 at all? I thought FSD was still limited to the U.S.?

I tend to think it's that Tesla is not able to blend front and B-pillar camera data into a cohesive picture (so far), and it's possible that when the cameras individually don't agree with each other, for safety, the car doesn't go. For example, if in my example where I'm turning RIGHT at a T, and the car is already facing toward the right, the B-pillar data shows OK to go, but front camera is saying I can't see sh*t, the car doesn't go.

We've already seen that the car likes to approach intersections straight on even if the route shows a right turn. This to me indicates that the front camera is relied upon way more (if not exclusively) for decisions.

Not sure you can cite first principles failure for the placement of the cameras, as we don't have the full picture of what constraints there might have been for placing side cameras on the front of the car. What were the tradeoffs? I'm not disagreeing that cameras further up front provides better visibility for blind intersections. I'm just more cautious about being critical of their current placement because I don't know the original constraints up front.

Once the cameras are calibrated, it should be a relatively straightforward transformation calculation to merge the two camera images.