March of 9s: The Beginning of the End

I have had enough experience now with FSD 12 to say that in my opinion, I believe Ashok is correct:

https://twitter.com/x/status/1774673326583210476

I would also stress we are at the beginning of the end. That is, we will hopefully now start to see significant improvement with the "known FSD issues" and "corner case disengagements" with every major FSD release. "The end" (level 3 /4 autonomy within a wide Operational Design Domain (ODD)) is almost certainly year+ away. I'd like to document, just from one consistent anecdotal use case that can be repeated over time, where we are "starting from."

I will be driving a 90ish mile (2-3 hrs) "loop" under FSD to cover a range of driving scenarios. (Mileage is approximate):

For privacy reasons, there are about 15 miles in my loop that are not included in the link above that takes me to/from my actual home.

Here is a Link to inspect the route in detail

This route takes me from NJ Suburbs into and out of Manhattan (NYC) and includes approximately:

• 10 Miles Suburban driving
• 65 Miles "limited access highway" driving (This should be using the FSD 'highway stack' which is not the same as the FSD 12 stack)
  • includes interchanges
  • Includes Tolls
  • Includes Tunnel
• 8 miles of other "highway type driving", (will probably fall under the FSD 12 stack)
• 6 miles of dense city driving...including areas around Times Square, Rockefeller center, etc which will have dense vehicle and pedestrian traffic.

I will not be recording with a phone or anything like that. However I will try and save dashcam footage of anything notable.

I will report on:

1. Interventions (Accelerator presses, particularly if safety related)
2. Disengagements (comfort or safety related)
3. Overall impressions

As we know Version 12.3.x does not support...(but will need to in the future):

1. Smart Summon or "Banish"..so what I call the "first 100 yards and final 100 yards" is not available to test. (Drop offs / pick-ups).
2. "Reversing" while on FSD is not yet supported

Finally, there are what I would say 2 well documented "comfort / safety" issues with FSD 12.3.x that I have also experienced regularly first hand:

1. "Lane Selection Wobble"...for example, approaching an intersection where the single driving lane splits into multiple lanes (turning vs. straight)...FSD may act "indecisively"
2. Unprotected turn (stop sign) behavior. Notably: stops early....then creeps. If no cars detected it may creeps into intersection instead of "just going". Further, if it has creeped into the intersection, THEN detects a car approaching, it may still hesitate and require intervention (accelerator press) to get it going.

In addition to those two consistent issues, I expect to encounter some issues related to routing, and any number of other 'corner case' issues. All things that will ultimately need to be handled, but we expect to see dealt with as we progress though the "March of 9s"...toward the "end of the end".

Although I have driven FSD regularly over the past 3 weeks...I have yet to take it into NYC.

Vehicle: Refresh Model S (2023), Vision Only, HW4. First test will be using:
Firmware version: 11.1 (2024.3.15)
FSD Version: 12.3.4

So...there's the set-up. I expect later today to drive the first loop.

 

[zoomer0056]

robotaxi

Tesla Cybercab

Tesla Reveals Robotaxi App and Names the Robotaxi the CyberCab

Tesla prepares to unveil its ride-hailing product Robotaxi or, as Musk called it, Tesla CyberCab.

www.notateslaapp.com

 

[Joey D]

Again, they resumed them,..

Exactly my point.

This is all to say that again, we can agree to disagree. This is by no means "not debatable."

I'll see your "they took our jobs" meme, and raise you several blocks of people lined up to get a free ice cream cone or slurpee.

 

[spacecoin]

As I said in the investor thread, RT needs to be significantly cheaper than Uber to take market share. There are many who dislike EVs and ADAS out of being extremist idiots.

It doesn't have to be cheaper. It can be better in other ways, and at the same cost.

In a Waymo you get:
1) A premium car
2) You play your own music
3) You dont have to chat or tip the driver
etc.

 

[spacecoin]

Exactly my point.

Ask Cruise if they are allowed to deploy driverless in California. They are not at this moment.

Not all companies are equal. It's a trust business.

 

[uscbucsfan]

Exactly my point.

This is all to say that again, we can agree to disagree. This is by no means "not debatable."

I'll see your "they took our jobs" meme, and raise you several blocks of people lined up to get a free ice cream cone or slurpee.

The point I was making is that this will not be a super smooth deployment and it's not for anyone with driverless vehicles. They are and will be more heavily scrutinized and so far the argument of exponentially less incidents hasn't been good enough. Cruise is banned (mostly for lying), but also for dragging a person in SF.

It's going to be a tough road for mass adoption and any small incident will be blown up like we see with AP/FSD now.

 

[uscbucsfan]

It doesn't have to be cheaper. It can be better in other ways, and at the same cost.

In a Waymo you get:
1) A premium car
2) You play your own music
3) You dont have to chat or tip the driver
etc.

None of those things are driving Waymo to be higher demand than Uber currently.

Please you don't have to tip in Uber/Lyft and with comfort you can play music and pick AC settings with a premium car.

In LV, most of the driverless cars were empty waiting for people when I was there 2 weeks ago while people waiting 15 min for an Uber/Lyft.

 

[Joey D]

Ask Cruise if they are allowed to deploy driverless in California. They are not at this moment.

Not all companies are equal. It's a trust business.

Of course. If you obfuscate an investigation for starters (like cruise did), you lose trust. Even in that case, they will be resuming operations.

 

[spacecoin]

Of course. If you obfuscate an investigation for starters (like cruise did), you lose trust. Even in that case, they will be resuming operations.

Not this year. They moved to Phoenix with safety drivers from what it looks like.

Speaking of starting from a place of negative trust, you should read the NHTSA ODI on Autopilot/FSD that come out today.

 

[Joey D]

The point I was making is that this will not be a super smooth deployment and it's not for anyone with driverless vehicles. They are and will be more heavily scrutinized and so far the argument of exponentially less incidents hasn't been good enough. Cruise is banned (mostly for lying), but also for dragging a person in SF.

It's going to be a tough road for mass adoption and any small incident will be blown up like we see with AP/FSD now.

I do understand what you're saying. I just believe that this technology (automated driving) is enabling to the point where that will overcome the challenges you cite. At some point, it will be accepted, just like automobiles were over horse and carriage...which faced similar bad press, FUD, etc.

 

[Joey D]

Speaking of starting from a place of negative trust, you should read the NHTSA ODI on Autopilot/FSD that come out today.

You should try using it.

 

[spacecoin]

None of those things are driving Waymo to be higher demand than Uber currently.

Yeah, I agree it's not a slam dunk and neither a high margin nor an easy business. Reminds me of auto manufacturing... That's why I think TSLA bulls are interesting in their analysis on TSLA hitting $2000 on deploying taxis.

 

[spacecoin]

You should try using it.

I did use v11 when I was in the US last time. For some reason Tesla won't update my car to FSDS (which I bought in 2019, with FSD). I tried Waymo in SF too.

 

[Joey D]

I did use v11 when I was in the US last time. For some reason Tesla won't update my car to FSDS (which I bought in 2019). I tried Waymo in SF too.

So we agree that v11 was no where near "getting acceptable". v12 is different.

 

[spacecoin]

So we agree that v11 was no where near "getting acceptable". v12 is different.

v12 is a lot better in terms of driving. Not reliability.

If I've learnt anything following the AVS-space for 10 years, it's that getting to "it drives and seems to work" is the easy part...

 

[Joey D]

v12 is a lot better in terms of driving. Not reliability.

Much more "reliable" in my experience. In any case this is the primary reason why I started this thread and put the effort in to documenting my experience. To show direct evidence (or lack thereof) of increasing reliability as new versions come out. With some measure of objectivity and also subjectivity.

Ver 11 (on city streets) was not worthy IMO of even being the "starting point" to start to look at this. V 12 is. Especially since Tesla is poised to iterate relatively quickly (via huge compute investments) and has "settled" on this architecture.

 

[spacecoin]

That is not my experience. In any case this is the primary reason why I started this thread and put the effort in to documenting my experience. To show direct evidence (or lack thereof) of increasing reliability as new versions come out. With some measure of objectivity and also subjectivity.

Ver 11 was not worthy IMO of even being the "starting point" to start to look at this. V 12 is. Especially since Tesla is poised to iterate relatively quickly (via huge compute investments) and has "settled" on this architecture.

I assume you're new to Tesla and FSD. I get the enthusiasm. But we've been hearing the "next version will blow your mind" and "Release X.Y should really be called version Z. so many times now since 2016.

That and the fact that the science isn't there yet for ML and CV makes me less hopeful. Let's hope you're right, otherwise you'll end up like me in five years. I hope you didn't pay $12k for FSD.

ps. Love the MS in ultra-red btw

 

[Joey D]

I assume you're new to Tesla and FSD.

Wrong on both counts. Why is it, that if someone has a different opinion than you, that that someone must be "naive" or "new?"

When Elon did his "Zuckerburg drive" about a year ago announcing / demoing the new architecture with Ashok, my first reaction was "it's going to be another few years of maybe getting a releasable version, and it will have tons more locality issues, etc." But here we are and after having used v12 myself, and seeing most others have the same experience, THEN I became more optimistic.

I get the enthusiasm. But we've been hearing the "next version will blow your mind" and "Release X.Y should really be called version Z. so many times now since 2016.

Yep, and not until version 12 did I think any of those kinds of statements were realistic.

That and the fact that the science isn't there yet for ML and CV makes me less hopeful. Let's hope you're right, otherwise you'll end up like me in five years.

What you are missing is that until this month, I was more like you than not.

I hope you didn't pay $12k for FSD.

I did not (and still have not purchased it). Why? Because I do not personally have a use case for automated city driving (yet). I purchased EAP for $6000 because I valued the highway assisted driving (and still do)...even if it never reaches higher than a level 2 ADAS.

I am on the fence with spending the now additional $2K it would cost me for FSD. I will at least be subscribing for a month when new major revisions come out so I can test and evaluate progress for myself. As an investor in TESLA, I don't believe what I read or hear about FSD. I need to experience it and evaluate it myself to judge its potential as a valuable technology.

ps. Love the MS in ultra-red btw

Thanks....once I decided to get the S....the Ultra red was a no brainer for me.

 

[uscbucsfan]

I've *sugar* on FSD my share of times and V12 is the first version where I've thought, "This can get there" as well.

It's really night and day from V11. Still has warts, but exponentially improved over V11. I want to wait until 12.4 to see how big the changes are with V12.

 

[spacecoin]

Wrong on both counts. Why is it, that if someone has a different opinion than you, that that someone must be "naive" or "new?"

Sorry about that. I just don't get that excitement anymore.

When Elon did his "Zuckerburg drive" about a year ago announcing / demoing the new architecture with Ashok, my first reaction was "it's going to be another few years of maybe getting a releasable version, and it will have tons more locality issues, etc." But here we are and after having used v12 myself, and seeing most others have the same experience, THEN I became more optimistic.

As I said before, in my opinion, removing the driver is a whole other ballgame.

Thanks....once I decided to get the S....the Ultra red was a no brainer for me.

I drive the X. It's been great. Right now I'm leaning towards a german car. I need to use the tow-hitch a few times per year for the boat, and I hate that I have to crawl under a $100k car to deal with that. Also I find the Tesla's (even the refresh) a bit too noisy for my taste.

I have free supercharging though so if I can bring that along, I might be persuaded to get a new MS. If I get an MS I will def get the ultra red.

 

[JB47394]

Tesla Cybercab

People who self-diagnose with Asperger's Syndrome (known for being socially awkward) should not assume the task of naming products.

Imagine what he'll do with the social order on Mars. "Okay, everyone with an even number is an Eloi. Everyone with an odd number is a Morlock. Oooo, this is gonna be awesome!"