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

Search results

  1. AlanSubie4Life

    FSD v12.x (end to end AI)

    Ok! 0/0 it is! Feeling confident. v12 is superbly good, but not 9/10 good. It has been quite good indeed on my turn. Probably at least 50% success rate.
  2. AlanSubie4Life

    FSD v12.x (end to end AI)

    0 for 1 on unprotected lefts on 12.3.1, @Daniel in SD. The 🍺 hangs by a mere thread. We'll see if he goes 9/9 on the next 9! I'm sure it was just the rain; those cars were super hard to see with their headlights and all. No normal human could perform this turn in these conditions. This...
  3. AlanSubie4Life

    FSD v12.x (end to end AI)

    (Words in brackets added by me.) If people can't see that these are very obviously different statements, and not equivalent at all, I am not sure I can help you. The first statement is true, per Tesla's data. The second statement clearly does NOT follow from the second. It may be true...
  4. AlanSubie4Life

    FSD v12.x (end to end AI)

    No, it’s not. I think accident rates were higher when headlights were on (when it was common to not have them on during the day). This is because there were more accidents at night.
  5. AlanSubie4Life

    FSD v12.x (end to end AI)

    Show me the data that demonstrates this.
  6. AlanSubie4Life

    FSD v12.x (end to end AI)

    Sure. But this does not follow, in any sense of the word “safe.” As has been pointed out, it is also safer to drive with your headlights off (when this was possible). Normally when talking about safer, we want to know whether or not you are less likely to be in an accident or not, in...
  7. AlanSubie4Life

    FSD v12.x (end to end AI)

    My car is fine. It has modern extremely capable hardware, capable of full autonomy: HW3. Drive on a road at 50-60mph with traffic lights every half mile or so and report back. I am curious if HW4 does better, actually, since I believe part of this could potentially be usable perception...
  8. AlanSubie4Life

    FSD v12.x (end to end AI)

    How do you figure? It’s not comparable. Describe exactly what it is that you are comparing.
  9. AlanSubie4Life

    FSD v12.x (end to end AI)

    No. It is subject to massive selection bias. There is no way to compare these categories. There is probably nothing wrong with the data - you just cannot compare the length of the bars - for fairly obvious reasons. This is the data I was referring to earlier. I am not sure why they...
  10. AlanSubie4Life

    FSD v12.x (end to end AI)

    This may be true - but just being better than poor drivers - what was demonstrated in these clips - is very possibly not sufficient.
  11. AlanSubie4Life

    FSD v12.x (end to end AI)

    I’m not opposed to a system that avoids collisions. As I made very clear, I was commenting on the poor reaction time in both cases. That could make a huge difference in a more difficult situation. It is important to compare against BEST human performance. This is a common error that...
  12. AlanSubie4Life

    FSD v12.x (end to end AI)

    Wow, those reactions were both pretty poor. Better late than never I suppose. Posting examples of poor FSD performance will never get old. One of these days it will finally be superhuman.
  13. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yeah I was seeing something like that this morning. However it was also (apparently?) resetting to other values in the background as I drove so it was confusing. So practically speaking for the original question here, just turn off ASSO mode, get all the benefits of ASSO (?), and know the...
  14. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yeah, just short handing it. I know it is just the ECU. Thanks for the confirmation. It’s a bit strange that other stations can’t provide compatibility but maybe that is because Tesla has not opened that standard to allow everyone to just mimic the SuCs? I was hoping that with NACS...
  15. AlanSubie4Life

    FSD v12.x (end to end AI)

    I know. Yeah. To be clear, as long as my car works at 3rd-party NACS DC chargers with the Tesla connector without needing a charge port update, I am not complaining, and I would agree Tesla won, by adopting CCS protocol. Otherwise, I would have to say my car has been left in the past because...
  16. AlanSubie4Life

    FSD v12.x (end to end AI)

    My understanding is that the Tesla proprietary standard was dropped by Tesla (though still supported on the Supercharger network), and CCS is the protocol going forward (the nice compact Tesla connector was preserved and will become commonplace in future, which is great). This is sad because it...
  17. AlanSubie4Life

    FSD v12.x (end to end AI)

    This DOES do something, but it is very laggy and unpredictable in ASSO mode as compared to manual mode. Honestly it is strange. It is very hard to describe the behavior. And it won’t address the original issue of speeding too fast in most circumstances, in any case. You’ll definitely need...
  18. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yes it is crucial. That and hills. 50mph limits on residential-feeding parkways and also surface streets through commercial areas are extremely common in San Diego. People go 55-60mph. Anyway it’s apparently very challenging for FSD v12.
  19. AlanSubie4Life

    FSD v12.x (end to end AI)

    I have not found that this works when ASSO mode is enabled (scroll down to an invisible 15mph or whatever). I guess I will try again. Maybe there is a long delay? (v12.3)
  20. AlanSubie4Life

    FSD v12.x (end to end AI)

    Turn off auto speed (ASSO). This gives you ASSO, but allows you to cap the speed with your desired manual set speed offset. Turning off ASSO does not turn off ASSO AFAIK. It just imposes a cap. (In other words the modes seem to behave the same if manual speed offset is +25% or whatever...
  21. AlanSubie4Life

    FSD v12.x (end to end AI)

    There have been a lot of versions lately, and they clearly just accidentally released the Cybertruck FSD version to the other vehicles. Nothing to do with relying on maps. This also explains the curbing issues which made no sense otherwise (obviously would not release something with known...
  22. AlanSubie4Life

    FSD v12.x (end to end AI)

    Awesome. Looking forward to it. Awesome! Looking forward to seeing everyone’s issues resolved. Not too surprising if this ends up being the case. Elon did say 12.3.1 would address it, after all.
  23. AlanSubie4Life

    FSD v12.x (end to end AI)

    OT, my last post on this: This is not quite true in the region of 55%-65% for NCA packs (it is a bit higher for NMC/LFP packs - curves are different but they show a similar peak in “returns”). This is pretty well established now and empirical data indicates this is how Teslas behave with...
  24. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yesterday it was worse without lead vehicles. (Though it is a mixed bag with lead vehicles, depending on how they stop. For some reason the car tracks the lead vehicle and not the traffic light.) When looking at this stopping behavior, be sure to be traveling at least 50mph towards traffic...
  25. AlanSubie4Life

    FSD v12.x (end to end AI)

    Sometimes this thread feels a bit like end-to-end Tulip mania. v12 is good, and makes some impressive moves, but it is not like it is a game changer - we’re not even close to a single order of magnitude improvement over v11 yet! Let’s at least wait for a second and maybe even a third data...
  26. AlanSubie4Life

    FSD v12.x (end to end AI)

    Commute: ~8 miles, mostly very light traffic ~12 disengagements (~9 for traffic lights (red) or stopped traffic; inability to ease off in time. 3 for curb proximity, unmarked residential streets) Unknown number interventions (not really important). v12.3 This is a really supremely tough...
  27. AlanSubie4Life

    FSD v12.x (end to end AI)

    I mean I certainly hope I have fewer disengagements but I don’t expect it. Lane positioning (not to be confused with lateral stopping position which is different and for which I had disengagements for safety) has been a little wandering for me but mostly ok (hitting bumps occasionally in...
  28. AlanSubie4Life

    FSD v12.x (end to end AI)

    I would be shocked if 12.3.1, 12.3.2, or 12.3.2.1 resolve the issues in 12.3 which require reports.
  29. AlanSubie4Life

    FSD v12.x (end to end AI)

    Good news, you’ll be able to see how my car is performing. I’m in the white Tesla. EDIT: 😉
  30. AlanSubie4Life

    FSD v12.x (end to end AI)

    I disengage for mistakes. I probably would have had only four if I had had MLC turned on this morning. It’s hard to cancel lane changes now due to no vibration and limited response to signal cancellation. So those can turn into disengagements. I’ll say they were dumb lane changes for...
  31. AlanSubie4Life

    FSD v12.x (end to end AI)

    Auto Max indicator appears/disappears when in ASSO “Ass*o**” mode and tentacle/snake length (subtle).
  32. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yeah that is no big deal; the metal part is just the edge of the battery pack, which is basically indestructible. Kind of a built-in rock slider. No one will care if that’s a little scuffed up. I warned people about this when I saw Omar’s curb kiss a couple weeks ago.
  33. AlanSubie4Life

    FSD v12.x (end to end AI)

    Commute: 8 miles, 6 disengagements (2 safety) 10-12 interventions. (v12.3) Robotaxi?
  34. AlanSubie4Life

    Is it better for my car to spend more time at 50% or have shallower discharge cycles but spend more time at a higher state of charge.

    Yeah the M3 LR is expected. But the Model Y is surprising since it gets the POS credit. I told my dad he was getting a Panasonic with his 2024 Y LR. Guess I was wrong? His energy screen method gives 79.5kWh, for what it is worth (probably the threshold). And he got similar numbers on...
  35. AlanSubie4Life

    Is it better for my car to spend more time at 50% or have shallower discharge cycles but spend more time at a higher state of charge.

    Yes, this EPA document is odd. Lookup website is down at the moment so cannot look at the document myself, though I trust @AAKEE is correct. Will need to keep an eye on this to see what happened. Maybe they decided to go back to the regular 2170s (rather than 2170L) for the faster...
  36. AlanSubie4Life

    FSD v12.x (end to end AI)

    It’s not like there is some huge untapped market of people out there who would willingly pay $200 a month on the regular (or intermittently) for FSD, but cannot stomach that first month out of pocket to try it out.
  37. AlanSubie4Life

    FSD v12.x (end to end AI)

    One month is far too short - just totally pointless for most people. Need 3 months minimum; 6 months would be more likely to convert the very small number of people who would not buy without a trial. In the end FSD is so expensive that only the completely price-insensitive people will buy it...
  38. AlanSubie4Life

    Elon: FSD Beta tweets

    Much better idea than the original though! This Tweet/X(Sh)eet only makes sense if they are rapidly moving everyone to the 2024 branch. Otherwise it is just pure desperation, long expected. Definitely, rolling v11 is not a selling point. My family has it free for a few months and I have...
  39. AlanSubie4Life

    FSD v12.x (end to end AI)

    Not nitpicky. This is a problem that some here claim to not notice in their use case, which makes me very worried! Could explain why this, below, somehow keeps happening! Always be disengaging, the moment it does something even slightly wrong (especially the wrong line)!!! Got very lucky...
  40. AlanSubie4Life

    Elon: FSD Beta tweets

    It has definitely been established that cameras must be clean. Soon they will implement the dewarping AI.
  41. AlanSubie4Life

    Elon: FSD Beta tweets

    With the emphasis on slowly. BTW when @sleepydoc was having problems getting his system to work, he was looking to park in spaces to the left of the car. I don’t think those are favored, are they? I assume it can still do it but maybe not as easy to get the option to show up?
  42. AlanSubie4Life

    FSD v12.x (end to end AI)

    No Yes
  43. AlanSubie4Life

    FSD v12.x (end to end AI)

    Yeah typical false and overblown commentary on YouTube. Anyone with patience to time the flickering of the AS symbol relative to brake application would be able to determine override timing. Looks to me it was right after max braking occurred but did not look closely. Yeah that puddle...
  44. AlanSubie4Life

    FSD v12.x (end to end AI)

    No. It failed to brake in a timely manner for an obvious and predictable slowdown (you can see the traffic stopped in front if you offset in the lane!!!) resulting in an unnecessarily small gap with high closing speed. Quite different than the typical early overbrake scenario you describe...
  45. AlanSubie4Life

    FSD v12.x (end to end AI)

    Look forward to your assessments of the “fixes.” Sadly they haven’t spoken of fixing the stopping and going (only addressing the maintaining), so I do not expect much there. This release of 12.3.1 bodes poorly for my bet with @Daniel in SD. Chuck did not ever get to 10 attempts on 12.3...
  46. AlanSubie4Life

    FSD v12.x (end to end AI)

    There was no emergency here. The car created the emergency by not easing off as soon as brake lights appeared (this is what normal attentive drivers do, even if they don't use the friction brakes). Videos are extremely misleading in distances to vehicles, deceleration rates, etc. It's the...
  47. AlanSubie4Life

    FSD v12.x (end to end AI)

    I think it certainly sees, perceives, and reacts to objects and events greater than 40 yards in front of the car, sometimes.
  48. AlanSubie4Life

    FSD v12.x (end to end AI)

    Again, that is consistent with what I said. It can certainly happen. But that’s super easy to control (assuming the detected speed is correct) - just turn off Auto Speed Offset, and set your manual offset to 0%, if that is what you want. AFAICT, you still get all the “benefits” of Auto...
  49. AlanSubie4Life

    FSD v12.x (end to end AI)

    No, not the issue. Your speed limit was likely set low enough it ended up being the control. Set the offset to +20% and you’ll probably find the modes are mostly indistinguishable except that in the old mode you can dial down the speed. Just luck. If you keep using it you’ll find the issues...
  50. AlanSubie4Life

    FSD v12.x (end to end AI)

    Hard to say. There’s thresholds for action, false negatives, false positives, etc. Hopefully Tesla looks at examples like this and can figure out why the NN was so slow to respond. Often it has pre-emphasis with v12. This was deemphasis, so quite different. In all cases I feel like...