Jump to content

Guadaloopy

Legacy Members
  • Posts

    11101
  • Joined

  • Last visited

Posts posted by Guadaloopy

  1. 2 minutes ago, jimmyjazz said:

    Today I parked in a lot next to what I assume was a small Tesla sedan.  It had no badge.  It also had no occupants.  I was checking some messages before going into the store I was patronizing and I was stunned to see this car pull out and drive towards the store entrance.  Zero humans.  It actually started to cut off another vehicle (with driver) who honked and stopped.  I'm sorry, but that's just very unnerving and not really ready for prime time.

    PS I'm a mechanical engineer who has done a bit of controls work in my career, I'm not a luddite.

    Actual smart summon still sucks.  It still has a lot of restrictions on it and doesn't operate like supervised FSD at all.  Once the "training wheels" come off of it and the full capability of FSD 13 takes over that feature, it will be much better.  People who use that function now in crowded parking lots are assholes. 

  2. 12 minutes ago, Captainant said:

    In this tweet, it shows the autopilot disengaging at the last second. There's been a huge disparity of accident reporting where "autopilot/FSD wasn't in charge!", even an NHTSA investigation about it

    https://www.motortrend.com/news/nhtsa-tesla-autopilot-investigation-shutoff-crash/

    They found a number of incidents where autopilot "on average aborted vehicle control less than one second prior to the first impact."

    Yep... that is a known issue that has been noted by NHTSA.  The fact that autopilot disengaged prior to impact isn't my main concern with that video. 

    My biggest concern is with how late he engages autopilot and the fact that it took three attempts to get it to activate.  If I had the time, resources, and talent to paint a Road Runner mural, I would love to recreate this and engage FSD a quarter mile out to see what would happen.  

  3. 2 minutes ago, Chopper said:

    https://www.theverge.com/tesla/631308/mark-rober-tesla-youtube-autopilot-lidar-fake-claims

    Seems like a lot of what you wrote has been debunked or shrugged at. But I'm no rocket scientist.

     

    My main concern with the video is the warning notifications that show up on the screen prior to impacts.  Without seeing those in focus and knowing what the warnings were actually for, it sure looks like the accelerator was engaged during some of these tests and overriding the system.

  4. On 3/12/2025 at 7:43 PM, Tailgate said:

    Tesla FSD (call it whatever you want) is unreal.

    The over-the-air updates even in the last few months have been game changing.

    I can go anywhere to anywhere with zero manual wheel, brake or acceleration assistance. I’ve been using it 10x more than this time last year.

    Just watch. Please bump this thread when they are no longer leading the EV race.

    There is nothing the argue here. Take your emotions to the cloak room.

     

    Leaps and bounds in the last couple of months alone.  The last OTA update (12.6.4) now turns into my driveway and pulls into my garage to park without prompting.  And I don't even get the full set of v13 features since my 21 Model 3 only has HW3.  

    • Hook 'Em 1
  5. 3 minutes ago, Captainant said:

    Lol 52-80 just posted a TLDR in the business board about 5 minutes before your post. Did the newsletter just go out or something?

    Nah...  Just took my time looking at the video.  I watched a couple of other analysis videos from yesterday, but they weren't all that compelling.

  6. On 3/16/2025 at 5:05 PM, Captainant said:

    Pretty good bit of real world proof on LIDAR vs cameras from YouTuber Mark Rober

    Screenshot_20250316_170528_RedReader.thumb.jpg.a73e959fe14701d17f2f173e6d1931cf.jpg

    I mean... y'all knew I wouldn't leave this one alone.  It was just a matter of time before this catfish ate the stinkbait.

    I did take my time and watch the entire video and did multiple viewings of all of the actual tests.

    You'll all be shocked to hear that I think the video is a bit disingenuous and manipulated to give pre-determined results.  Here are my observations that inform my skepticism:

    • Rober doesn't use any FSD build and limits his testing to EAB and Autopilot.  Yes, the sensor suite and hardware is the same, but the software is vastly different between autopilot and FSD 12+.  Rober pits the latest version of Luminar's system against an older generation Tesla system.

     

    • In each test where the screen is visible (except for the EAB test), there is an out-of-focus warning notification present prior to impacting the dummy or the wall.  Anyone who uses the Tesla TACC, autopilot, or FSD system will tell you that if you have the accelerator pressed and are going above the set speed, you get the following warning:  image.png.d0715c7ff9482cc3ab207878aa4e5f94.png          Even EAB will keep going when the accelerator is pressed, which is what appears to happen in the video.  It won't give the visual warning notification, but it will give an audible alarm which you hear about two seconds prior to impact in the EAB test in the video.  To my eye, it looks like Rober is pressing the accelerator and not actually allowing the driver assist systems to maintain speed on their own.

     

    • In the raw footage that Rober released in response to criticism, rather than setting the autopilot early in the test which is easily doable, he waits until very late (<5 seconds) to attempt to engage FSD and fails on two attempts to engage before getting it to engage on a third attempt.  There's no reason he couldn't have engaged autopilot much earlier.  It's just really shady and looks like he is intentionally engaging late to confuse and cause errors in the system. https://x.com/MarkRober/status/1901449395327094898

     

    • The "fog" test was not fog, but smoke.  LIDAR is significantly degraded in actual fog when water in the air scatters and absorbs the lasers.  Credit to the LIDAR system for "seeing" through the smoke, but that is an extremely uncommon scenario for actual driving (no safe driver proceeds at full speed into an obscuring smoke cloud) that was engineered to give the LIDAR system an easy "win" in this comparison.

     

    • In the "rain" test, which did not simulate a realistic rain scenario that one would actually drive in, the LIDAR system stopped because it read the significant water spray as a solid object, not because it saw the dummy.  Judge's review does not award the point to LIDAR in this comparison.  LIDAR is degraded in rain just like in fog due to attenuation.

     

     

    • The CEO of Luminar (the LIDAR system featured in the video) is present in the video and is a previous collaborator and admitted friend of Rober.

     

    For posterity:  Motherfuck Elmo for being a nazi asshole.

    You may now proceed with the knob-slobbing gifs.

    • Haha 4
  7. 2 hours ago, Texas Jeff said:

    A little more info on the Waymo accidents:  https://www.damfirm.com/waymo-accident-statistics.html

    Waymo has indeed been in about 700 accidents from 2021-24, but 632 of the 700 accidents involved no injuries.  Waymo was not at fault in many of these accidents.

    Only three involved serious injuries and there has been one fatality:

    * One in SF involved a Waymo stopped at light.  Another car crossed a double yellow light and crashed into a SUV, which hit the stopped Waymo
    * One in LA involved a Waymo following a box truck.  The Waymo had a driver.  The Waymo was hit from behind by another passenger car.  That car may have been changing lanes.
    * One in SF involved a Waymo stopped at a light, that proceeded into the intersection on green.  A cross-traffic car ran the red light and crashed into the Waymo and another car.  The red light car hit a total of two cars plus pedestrians.

    The fatality involved a Tesla in SF traveling at a reported 98 MPH that hit a Waymo, plus other vehicles.  Everyone but the Tesla was stopped at an intersection.  The Waymo was unoccupied at the time. The person killed was in a Lexus that was hit by the Tesla.  The Tesla had a human driver at the time.

    Eventually self-driving cars are going to be much, much safer than humans, either Tesla or Waymo or whatever.  We may already be there. Crashes and injuries and fatalities are going to go way, way down.  DWI is going to go hopefully to zero.  It can't get here fast enough, IMHO.

    Waymo’s not going to fuck you, dude.  
     

    J/K… good info.  Thanks for sharing.  

    • Hook 'Em 1
  8. 49 minutes ago, Js1 said:

    Yes. I don’t see how you can have Ted Lasso without also including all of Coach Beard, Keely, Roy, Rebecca. Even Jamie 

    Jamie will not be a regular due to scheduling issues.  Coach Beard and Roy Kent are in as EPs for the show, but not confirmed as cast members. 

    I'm all in on keeping Hannah Waddingham employed and on my TV screen.

  9. 5 hours ago, Longboard Horn said:

    Gl9KcfcbYAE0xPT?format=jpg&name=large

    Gl9KcfYbYAU-7Im?format=jpg&name=large

    Gl9KcfYbsAAkbED?format=jpg&name=large

    Gl9KcfZa8AAyth8?format=jpg&name=large

    trip.gif.209380d1f8a10c03e21167f254e4d6e5.gif

    trip2.gif.f61decc4bc2d4538c0a954004e47f5f3.gif

    Feels like our normal supply of aggy misery has been spiked with some of that Chinese fentanyl over the past four months.

    • Haha 1
  10. 2 minutes ago, Pig Bellmont said:


    I won’t speak for the person you’re quoting, but I’d guess it has something to do with this:


    As of October 2024, there have been hundreds of documented nonfatal incidents involving Autopilot and fifty-one reported fatalities, forty-four of which NHTSA investigations or expert testimony later verified and two that NHTSA’s Office of Defect Investigations verified as happening during the engagement of Full Self-Driving (FSD).”

    https://www.forbes.com/sites/stevebanker/2025/02/11/tesla-again-has-the-highest-accident-rate-of-any-auto-brand/

     

     

     

    Auto-pilot ≠ FSD. 

    Two fatalities have occurred with FSD engaged.  Over three billion miles have been driven with FSD. 

    Waymo has had nearly 700 accidents and one fatality with somewhere around 25 million miles driven.

    • Hook 'Em 1
  11. 10 minutes ago, choripan said:

    This is incorrect. Waymo operates on freeways in at least California and Arizona. Google it so I don't have to do it for you. Waymo has a far more comprehensive set of sensors and is WAY ahead of Tesla, which is not even #2 in the AV game.

    The difference is that Waymo generally works with regulators to establish agreed-to geofences and parameters that allow safe(r) deployment and development of the AV solution, rather than putting erratic "beta" machines on public roads and selling them as "full" self driving, with untrained drivers at the wheel.

    Source: Regulatory lawyer who has spent hundreds of hours on this very issue (not for Waymo, btw. No dog in the fight.).

    I stand corrected on the freeway issue.  Waymo has been testing on freeways in Phoenix since January 2024 and just moved onto LA freeways this January.

    I'm curious as to the measure used to state that Waymo is "WAY ahead" of Tesla.

    For the bolded, this is continuously referenced as some horrible unchecked danger to the public.  The data do not bear out this argument.

     

×
×
  • Create New...