Hi, Shrewd!        Login  
Shrewd'm.com 
A merry & shrewd investing community
Best Of Politics | Best Of | Favourites & Replies | All Boards | Post of the Week!
Search Politics
Shrewd'm.com Merry shrewd investors
Best Of Politics | Best Of | Favourites & Replies | All Boards | Post of the Week!
Search Politics


Halls of Shrewd'm / US Policy
Unthreaded | Threaded | Whole Thread (4) |
Post New
Author: onepoorguy 🐝  😊 😞
Number: of 48448 
Subject: self driving cars
Date: 03/15/2025 8:33 PM
Post Reply | Report Post | Recommend It!
No. of Recommendations: 5
I was a big believer in this at one point. Not so much anymore, after I starting paying a lot more attention to the tech. Anyone who thinks it will happen in a few years really needs to watch the NOVA program "Look Who's Driving". It is so much more complicated than you might think. And road construction, or bad weather, amplify the complications.

This leads into Mark Rober testing his Tesla versus a LIDAR-enabled vehicle. Tesla relies solely on cameras. You might be able to guess where this goes. It's a cool video. I believe the Waymo (which I see around town here occasionally) uses LIDAR.

https://www.youtube.com/watch?v=IQJL3htsDyQ&pp=0gc...
Print the post


Author: Mark   😊 😞
Number: of 48448 
Subject: Re: self driving cars
Date: 03/17/2025 10:51 AM
Post Reply | Report Post | Recommend It!
No. of Recommendations: 3
This leads into Mark Rober testing his Tesla versus a LIDAR-enabled vehicle. Tesla relies solely on cameras. You might be able to guess where this goes. It's a cool video.

It's also a "produced" (mostly fake) video to create content for his sponsors/advertisers! There are so many things:
1) He does NOT use the FSD feature (it is clear in the video that he is using Tesla's version of cruise control and lane keeping, not the FSD feature, they are two entirely different things)
2) He's using their cruise control, BUT it is turned off shortly before hitting the fake wall!
3) The guy with him in the car is wearing a Luminar T-shirt. That's one of the sponsors.
4) The video has been edited. The Youtube video is clearly different than the raw video he posted on X while defending his actions.
5) He even faked the camera used to video it! He used an iPhone Pro model, but photoshopped in a Google Pixel logo [incorrectly] onto the back of the phone. That's because Google Pixel is also a sponsor of his content.
Print the post


Author: onepoorguy 🐝  😊 😞
Number: of 48448 
Subject: Re: self driving cars
Date: 03/18/2025 9:31 PM
Post Reply | Report Post | Recommend It!
No. of Recommendations: 0
Yeah...it wasn't a "fake" video. He showed both normal and "self-driving" modes, and commented that the self-driving mode brakes much too frequently at nothing (because it assumes you aren't driving). Rober doesn't do fake videos, he explores the engineering of something. He was have no following if he faked all this stuff. Though, I'll take your word about the sponsors. Even though I know that he used iPhones for his glitter bombs for porch-pirates. Logos prominently displayed in those videos.

If you had seen the Look Who's Driving Now program, you would know that the heavy mist/rain is a problem for camera-only systems. Also snow. They can't see through it. Even we can't see through it. But the LIDAR was able to see enough to stop.
Print the post


Author: Mark   😊 😞
Number: of 48448 
Subject: Re: self driving cars
Date: 03/18/2025 11:14 PM
Post Reply | Report Post | Recommend It!
No. of Recommendations: 3
He showed both normal and "self-driving" modes

He never entered the self-driving mode! He engaged adaptive cruise control with lane keeping (what Tesla calls "autopilot", a terrible name for what others call adaptive cruise control). Tesla also has a much more advanced feature that they call "FSD" (also a terrible name because it is neither F nor S yet). I use both features regularly and they are completely different features and have nothing at all to do with each other. Part of the problem is that Tesla picked god-awful names for these features, and another part of the problem is that people refuse to educate themselves what each of the features are and what they each can do.

Then he began the test by driving the car up to some medium speed, 42 mi/hr in one of his takes, and 40 mi/hr in another of his takes. Yes, Rober later posted that they did multiple takes, and rebuilt the painted styrofoam wall in between each take. Then with the car sped up, while aimed at the fake wall, he engaged the adaptive cruise control 4-5 seconds before reaching the painted styrofoam wall.

Now I suspect that any car on adaptive cruise control could be made to drive into a fake wall if you speed up enough and engage the system late enough. A better test for Rober to have done would have been to go a bit over half a mile up the road, engage the adaptive cruise control at 40 mi/hr, let the car drive the little over half mile over the next 40-50 seconds, and see if it slows and perhaps stops as it approaches the painted styrofoam wall.
Print the post


Post New
Unthreaded | Threaded | Whole Thread (4) |


Announcements
US Policy FAQ
Contact Shrewd'm
Contact the developer of these message boards.

Best Of Politics | Best Of | Favourites & Replies | All Boards | Followed Shrewds