Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Tenchusatsu who wrote (1578239)	12/18/2025 5:39:30 PM
From: Eric	Read Replies (2) \| Respond to of 1579255

T The irony is since Trump was reelected virtually all the posters now on the Tesla thread Subject 58139 is that none of them actually own a Tesla! Except myself. MAGA is alive and posting like mad over there. They found their new savior and hero with Musk, the latest AfD Supporter in Chief. E

To: Tenchusatsu who wrote (1578239)	12/18/2025 5:56:04 PM
From: pocotrader	Read Replies (1) \| Respond to of 1579255

Of the rides I have been on my trips to Disneyland and Knotts Berry farm, Space mountain at Disneyland and the parachute drop at Knotts berry farm were the worst.

To: Tenchusatsu who wrote (1578239)	12/18/2025 6:12:46 PM
From: combjelly	2 Recommendations Recommended By pocotrader Tenchusatsu Respond to of 1579255

A competent engineer wouldn't ignore any source of data like lidar or radar. Much less do it on the basis of "vision". Relying on vision alone means you bring in all of the failure modes that happen with vision, like what Tesla has run into. Like the Looney Tune example. And the "fix" smacks of desperation, the kind of thing you do when your dog and pony show has gone off the rails. The fact is that the Cybertaxi keeps missing its goals. I have seen articles where Tesla has ramped production like they did with the Cybertruck. In July Musk was claiming that half the US population would have access to service by the end of the year. To do that, they would have to have complete coverage in the top 40 metro areas. They don't even have complete coverage in 1, but if they were shooting for that goal in July, they would have to be pumping the things out like there was no tomorrow. I mean, it takes time to organize something like that. So they probably have bunches of the things stored somewhere. Or it was all a lie. Obviously, they aren't going to be covering half the population of the country by the end of the year short of triggering a nuclear war and shrinking the population down, like in Fallout. But, if you have drunk deep of the Kool-Aid, well I guess it all looks bright and shiny.

To: Tenchusatsu who wrote (1578239)	12/19/2025 1:02:14 AM
From: i-node	1 Recommendation Recommended By longz Read Replies (2) \| Respond to of 1579255

That claim was disproved. You should be able to acknowledge the pace of development of FSD and hardware/software transitions and updates in the software development environment. However, because I know you wouldn't be satisfied with mere truth, I'll give you the article: =================== Mark Rober's main claim in his March 2025 video—that Tesla's camera-based autonomy (specifically, what he presented as its "self-driving" capabilities) could be easily fooled by optical illusions like a painted fake wall or child mannequin, while LiDAR systems weren't—has been widely disputed and effectively disproved by subsequent recreations, analyses, and community testing. Key Reasons and Evidence: Outdated or Misrepresented Tesla Tech: Rober used Tesla's basic Autopilot (a driver-assist feature, not true self-driving), not the more advanced Full Self-Driving (FSD) system on newer hardware (e.g., HW4). Critics pointed out the video's title and framing were misleading, as Autopilot isn't designed for full autonomy. yahoo.com +2 Multiple recreations with FSD engaged (versions like v12 or v13) showed the car detecting and stopping for the same obstacles, contradicting his results. carscoops.com +1 Recreations and Debunks: Tesla owners and YouTubers replicated the tests using proper FSD settings and found the car avoided the illusions. For instance, one test showed FSD v13 passing the "painted wall test" while an older v12 failed, highlighting software evolution. teslaoracle.com Videos like "Mark Rober Debunked - You can't fool a HW4 Tesla using FSD!" (by creators such as Kyle Paul) went viral, demonstrating successes where Rober's setup failed. Another called it "BS," alleging editing tricks or disengagement of features before impact. youtube.com +1 Bias and Sponsorship Concerns: The video was accused of being a veiled promotion for Luminar (a LiDAR company), with Rober's ties to its founder noted. Luminar's CEO was later ousted amid an ethics inquiry, adding fuel to skepticism. reddit.com +3 Rober released raw footage to defend himself, but it didn't sway many, leading to backlash like lost subscribers. Ongoing Discussion: As of December 2025, X posts and articles still reference it as "debunked" or "faked," with Tesla enthusiasts calling it rigged for clicks or anti-Tesla bias. facebook.com +1 While not every critique is unanimous (some defend the video's intent to highlight tech limitations), the preponderance of evidence from real-world FSD tests shows his specific claim doesn't hold up. In short, yes—his claim has been disproved through better-controlled experiments using Tesla's actual self-driving tech. It's a reminder that autonomy software evolves rapidly, and tests need to reflect current capabilities. I'm quite sure these sources aren't enough for you but you will need to do your own digging if you want to argue the points above.