Repeating the experiment by Mark Rober on autonomous driving systems, but with variations, yielded outcomes indicative of the original, albeit with some distinct differences.

A recent replication of the Tesla "wall test" by YouTuber Kyle Paul has reignited a debate about the capabilities of the electric vehicle manufacturer's Full Self-Driving (FSD) feature. In this test, Paul used two vehicles: a 2022 Model Y with FSD Hardware 3 (HW3) and FSD version 12.5.4.2, and a Cybertruck with HW4 and FSD version 13.2.8.

The original test, conducted by Mark Rober, featured a Model Y running on Autopilot, not FSD, which failed to recognise a printed road scene on a wall in time, necessitating manual driver intervention to prevent a collision. However, in Paul's recreated test, the Model Y with HW3 and FSD version 12.5.4.2 performed similarly to the original test.

In contrast, the Cybertruck with HW4 and the latest FSD version 13.2.8 recognised the same wall from a safe distance and stopped on its own without any driver input. The wall in the Cybertruck's test stood out more clearly against its surroundings, potentially giving it an edge over the Model Y.

However, it's important to note that environmental factors may have played a role in the improved result of the Cybertruck's test. The test was conducted near dusk, with more pronounced lighting and environmental contrast, which could have aided the Cybertruck's performance.

Rober's initial test may have been flawed due to differing test conditions. The changed factor in Paul's repetition of the Tesla test compared to Rober's original was likely the testing conditions or methodology, which could lead to different results; however, specific details about this change or its exact impact were not found in the search results.

The Tesla "wall test" has sparked a lively discussion about the capabilities of FSD. Some argue that the test results highlight the potential improvements in FSD with the latest hardware and software updates, while others question the validity of the test due to the varying conditions and methodologies.

For those interested in experiencing FSD for themselves, a referral link is provided that can be used to get three months of Full Self-Driving when buying a Tesla.

Given the ongoing debate, a retest under identical conditions using Hardware 4 (HW4) and the latest version of Full Self-Driving (FSD) is suggested to provide a more definitive comparison between the two vehicles. Only then can we truly understand the advancements in Tesla's FSD technology.

Repeating the experiment by Mark Rober on autonomous driving systems, but with variations, yielded outcomes indicative of the original, albeit with some distinct differences.