@Skarrow9 The update is 148.4 GB. To download it within 4 hours you would need to average ~82 Mbps.
For context, the median download speed in the US is ~200 Mbps.
Yes it would register some of the system’s torque, however we see a couple Nm of torque applied for a short duration, followed by a large spike in torque at the exact time FSD disengaged, which it’s designed to do if the driver applies a large torque to the wheel.
That torque was then maintained after FSD disengaged, meaning it was the driver inputting the left steering input, not the car, for multiple seconds after.
So no, we do have a very clear indication of what applied these CCW torques into the steering column, ultimately leading to the crash.
@alias23b@DevinOlsenn FSD disengaged simultaneously to the spike in torque.
Noise in the torque graph isn’t anything unusual, as the steering column still moves when FSD has control. Torque sensor in the steering column would register “resistance” in that case.
@uhohitsa_burner @Space_Josiah@bscholl@DJSnM Linked a NASA paper summarizing this topic. Long story short, most aero structures are designed with ultimate failure at 1.4 & yield at 1.25, and then iteratively tested and refined up as needed (statistical failure rate, fatigue, etc)
https://t.co/UPk4Bwjpt5
Commercial aircraft are designed with 1.25 to usually at most 2.0 SF (part dependent).
Aircraft are limited by pressure cycling fatigue, not so much creep.
The purpose of LM’s test was to demonstrate resistance to creep from internal pressure, which for a permanently inflated space module is kind of important.
Commercial aircraft are designed with 1.25 to usually at most 2.0 SF (part dependent).
Aircraft are limited by pressure cycling fatigue, not so much creep.
The purpose of LM’s test was to demonstrate resistance to creep from internal pressure, which for a permanently inflated space module is kind of important.
Commercial aircraft are designed with 1.25 to usually at most 2.0 SF (part dependent).
Aircraft are limited by pressure cycling fatigue, not so much creep.
The purpose of LM’s test was to demonstrate resistance to creep from internal pressure, which for a permanently inflated space module is kind of important.
It seems obviously illegal for an American agency to order a Chinese company to block American citizens from sharing firearms blueprints. The legal precedent is extraordinarily clear, this is a blatant violation of the 1st Amendment with a foreign adversary as the agent.
Horrible. Don’t attack protesters. Thankful no one was hurt
Liberals may clap for violence against Teslas because they see it as justified, but once you start a cycle of violence the escalation will not always be under your control
that is why all violence is wrong
Tesla Cybertruck with HW4 using FSD v13.2.8 did not hit the wall in a new recreation of Mark Rober's "Wile E. Coyote-Style" self-driving test. However, a Model Y with HW3 using FSD 12.5.4.2 was going to hit the wall.
Nice test @alsetcenter
Absolutely BODIED by community notes.
I checked the YT video myself, the note is accurate. In the "raw" footage Mark posted, autopilot engages at 42pm vs 40pmh in the YouTube video.
But it gets WORSE. See the next posts in this thread.
This is honestly really sad. A former NASA engineer who was a real inspiration to me and many others, blatantly lying, faking tests and being paid to promote a failing business.
I really did not expect this from Mark 😥
@DrChrisCombs From experience, any Aero Professor would be shitting bricks if a fully functional, manufacturing-oriented / scalable, and capable drone platform came from a senior design team in a single semester (that took years for a team of paid engineers to develop in reality).