We keep hearing about sudden acceleration, Toyota, or Model X's running into gyms. This is a very cool podcast, available on TuneIn. All are worth a listen. Revisionist History Episode 08
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
AMPd
Well-Known Member
Didn't listen
No such thing as sudden acceleration.
Car did what the driver instructed it to do, they hit the accelerator, car accelerated. End of story
No such thing as sudden acceleration.
Car did what the driver instructed it to do, they hit the accelerator, car accelerated. End of story
aesculus
Still Trying to Figure This All Out
Small correction. It was a Model S that ran into the gym. A Model X ran into a nail salon I think. We need to keep this straight.
RobW2428
Member
mblakele
FSD Beta (99)
I don't have a dog in the fight, but this rebuttal may also be worthwhile: A Tipping Point | Safety Research & Strategies, Inc.
Phillip L
Gas Passer
Thank you Very much for the link.
I had really enjoyed the Revisionist History Podcasts but I have read this long rebuttal of the one on Unexplained Acceleration and it is very convincing to me that Gladwell has done a disservice here to a complex story. Before reading this rebuttal I had totally bought into Gladwell's thesis but this article is very convincing. Sure there are human error case, maybe even the majority of low speed ones, but that does not negate the evidence for a large number of mechanical / electronic incidents.
hybridbear
Active Member
It would be interesting to see the rebuttal to the rebuttal link.
The professor who was soldering jumpers on the computer bits of cars, then claiming the product was somehow related to reality in the field was crazy. You can break anything by soldering on jumpers, see the TV show Burn Notice. Lol
That experts complain about code quality, let me assure everyone that new code looks like crap to every programmer. Having made my bones dealing with problems like this, if you locked me in a room with code for months, looking for the part where there was a case there full acceleration was stuck on forever, and I couldn't find it? What does that say? They aren't running the code with a debugger, on a computer? I guess it could be so complex, it was impossible, but that seems unlikely to me. Does this alleged acceleration bug also stop the driver from breaking and turning off the car? I don't like this answer at all. If the code is so bad, how does Toyota have such reliable cars? Surely, you couldn't have one terrible, bad code bug, from a terrible code base, but have no other major bugs.
This is what I would like to know. Every car has a black box that records the drivers actions before the crash. Each Tesla crash has news articles written about it, with a full analysis of driver actions. If this is so terrible, why isn't there a full black box breakdown? Tesla makes it look so easy.
Who funded the rebuttal? Do they have a horse in the race?
I have no clue what is right.
The professor who was soldering jumpers on the computer bits of cars, then claiming the product was somehow related to reality in the field was crazy. You can break anything by soldering on jumpers, see the TV show Burn Notice. Lol
That experts complain about code quality, let me assure everyone that new code looks like crap to every programmer. Having made my bones dealing with problems like this, if you locked me in a room with code for months, looking for the part where there was a case there full acceleration was stuck on forever, and I couldn't find it? What does that say? They aren't running the code with a debugger, on a computer? I guess it could be so complex, it was impossible, but that seems unlikely to me. Does this alleged acceleration bug also stop the driver from breaking and turning off the car? I don't like this answer at all. If the code is so bad, how does Toyota have such reliable cars? Surely, you couldn't have one terrible, bad code bug, from a terrible code base, but have no other major bugs.
This is what I would like to know. Every car has a black box that records the drivers actions before the crash. Each Tesla crash has news articles written about it, with a full analysis of driver actions. If this is so terrible, why isn't there a full black box breakdown? Tesla makes it look so easy.
Who funded the rebuttal? Do they have a horse in the race?
I have no clue what is right.
WannabeOwner
Well-Known Member
That can definitely be a problem in an unfamiliar car - e.g. if it has no key, and instead a "press and hold" or somesuch mechanism (to prevent accidental use)
As I see it the problem is that most people don't know what the, new, side effects are in a new version of technology. I get a Tesla and put it on AP. I have no idea that it cannot see parked cars, or gets squiffy if I go over a rise.
Or the Throttle and Brake are now fly-by-wire, and no one tells me this. There is, also, no longer any key that I can "turn off" as I am used to ... so I have to know / learn that there is a push-and-hold to turn off. How would I ever know that, without reading the whole manual or research or somesuch? Hoping to discover the Right Way in the first full-on emergency is a pretty hopeless solution.
Roll on full autonomous so I can stop having to keep up-to-speed with all the blessed changes ...
... I'm sure there is a good reason why "swipe to logon" on my iPhone has been retired ...
You are back to Gladwell's conclusion them? When the driver doesn't know how to operate the car correctly, that is when these weird crashes happen?
WannabeOwner
Well-Known Member
Sorry, I wasn't clear on that point. No, I'm sat on the fence. Biggest thing for me is "Why did Toyota pay out all that money / suffer all that bad PR if there was no software bug and it was only ever driver error". ( I am not familiar with the details, maybe Toyota paid out purely for "heavy, loosely fitted, mat caught on accelerator pedal"? and that was demonstrably their fault)
I've read reports that Tesla say the unexpected acceleration in Model X (maybe Model S too? I can't remember) was logged as accelerator pedal usage. I'm even sceptical about that ... maybe it logs like that, but can we be sure that the only cause can be pressure on pedal? Even with redundant sensors perhaps the logging can still be fooled into recording it as pedal-travel rather than software error.
I have a friend who drove his Porsche 911 into a wall. Wet day, wet shoes, foot slipped off brake onto accelerator. He's honest about it (even on his insurance claim!) but maybe that could happen to some people and them not realise it.
Similar threads
