Slashdot Mirror
Consumer Reports Recommends Tesla's Model 3 After Braking Fix (reuters.com)
Consumer Reports said on Wednesday that it now recommends Tesla's Model 3 sedan after its latest tests showed that a firmware update improved the car's braking distance by nearly 20 feet. From a report: The magazine last week flagged "big flaws" in the car, including braking slower than a full-sized pickup truck, while also highlighting many positives. In a tweet, Mr. Musk said he really appreciates "the high quality critical feedback from @ConsumerReports. Road noise & ride comfort already addressed too. UI improvements coming via remote software update later this month."
The Slashdot crowd isn't going to like this...
They said CR was wrong but now apparently they say CR was right...
Browsing at +1 - no ACs, I ignore their posts. So refreshing!
... and over-the-air update can also break it. Or take away the "feature" once the car leaves the showroom. If it were so easy of a fix, one has to wonder why Tesla didn't recognize and fix the problem in the first place? Why did it take a third party tester to find it?
...CR just gots PAID!!!
The appeal of this car is a lot less knowing that the auto-drive feature drove directly into barrier killing the driver because of bad lane markings. Doesn't the car take into account all of the other cars in front of it? Or does lane markings trump the direction other cars are going?
Namaste
If they are consumed, that will mean you're a eunuch.
Eunuchs freaks need not apply!
>> Consumer Reports Recommends Tesla's Model 3 After Braking Fix
I've never seen "Braking Fix" as a euphemism for "Payola" before.
CR said: Breaking distance is > 150'
Tesla said: Our testing says 133'
CR said: Same, on our first try, subsequent tries were longer.
Tesla said: Oh crap, that's probably a bug in our regen breaking stuff--thanks for pointing that out.
Tesla rolls out a fix and CR verifies the fix. It seems like everyone was well-behaved all the way around.
Is this really the first time you've ever seen a bug in a product? Or are you shorting TSLA?
How can Tesla explain why it took CR's reviews to analyze braking distance and come up with the correct algorithm to stop the car in the shortest distance possible?
Isn't it more likely that they took CR's data and modified the braking approach to optimize for CR's experimental setup but presumably results in a worse braking distance in some other real-world experience?
The real story here is that Tesla has XYZ,000 cars out on the road designed by people who collectively decided that the proper response to the driver flooring the brake pedal should be something other than bringing the car to a stop as quickly and safely as possible.
Oh, and as soon as someone brought that clearly unintuitive design requirement to their attention, they somehow were able to address it in a matter of days. Just like any other pesky little bug fix, right?
This is madness.
Consumer Reports Recommends Tesla's Model 3 After Braking Fix
Seems counter-productive. :-)
It must have been something you assimilated. . . .
What happens when another firmware update breaks the fix? No pun intended. Any vehicle with this amount of change possible in firmware should be tested regularly. Tesla certainly won't let anyone know about it.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Wow, I thought Nerds educated themselves.
The amount of control the ECU or ABS controller has over the braking of a modern day car is stupid. It handles the amount of pressure sent to each individual brake caliper on each wheel. This allows that controller to act as an ABS controller, a traction control device and a bias adjustment all in one. To work properly it also needs to have input from the rest of the CAN BUS in the car so that it can do its job. So if you want to tweak the settings that the brake controller receives from the CAN BUS you can affect how the brakes perform.
Should it be this way? Meh, I prefer a nice physical limited slip differential to any pseudo brake magic stuff. But for most drivers who just want to get in a car and show up somewhere all while sneaking a look at their cell phone every 2 minutes, the digital stuff is fine.
Now get off my lawn.
Dude, Jimmy Fallon creeps me the fuck out. His laugh seems so fake and forced in my own opinion.
Listening to Elon speak makes me think he is actually very introverted and has stepped out of his shell quite a bit.
All it means in effect is that Tesla's own testing and QA is so shoddy that it took a 3rd party to point out how dangerously bad the braking distance was. It's good that it can be rectified over the air but it doesn't absolve them putting it out in that condition to begin with. What else did they not bother to test, or allowed to slip past QA for fear of missing their targets?
It is ON by default, it reduces brake wear significantly and reduces brake heating and energy wastage significantly.
But they turned it off while testing Tesla. Why? To be "fair" to the ICE cars? To be "consistent" with earlier testing of gas cars? The test involves 60 to 0 braking hard, five times in a row with one mile of driving in between to cool.
This test is probably designed to promote disk brakes over drum brakes, by disk brake makers, by lobbying SAE to use this as the "standard".
Now EV makers should come up with a test that is impossible for ICE cars to pass because of the lack of regen braking. With ABS all braking distances are limited by the tire not the braking power. So they should do five consecutive 60 to 0, reducing the "cooling run" from 1 mile to 15 seconds. With regen EV can do it without heating the discs. ICE will set their rotors on fire.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
I hope this update doesn't crash
"Tesla in Autopilot mode crashes into parked Laguna Beach police cruiser" A Tesla sedan in Autopilot mode crashed into a parked Laguna Beach Police Department vehicle Tuesday morning, authorities said. The collision happened at 11:07 a.m. at 20652 Laguna Canyon Road, according to Laguna Police Sgt. Jim Cota. The officer was not in the cruiser at the time of the crash. The Tesla driver suffered minor injuries, but refused transportation to the hospital. “Thankfully there was not an officer at the time in the police car,” Cota said. “The police car is totaled.” http://www.latimes.com/local/l... "Tesla in Autopilot sped up before Utah crash, police report says" A new police report says a Tesla that crashed in Utah while in Autopilot mode accelerated just before it smashed into a stopped firetruck. https://www.sltrib.com/news/20...
Respond to criticism by fixing the problem? What a novel idea! It also seems to be working better than the usual Tesla approach of shitting all over anyone and anything that isn't 100% enamored.
I'm assuming there's some trade between shortest stopping distance, optimal safety in anti-lock braking, Optimal safety in steering while braking, and optimal regenerative braking. One could always shorten the stopping distance of ANY car. Just put on grippier tires and detune the Antilock breaking. But the car may lose some handling while braking or slide on slippery surfaces.
it's all trades.
Some drink at the fountain of knowledge. Others just gargle.
With ABS all braking distances are limited by the tire not the braking power.
Considering that Tesla just reduced the braking distance with a firmware update, this statement is wrong: it is also limited by how aggressive the ABS is designed. Also, in an emergency panic stop regenerative braking is relatively inconsequential.
My guess is that there is a trade-off between braking stopping distance vs regenerative energy recovered while braking.
In a traditional car, if someone stomps on the brake pedal, all of the forward momentum is converted to and dissipated as heat when the brake pads are squeezed against the rotors to stop the car in the shortest possible distance.
Regenerative braking relies on the wheels to continue turning the electric motors causing them to act as generators feeding current back into the battery system. Hence, slightly longer stopping distances, translates into more energy recovered into the battery than lost as heat. It would not surprise me that the car was performing as intended until CR compared its stopping distance to other vehicles.
The issue then becomes, determining when safety (e.g. need for short stopping distance in a sudden braking event) overrides the desire to be energy efficient and how the vehicle's computer tunes the braking system when under "manual control" of a driver. I wonder if the car can detect when there is an obstacle in its path vs not.
Maybe the computer had the capability to override the driver and soften the braking force applied when it detects there is not a immediate need to stop (recovering more energy) vs applying full braking force when it senses an object in its path? It would be interesting to see if there were significant differences in braking distance between a simulated collision vs just jamming on the brakes for "no apparent reason"?
No. A couple people pointed out that if the brakes were physically too small then firmware won't do anything
Not sure which thread you were reading, but the last article here on Slashdot talking about Consumer Reports not recommending the Model 3 there ABSOLUTELY were people unbelieving of such a software update being possible. Seriously, go take a look.
Slashdot is just one of the homes of TTBs and Musk-bros and it's a big circle jerk.
Okay, we DEFINITELY aren't reading the same Slashdot.
We've seen a whole lot of holier-than-thou criticism (from people who probably write shitty code day-in and day-out) that "cars shouldn't have bugs", a lot of implication that Tesla is the only manufacturer ever to have a bug found in the field, and lot of exaggeration about the magnitude of the bug, and then some clearly insane accusations that Tesla just paid off Consumer Reports--but you're the first person I've seen make a reasonable engineering criticism.
Bugs are inevitable, they are attracted to complexity. A self-driving system is going to have bugs. The infotainment system is going to have bugs.
But, the brakes should not have bugs. It's such a critical system that the code should be blindingly simple, trivial to formally verify, and the engineers should be *shocked* when there's a bug found. If not, then the system should be simplified until that's the case.
That's probably naive, because with ABS braking, auto-gen, etc.--there are probably good reasons for some complexity separating the user pushing their foot down and the brakes being engaged--but I agree that your question is the right one to pose.
Still not clear what exactly was changed in the firmware update.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
This reminds me of a CR article I read about testing VHS tapes. They mentioned that most VCRs have noise reduction circuits, and so they disabled them so that they could see the actual tape performance. This seemed like a mistake to me: If one brand of tape happens to suffer twice as much from a problem that the common noise reduction circuits can deal with, but half as much from some problem that they cannot, then won't that tape be preferred by consumers? Maybe, but it wouldn't have been by CR's testers, as they weren't looking at the tape's performance the same way that ordinary users would.
I also once read an article by someone who makes refrigerators about how they intentionally make them less efficient than they can be, just because CR tests refrigerators in a room that's 85 degrees. I suppose that to CR, this makes sense since they're giving the refrigerator a higher work load and thus maybe amplifying any problems, but this guy was saying that when you design refrigeration equipment, you design it to operate over a certain temperature differential. If most people's houses are 75 degrees, then you get a more efficient refrigerator if you design it with that in mind, but they had to design for the 85 degrees that CR tests at.
It's kind of a problem with "smart" people that they think that any idea that "makes sense" is automatically a good idea. This "ICE cars only have friction brakes, so we should compare the Tesla to them with its regenerative braking disabled" is something that I'm sure makes sense to some "smart" people, but it doesn't reflect what the consumers are actually going to experience if they buy a Tesla.
Consumer reports is trash. To recommend people buy a car that Tesla goes to great lengths to stop independent mechanics from repairing and refuses to do recall safety repairs on if it has been salvaged is insane.
Only idiots buy this shit.
Pickups and such have above average brakes, sometimes they need to stop with 500kg of extra junk in the trunk. Or more, depending on the type.
"You people need to grow up and get a life." - pot kettle black.
"The hands that help are better far than lips that pray." - Robert Ingersoll (1833-1899)
This test is probably designed to promote disk brakes over drum brakes, by disk brake makers, by lobbying SAE to use this as the "standard".
You've obviously never been driving down a mountainside and experienced brake fade.
With ABS all braking distances are limited by the tire not the braking power
You obviously have no idea what ABS actually is.
ABS independently measures the speed of each wheel, runs a hydraulic pump, and uses valves to independently brake each wheel. All of which is controlled by an ECU.
For a start ABS performs cadence braking (far more rapidly than a human operator can). The on/off intervals can affect braking distance, therefore firmware can affect distance.
Secondly it applies different braking to each wheel. Firmware could mean some wheels are braking less than they should.
Wheel speed is measured via sensors which generate a signal from an EM field. The interpretation of the voltage emitted by sensor is affected by the firmware and can result in more or less braking than is ideal for a given speed.
There's plenty of places a firmware update can affect the ECU's behaviour and therefore the braking distance.
Or did you think there was only one ABS ECU design? Because that'd be wrong as each one is tailored to the individual characteristics of the car.