Sam

3.2K posts

Sam

Sam

@ideas9319

Just yapping

Chicago, IL Katılım Aralık 2019
557 Takip Edilen165 Takipçiler
Sam
Sam@ideas9319·
@Noahpinion This is the Republican version of that woman saying “Obama is gonna open up my purse and give me money” in 2008. Just as stupid, and arguably much more destructive
English
0
0
1
136
Sam
Sam@ideas9319·
@ajlamesa That's only extremely scary if you think those drivers weren't going to be distracted without FSD. The numbers don't really bear that out. FSD is, at worst, harm reduction that mitigates the impact of otherwise-dangerous behaviors that people were already engaging in.
English
0
0
0
4
Sam
Sam@ideas9319·
@moreland_shane @wholemars You would not believe how many people act like it's insane for states to require front plates. People get really pissed off about it. Makes no sense to me, you're exactly right that this is a great example of why some states require it.
English
1
0
1
13
InShane76
InShane76@moreland_shane·
@wholemars This is prime example of why states like Texas require front plates. Just saying.....
English
2
0
2
105
Wes
Wes@wmorrill3·
Every one of these dots is an actual crash from the fleet. Real world speeds, collisions, and people. Not just the regulatory test cases. The richness of this data is what enabled the result. With simulation we can replay the crashes and measure the forces on the human body model. Then sweep through restraint deployment times to find that deploying earlier gives the time for the bag to be inflated optimally and seat belt pretension before the occupant has moved out of position. But it takes time for crash accelerometers to be certain. Lowering that time threshold risks unwanted deployments. Using vision gives the vehicle confidence to reduce that timing. The camera sees the impending impact and together with the sensors tell the restraint controller to reduce the filter and act sooner. The Y Axis shift in Predicted Injury Severity is based on sensors in the human body models from rerunning the crash simulations with the faster detection threshold. Such a reduction in injury severity across the spectrum is unheard of, let alone doing this via an update over the air. I'm extremely proud of the analysis team's work and dedication. Going above and beyond to ask "we have the safest car on the road but can we make it even safer?" And then working with the vision team to build the predictions needed to make it happen. Rigorously tested in simulation and then in physical crash testing. Now deployed and improving lives. I watch the video on loop and just imagine each dot, a person.
Tesla@Tesla

Tesla Vision allows us to deploy airbags up to 70 milliseconds earlier if your Tesla detects an unavoidable collision This can be the difference between serious injury & walking away from a crash

English
385
832
6.1K
15.9M
Sam retweetledi
Raines
Raines@raines1220·
v14.3.2 is the first version where my reaction is no longer, “Wow, that feels very human-like.” Now it is more like, “Wow, a human could never do that.” Usually, I try to understand the “personality” of an FSD version. Is it chill? Is it too dramatic? Is it overly cautious? But v14.3.2 feels different. It almost has two personalities. In normal daily driving, it is smooth, natural, and human-like, almost like a friend driving you around. But when an emergency happens, it no longer feels like a friend. It becomes something else: a cold-blooded machine. Calm, smart, and fast. It handles dangerous situations correctly before a human even fully understands what is happening. That is why it feels trustworthy, but also kind of “scarily capable.” In certain aspects of driving, it is already more capable than even the best human drivers. And honestly, v14.3.2 is the first version that makes me seriously think FSD could save millions of lives that human drivers would never be able to save.
English
12
29
330
11K
Sam
Sam@ideas9319·
@DouthatNYT There is a certain type of person for whom pulling a prank like this would be very funny
English
0
0
0
25
Ross Douthat
Ross Douthat@DouthatNYT·
Category of weird things I'd like to understand: Why someone affiliated with or pretending to be affiliated with the U.S. government decided to rent a rural Airbnb to one-shot a group of Pentecostalist pastors with a very specific theory of UFOs: x.com/UAPDr/status/2…
Dr. Dan@UAPDr

Instead of speculating about a “government disclosure prep meeting for pastors,” listen to two guys who were actually there walk through the whole conversation. Not broad prep for churches. Something much smaller. And much stranger. #UAP #Disclosure #aliens

English
86
60
730
111.6K
Sam
Sam@ideas9319·
@mikepat711 Glad you're enjoying it! 14.3.2 has been my favorite build yet. I have to babysit its tendency to pass when it shouldn't (like in diverging lanes on the interstate), but other than that it's amazing.
English
0
0
0
17
Mike P
Mike P@mikepat711·
First 50 miles on 14.3.2 seem much more predictable (in a good way) than 14.3.1. Idk I was doing work the entire time but the car must’ve done a great job because I’m parked at my destination. FSD is such a mindfuck dude. They really did it. Imagine 5 years from now lmaooo
English
15
1
146
3.7K
Sam
Sam@ideas9319·
@WorldlyReviewer Every single issue I've had with 14.3.2 is a result of it trying to pass someone when it shouldn't. It consistently tries to pass people on I-75 in the I-85 exit-only lanes, causing it to get stuck and forced to go the wrong direction on I-85, for example.
English
0
0
1
19
Abhimanyu Yadav
Abhimanyu Yadav@WorldlyReviewer·
Here is v14.3.2 merging like an idiot. There was a clear sign that said lanes are merging too. This was too close. I was about to disengage but the car reacted fast. Seems like this version is doing these erratic things because its reaction time is so fast and it can recover and it leads to only close calls. Merging issues like this are common with this version while v14.2.2.5 was almost flawless in that regard. Hopefully future releases polish stuff like this. The random behaviors like this what is stopping me from calling v14.3.2 almost perfect @Tesla_AI @aelluswamy @yunta_tsai
English
100
6
272
28.7K
Sam
Sam@ideas9319·
@mikepat711 My assumption is that anything marked "preference" is lowest priority for getting looked at or fixed.
English
0
0
0
19
Mike P
Mike P@mikepat711·
When FSD gets into a parking lot and passes tons of perfect spots like an idiot because it wants to pointlessly get to the nav pin before it looks for parking, do yall want me pressing nav as the reason for disengagement, or preference?
English
75
1
229
14K
Sam
Sam@ideas9319·
@mikepat711 You’re gonna get sick of hearing this, but…it’s a lot better on 14.3.2 lol
English
0
0
1
76
Mike P
Mike P@mikepat711·
Nag is definitely fucking worse on 14.3.1. So annoying. 14.2.1 nag was the dream
English
31
0
79
6.3K
Sam
Sam@ideas9319·
@pbeisel @mikepat711 14.2.2.5 didn’t seem to care unless the deer entered the road. 14.3.2 proactively slowed down because it knew the deer could easily enter the road, even if they weren’t giving any sign that they would.
English
0
0
0
17
Sam
Sam@ideas9319·
@pbeisel @mikepat711 Great example of RL: Last night my car was driving below the speed limit on a two lane road for seemingly no reason. Then I realize there are four deer maybe 20 feet off the road up ahead. Not in the road, not moving towards the road, but FSD knew they were trouble.
English
1
0
0
31
Mike P
Mike P@mikepat711·
My current feelings after a few weeks of 14.3.1 is that it was pretty significantly over-hyped. There are hints of higher intelligence during certain edge scenarios, but in routine day-to-day driving, I find it less consistent than 14.2.2.5. This post is going to sound negative, but please remember that it’s still V14, and still mostly incredible. I am just judging it against the hype, and relative to 14.2.2.5. There are a lot of “I’m going to go, OH FUCK, maybe I shouldn’t” moments that did not exist for me on 14.2.2.5 at all. Or “I’m going to go. AHH PROBABLY SHOULDNT HAVE” moments. Just not as polished. Also random moments where it’ll miscalculate deceleration curve and AEB it at the last second of a deceleration because it realizes it’s not slowing down fast enough. These are not super frequent, but frequent enough to notice it when it really never happened on 14.2.2.5. Parking seems better I guess, but honestly not that much different. At first it felt a lot better, but after more testing has ended up feeling pretty much the same as 14.2.2.5. Sometimes it crushes it, sometimes it’s annoying af. I think whatever was significant about 14.3 to make it “the final piece of the puzzle” either didn’t actually happen yet, or just exists under the hood today and isn’t visible to the customer yet. Definitely eager to see how this series evolves with point releases.
English
53
4
162
11.6K
Sam
Sam@ideas9319·
@mikepat711 I’m surprised that you aren’t on 14.3.2 yet, I’ve been really happy with it. It’s not quite as smooth 14.2.2.5, but close. The speed of decision-making seems significantly better than 14.2.2.5. It just feels so…decisive? Like input and output happen at the same time.
English
0
0
1
53
Sam
Sam@ideas9319·
@Brady_H Sugar is yummy, and I refuse to pretend that it’s not
English
0
0
2
186
Brady Holmer
Brady Holmer@Brady_H·
I actually look forward to taking gels on runs. They taste great. Give you energy. Make you faster.
English
21
2
175
17.1K
Sam
Sam@ideas9319·
@SenWarren Senator, I would not bark up this tree if I were you lol
English
0
0
0
3
Elizabeth Warren
Elizabeth Warren@SenWarren·
Spiking fuel prices from Trump’s war was the nail in the coffin for twice-bankrupted Spirit airline. FWIW, JetBlue merger failed because a judge, appointed by Ronald Reagan, said the deal was illegal. Republicans are desperate to shift blame from higher costs hitting families.
English
14.4K
1.5K
8K
2.5M
Sam
Sam@ideas9319·
@NWSAtlanta Bone dry here in Marietta.
English
0
0
0
22
NWS Atlanta
NWS Atlanta@NWSAtlanta·
☔️Light to moderate rain is pushing into Georgia this evening and will continue into Saturday morning, especially near & south of I-20. 🌧️When the rain ends by late morning into early afternoon, the highest totals will be in central GA where amounts over 1-1.5" are likely. #gawx
GIF
English
2
9
42
3.5K
Sam
Sam@ideas9319·
@DeGrawMuff @JeremiahDJohns Chicago drivers are the most antisocial pieces of shit I have ever seen. It’s like they are actively trying to kill someone.
English
0
0
0
10
Muffin DeGraw
Muffin DeGraw@DeGrawMuff·
@JeremiahDJohns Dangerous drivers should be treated like home invaders. Drive around the entire Chicagoland area for work. More than once someone has been genuinely trying to ram their car into mine. I am not aggressive, I do not start anything. My crime is going a little too slow
English
1
0
2
561
Jeremiah Johnson 🌐
Jeremiah Johnson 🌐@JeremiahDJohns·
If someone has MORE THAN A HUNDRED VIOLATIONS, the time has come to permanently take their driver's license and throw them in jail. This is "300 professional criminals do the majority of all shoplifting" but for cars. Don't let insane people kill us with their cars.
Governor Kathy Hochul@GovKathyHochul

If you’re speeding over and over again, it’s not a mistake. You’re making our streets dangerous. I’ve got a plan to crack down on super-speeders and make New York City safer. Let’s get it done.

English
99
476
7K
330.8K
Sam
Sam@ideas9319·
Today has to be a top-ten meme day on Twitter, right?
English
0
0
2
20
CryptoProjection
CryptoProjection@CryptoQuantum42·
@ramit The last thing our education system needs is more money or tax breaks. We need upper management fired. More teachers hired disruptive kids removed from schools and real education taking place - not activists raised to over throw the system.
English
19
0
3
4K
Ramit Sethi
Ramit Sethi@ramit·
26-year-old Republican Franceska is shocked that Trump has defunded education instead of "tax breaks or helping with the education system" Let's see what else MAGA voters have to say
Ramit Sethi tweet media
English
107
182
2.6K
1.7M
Sam
Sam@ideas9319·
@ChuckCook It was like 6 inches away from the pole and wasn't stopping, I'm pretty confident it would have backed into it. I'm glad that I can get Tesla to look at that rather than someone being unhappy about it not shooting the gap on a random left turn or something.
English
0
0
1
11
Sam
Sam@ideas9319·
@ChuckCook I'm happy that we can at least designate interventions as "critical" for further review. On my first drive w/ 14.3.2 today, it almost backed into an unusually-placed light pole (seems to be the biggest issue for Robotaxi too). There was a big raindrop on the camera, but still!
English
1
0
1
35
Chuck Cook
Chuck Cook@ChuckCook·
Agreed
JS@JSX423

The biggest issue I have with the disengagement reason menu probably is the average person paying $100/mth just wants FSD to work. Not be a mandatory tester otherwise having UI elements blocked if FSD does something wrong. My 77 year old mom ain't your tester, @tesla_ai. 4 categories don't even come close to being enough to work. Actual capable testers, interested in these things, should sign up and get a full screen with 50+ categories.. @ChuckCook @DevinOlsenn @DirtyTesLa @AIDRIVR Potential categories - 1.Speed Issues Wrong speed (S) Ignored speed limit (S) Too fast (S) Too slow (C) 2. Highway / Freeway Phantom braking (S) Inconsistent traffic speed (C) Merge yield fail (S) Tailgating (S) Hesitant passing (C) Bad construction behavior (S) Missed exit (C) Aggressive lane change (S) Lane change fail (S) 3. City / Urban Ran red / late stop (S) Crept intersection (S) Poor left turn (S) Pedestrian yield fail (S) Unnecessary stop (C) Bad route (C) Hesitant at stop (C) Missed school zone (S) Hit curb (S) Emergency vehicle fail (S) Drifted lane (S) Missed hand signal (S) 4. Parking / Low-Speed Bad parking spot (C) Poor parallel park (C) Driveway curb hit (S) Missed low obstacle (S) Garage door fail (S) Bad reverse distance (S) Speed bump issue (C) Parking lot pedestrian (S) 5. Intersections & Turns Turn too wide/tight (C) Bad turn alignment (C) Rolled stop (S) Cut off traffic (S) Slow post-turn accel (C) Bad roundabout (S) Ignored turn lane (S) Late turn signal (C) Wrong stop position (S) Bad yellow light (S) 6. Weather / Conditions Traction loss (S) Hydroplaning fail (S) Low visibility drive (S) Insufficient ice slow (S) Bad unpaved road (C) 7. General / Other Too close cyclist (S) Ignored debris (S) Ignored pothole (S)

English
43
11
318
27.8K