The distinction is that larger installations cannot form a single network. Before xAI's new network architecture, only around 30k GPUs could train a model simultaneously. It's not clear how many can train together with xAI's new approach, but apparently it is >100k.
If you define 'speeding' as driving above the speed limit, then FSD absolutely speeds.
"FSD (Supervised) can contextually drive above the speed limit or the predicted road speed to match the flow of traffic. [The Max Speed Offset Setting] specifies the maximum speed it is allowed to drive as a percentage offset applied to the speed limit of the road or the predicted road speed."
People can configure this to 40% above the speed limit if they wish. I'm not sure what the default is.
I own FSD. FSD will travel at the speed of traffic. The % above the speed limit you set doesn't change how it drives for the most part.
The only way FSD will go wildly faster than traffic is if you push the accelerator on purpose, at which point the car beeps and reminds you that it won't stop while you're pushing it.
Actual materials chemist here. A PhD in physics is not that relevant to understanding the real electrochemical nuts and bolts if a battery.
Just look up what a solid electrolyte interface (SEI) layer is to see how bush league current battery technology is. Compared to CMOS tech, which has benefited from trillions of dollars of R&D, batteries are in the stone age.
From the website:"the number of cars involved in a fatal crash were normalized by the total number of vehicle miles driven"
They assume every car drives the average number of miles. Of course a CR-V is going to score high, since they likely drive more than an average number of miles.
A 911 on the other hand probably drives a small fraction of the average.
This study is fairly low quality. The IHHS list 13.3 fatalities per billion miles as the actual number.
Are you talking about the debunking videos that highlight that the Cybertruck only beat the Porsche in the 1/8 mile and not the whole quarter mile? And that Cybertruck is still basically a 10 second truck?
It's honestly a absurd and funny to me how fast the truck is. It's like seeing an elephant fly. It's even funnier to see people get angry at it.
I definitely don't like marketing and ads, but that's literally how all ads work. It doesn't debunk some of the insane performance characteristics.
> Are you talking about the debunking videos that highlight that the Cybertruck only beat the Porsche in the 1/8 mile
No, I'm talking about the debunking videos and articles that show:
"We ran six quarter-mile drag races, and each one had the same outcome: The Porsche 911 Carrera T wins and the Tesla Cybertruck Beast loses... it’s not a particularly close race, either."
Re the 1/8: "We can say confidently that Tesla didn’t show the Porsche 911 Carrera T’s quickest possible run. In four out of six MotorTrend drag races, the Porsche 911 Carrera T beat the Cybertruck to the eighth-mile mark."
"The manual-transmission Carrera T has a 3,500-rpm limiter at standstill, and on a sticky, prepped drag strip, launching quickly requires getting off the line without letting the revs fall. Drop the clutch too fast, and the engine will bog, falling out of its powerband. It takes a slow, carefully modulated clutch release to get the perfect launch, which keeps the engine on boil and extracts a small amount of slip from the tires."
I'm sure Tesla was quite eager to make sure that they launched the Porsche optimally.
Yes. I get it. EVs have amazing acceleration. My brother in law has worked for, and owned, both Teslas and Rivians, so I'm no stranger to this.
But this was just another Tesla self-congratulatory event that needed "simulation" and "we didn't actually do it but we think it would go this way" puffery.
Hilarious that they used the base 911 as well, the one with slightly less horsepower than a BMW 340i. We’re talking Golf-R-with-a-chip-tune / stage 2 WRX levels of power here. Not exactly impressive in terms of actual sports cars.
Then you get to lateral grip and uhh, yeah 0.76G is worse than the average minivan.
911s are not designed for drag, especially the base model. If they wanted to compare it to something that’s good in a straight line but no good in corners, a Dodge Demon 170 stickers for under $100k. Comparable horsepower, too.
I have a HW3 2019 Tesla Model 3 with FSD it's amazing.
I think you can pick one up used for 25k or so. If you buy used from Tesla it sometimes comes with FSD for a super reduced rate (at least they used to). Check it out.
The worry is that HW3 (only 144 TOPS) will not be enough for true self driving. Today they committed to upgrading HW3 cars with FSD for free if they solve self driving with more powerful HW4. Take one for a test drive they're dope.
Not Tesla exactly, but Musk has gone all-in trying to get a man elected to be US President who consistently says climate change is a hoax, or words to that effect.
US oil production under the current administration is at 13.5M barrels per day. The highest ever. The US is shitting the bed on the energy transition. Meanwhile global solar cell production is slated to hit 2TW/year by the end of 2025 @ under 10cents/watt. China, the land of coal, is on track to hit net zero before the US. Both parties and all levels of government have a disgraceful record on climate change.
PS: For context 2TW of solar can generate about 10% of global electricity. Production capacity will not stop at 2TW. All other forms of electricity are basically doomed, no matter what the GOP says about climate change.
Both parties have a disgraceful record on climate change, but the GOP is still clearly much worse. High as US oil production is, Republicans complain that it should be higher. And Trump making the hoax claim dogma for his followers is incredibly damaging.
There are also endless videos of teslas driving into pedestrians, plowing full speed into emergency vehicles parked with flashing lights, veering wildly from strange markings on the road, etc. etc.
"works for me" is a very strange response for someone on Hacker News if you have any coding background - you should realize you are a beta tester unwittingly if not a full blown alpha tester in some cases
All it will take is a non-standard event happening on your daily drive. Most certainly not wishing it on you, quite the opposite, trying to get you to accept that a perfect drive 99 times out of 100 is not enough.
Considering HN is mostly technologists, the extent of Tesla-hate in here surprises me. My best guess is that it is sublimated Elon-hate. (Not a fan of my former neighbor myself, but let's separate the man from his creations.)
People seem to be comparing Tesla FSD to perfection, when the more fair and relevant comparison is to real-world American drivers. Who are, on average, pretty bad.
Sure, I wouldn't trust data coming from Tesla. But we have government data.
That seems an odd take. This is a technologist website, and a good number of technologists believe in building robust systems that don’t fail in production. We don’t stand for demos, and we have to fight off consultants peddling crapware that demos well but dies in production.
I own a Tesla, despite my dislike of Musk, because it is an insanely fun car. I will never enable FSD, did not even do so when it was free. I see even the best teams have production outages. Until Tesla legally accepts, and the laws allows them to, legal responsibility, and until it’s good enough that it doesn’t disengage, ever, then I’m never using it and nobody else should.
A minimum bar, for societal harm, would be against an identical data set of US drivers. The data for human drivers covers vastly more situations than FSD does. FSD refuses to activate in those situations. So an apples-to-apples comparison doesn't exist. The FSD data is effectively cherry picked for ideal driving conditions. Tesla's claims that FSD is safer than the average driver are not supported by their data, and as others have said, either their statisticians are incompetent or liars. This is basic stuff.
However the minimum bar for me to activate it is "compared to me". I've never come close to driving under a truck or into a divider. I slow down driving into the sunset and use a baseball hat if necessary to make sure I can see.
How many miles does it have on the latest software?
Because any miles driven on previous software are no longer relevant.
Especially with that big change in v12.
The miles driven are rising exponentially as the versions improve according to company filings. If the miles driven on previous versions are no longer relevant how can the NHTSA investigation of previous versions impact FSD regulation today?
Given that the performance has improved dramatically over the last 6 months, it is very reasonable to assume that the miles driven to fatality ratio also improving.
Using the value of 1.33 deaths per 100 million miles driven vs 2 deaths in 2 billion miles driven, FSD has saved approximately 24 lives so far.
There's a lot of misleading going on, Elon trying to placate Russia until the Trump card is ready to be played, which must be launched very quickly all at once.
