They need to add radar back. Ability to "see" slow downs or stopped traffic even if your vision is obscured is a great benefit for self driving.

They need LiDAR. The biggest problem for FSD is still around bounding box detection.

I just don't believe that you can infer the dimensions of objects using stereoscopic images with a reliability that you need to make FSD work.

How do our eyes work?

Far more complex sensors than cameras attached to some of the most complicated and least understood processing systems running a general intelligence with millions and millions of years of training.

Also, when you're in the car, you probably move your head more than you realize. Moving your head around and looking around also gives you better understandings of distance. Far more than a couple of cameras feeding a few megapixel images into a ML model.

Why make the difficult task of self-driving any harder, by artificially limiting the sensors you use? Planes have rigid aluminum wings and jet engines instead of bone muscle and feathers.

By us constantly moving them around in three dimensions unlike a car.

And also having a computer behind them that deeply understands what an object is, the forms it should take and what its expected behaviour should be. We don’t ever confuse billboards for real people.

>By us constantly moving them around in three dimensions unlike a car.

Fair but to compensate the car has many more cameras than we do.

For looking forward Tesla's have three fixed cameras looking from the rear view mirror. All the other cameras do not have stereoscopic vision and are not looking forward. A camera looking out from a side pillar isn't helping gauge the distance to something far up the road.

On top of that these cameras can't move, can't be re-aimed, and have generally far worse dynamic range than our eyes.

Our eyes work in concert with a supercomputer accessing huge stores of contextual data (far eclipsing anything in a Tesla neural database) to understand and react to unique situations.

Rather poorly, given the rate of traffic accidents by humans.

That's your opinion, but Tesla does not "need" to add anything from the point of view of actual Tesla customers!

Tesla cars are selling faster than they can make them and adding a Radar will slow down the manufacturing, not make it faster.

Tesla is currently being sued by actual customers based on this exact issue. They've promised FSD, and can't deliver it - especially without lidar.

> adding a Radar will slow down the manufacturing, not make it faster

Tesla told the FCC that it plans to market a new radar starting next month.

https://electrek.co/2022/12/06/tesla-radar-car-next-month-se...

> Tesla cars are selling faster than they can make them

If this is true, why are they reportedly cutting production and offering incentives? If they were booked solid, no incentives would be needed.

https://www.reuters.com/business/autos-transportation/tesla-...

> Radar will slow down the manufacturing, not make it faster.

? They are adding radar, though: https://electrek.co/2022/12/06/tesla-radar-car-next-month-se...

But Tesla's software version is 10.69.x....69...get it...LOL!!

I agree. Tesla FSD has so many obvious limitations that can be worked out on closed courses that it has no legitimate reason to be tested on public roads. For example, it cannot drive directly into the sun. That's a flaw they could work out on the test track (by adding different sensors). There needs to be regulatory intervention to force FSD off the road.

Curious you mention this I was using it this morning driving directly into the Sun... even without our lanes painted after they repaved the roads it kept me in the right part of the road and even engaged the turn signal automatically as it stopped at the traffic light before our right turn... I know they had issues with direct Sun few years ago but it would seem to me driving with it enabled into the sun works just fine...

Might’ve been because it perfectly obscured the straight-ahead view, given the hills where he’s driving. While not perfect, the lenses on Tesla’s forward-facing camera is good at retaining detail if the sun is above eyeline. https://twitter.com/greentheonly/status/1200626377097129984?...

Yeah, I see that is a pretty intense road and situation... my road is much much straighter... I'll try to capture it and share definitely not difficult...

What is a "pretty intense road and situation" for you is some people's daily commute.

How is this an intense road? It looks pretty wide and clear of traffic with a few parked cars. If it can't handle this, it's got no chance in an average city in Western Europe.

That looks like an extremely normal residential road with no other drivers or pedestrians.

I don't see how that's supported here? I mean, clearly Tesla does its own closed-track testing. I've seen coverage of that in the past. They don't release public software the doesn't pass these kinds of tests. Likewise it's not like Waymo restricts their testing to closed tracks, they have vehicles on the streets too.

> One of the scariest parts about Tesla is that they don't even seem to know what they don't know.

I'm curious what the reference here is? Again, are you taking coverage of Waymo's test environment as evience of its absence at its competitors?

Tesla claims they are safer than human drivers by fudging statistics about accidents per mile.

Waymo describes in detail how they test their algorithm against specific scenarios and makes statements about those experiments.

One company is trying to sell you cars and telling you what you want to hear, the other company is extremely careful with their statements.

> fudging statistics about accidents per mile.

How? The only problem with this page[0] is that they haven’t released 2022 stats. Otherwise:

> To ensure our statistics are conservative, we count any crash in which Autopilot was deactivated within 5 seconds before impact, and we count all crashes in which the incident alert indicated an airbag or other active restraint deployed.

0: https://www.tesla.com/VehicleSafetyReport

That "vehicle safety report" is 70 words, none of which answer the question of "how many crashes were there" or "how many people were injured or killed." One of the few numbers they give us, the percentage of incidents/hour for driving "with active safety but without autopilot," bounces around by like a factor of two from quarter to quarter, which doesn't make any sense. Literally every provided number is in Tesla's favor, and context is only provided in ways favorable to Tesla.

This is a press release, not a safety report.

I'm honestly curious what test tracks Tesla uses. Do they just rent private tracks for their road testing before it gets to public beta? It clearly isn't done onroad by safety drivers.

GM has Milford and Yuma + others, Waymo has Castle, Zoox uses Altamont Raceway, Nuro has a track at the Vegas speedway. I can't imagine Tesla's small Fremont loop and their winter track in Fairbanks are sufficient.

While both Waymo and Tesla are testing vehicles on the streets, and have done so for some time, their approaches could hardly be more different.

Tesla’s FSD is so terrifyingly bad at routine tasks (I used it for six months before giving up) that it’s natural to assume whatever closed track testing they did was ineffective. Or perhaps they did a lot, I don’t know—but it doesn’t feel like many of those lessons learned made it into the “production” system.

I really like AP in general, but for FSD Beta, I tend to agree. I've seen enough mistakes from even the really careful Youtubers that I don't understand why its still in the field.

And those are the mistakes that they were willing to show! To be clear, I mean situations where the driver needed to take over but either didn't, or didn't in sufficient time to avoid an illegal or dangerous maneuver.

From what I can tell of people posting FSD videos on YouTube, they are actively seeking adversarial conditions with a desire to show where it fails. I’m sure there are some YouTubers that are trying to sugarcoat FSD, but I haven’t seen any.

I partially agree. I mean, I find dirty tesla's videos to be pretty fair.

Having said that, he's also been really clear that the video doesn't always make it obvious just how many aspects of FSD are just plain weird. Even when its not actively failing, it moves in odd ways that are uncomfortable.

He's also said that earlier versions resulted in curbed wheels.

Yep, my mate curbed his front passenger wheel that way. But I dunno how you fix that without LIDAR sensors mapping the terrain around the vehicle.

Agreed, that's one way. The other way is multiple downward facing cameras with stereoscopic vision of the area immediately around the car. Mobileye demonstrated a system like that and it worked really well.

Tesla's cameras have huge blind spots near the car and aren't stereoscopic in all directions. They had ultrasonic sensors at one time, but those had blind spots and major limitations too.

There's no reference. It's just fashionable to hate on anything associated with Elon at the moment.

> They don't release public software the doesn't pass these kinds of tests.

Do we actually know that this is true? October of last year "internal QA" found regressions on an already public release and they rolled out an update in less than 24 hours after that was published. Both the fact that it was already public when QA found an issue and that they were able to push new public version in such a short time seems to me like they don't necessarily have a release gauntlet for each version, or at least not a very robust one.

https://www.cnbc.com/2021/10/25/tesla-rolled-back-fsd-beta-v...

And one of the scariest parts of Waymo is that they may never ship and it'll be 20 billion dollars down the drain

I mean, that’s good though right? If it doesn’t work it doesn’t work. Worst comes to worst taxi drivers still have a job.

I think you mean, worst comes to worst 1.35 million people still die in traffic accidents every year.

If companies really had a noble causing of reducing traffic fatalities I'm pretty sure the billions they spend could be better spent advocating for light rail and buses. And hey, thousands less people would die of lung diseases to boot.

I don't understand what you mean? Doesn't the Phoenix launch count? They've definitely shipped something to prod.

The thing is – nobody cares!

People can scream bloody murder till the cows come home, but Tesla for all its faults is king of the hill - literally more valuable the rest of the car industry combined. Musk is the invincible and irrespective of any shortcomings he's literally the richest person on Earth.

Tesla's approach is demonstrably the best approach in the court of the customer, irrespective of what we (others on the road) think or are put at risk.

"we (others on the road)" is not "nobody", and we definitely care. The law is supposed to intervene in tragedy of the commons cases like this for precisely this reason. The free market is known to have a few blind spots and this is one of them.

> literally the richest person on Earth

Not anymore https://www.bbc.com/news/business-63963239

Market cap has absolutely no relation to the "court of the customer", nor should it in terms of what safety features we allow or disallow.

"invincible"? He has literally lost over $100B this year.

Tesla's market share is eroding every year, and is expected to decline to less than 20% by 2025.

https://www.cnbc.com/2022/11/29/teslas-dominance-of-evs-is-e....

There are plenty of people like me, who are choosing non-Teslas when they buy EVs.

He sucks

Which is consistent with what they had said at the time when they removed the low-resolution radar they had been using. Specifically, they had said that radar would be useful if it were sufficiently high resolution. The radar in the works and rumored to be added soon is anticipated to be high resolution.

The really scary part is actually the one nobody in these comments seems to address. We are letting humans drive these things with barley any training on roads that are terrible designed for safety of the people around the vehicles.

This utter disaster of a situation leads to 10000s of deaths that has well known solutions.

But instead of preventing these deaths with well known low tech solution, a super expensive technological holy grail is gone somehow fix those problems.

Fully agree here but what's the "well known low tech solution", besides public transit?

I agree with your and the parent poster's opinion that the best safest option is to encourage and fund buses and trains