One thing Tesla and Comma.ai overlooked in self-driving • Hacker News Board

Re: One thing Tesla and Comma.ai overlooked in self-driving

#11

Post by artificialprint » Wed, Jun 04, 2025, 10:30 PM UTC

hyuuu wrote: post #6

one point the author seems to be missing as to why we need full self driving everywhere is in the logistics industry. We are talking about interstate driving, from point A to point B, in this case it might involve industria environments such as a truck stop, weighing, loading/unloading etc, something that requires decision making across different environments (urban, highway, dense residential etc). Tesla FSD has the…

I remember Waymo had their "Via" division, seems like they been hitting brakes on that one. I think they had people driving the last mile, which makes sense. Again, freeway crashes are always gonna be the most lethal and the worst, especially if we are talking about semis.

Government should make an AV lane for trucks!

Re: One thing Tesla and Comma.ai overlooked in self-driving

#12

Post by laweijfmvo » Wed, Jun 04, 2025, 10:31 PM UTC

quantified wrote: post #2

If Tesla's Robotaxis become Bing to Waymo's Google, that'll be quite an impressive leap from where Tesla is today.

as a Waymo lover and a Tesla owner who won't use FSD, I think they have a long way to go.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#13

Post by UltraSane » Wed, Jun 04, 2025, 10:31 PM UTC

I love how people ignore the fact that Tesla's FSD is no where near good enough to do what Waymo does.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#14

Post by modeless » Wed, Jun 04, 2025, 10:31 PM UTC

Ataraxic wrote: post #8

hartator wrote: post #5

The main downside is not just a cost thing but its waves are blocked by moist in air, rain, and dense fog. So, it will also degrades to just use cameras in this conditions.

From the recent tesla acme test, where the tesla runs into a wall containing a picture of a road, they also tested rain and water blocking two mannequins. The lidar equipped vehicle managed to stop before hitting the mannequins while tesla's autopilot did not. The distance before automatic braking triggered was reduced iirc. Sure lidar is degraded by rain, but I think the conclusion that when it rains the whole syste…

That test was not using FSD. When retested on the latest version of FSD, the car stopped. https://www.youtube.com/watch?v=7cxTO8g47_k

Re: One thing Tesla and Comma.ai overlooked in self-driving

#15

Post by UltraSane » Wed, Jun 04, 2025, 10:32 PM UTC

modeless wrote: post #3

> it will probably not be very safe, otherwise nothing would stop Tesla from rolling out “the safe” autopilot to car owners today, unless you think that something will change in one month and we are getting some kind of breakthrough in software, so it seems like taxi will run today’s software. The last version of FSD to increment the major or minor version number was released last year. There haven't been any big upg…

The most recent release of FSD is no where near good enough to do what Waymo does.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#16

Post by henry2023 » Wed, Jun 04, 2025, 10:32 PM UTC

ifdefdebug wrote: post #9

Animats wrote: post #7

Tesla is probably going to fake it using human remote drivers. Tesla has been running ads to hire the drivers.[1] That has all the costs of the self-driving hardware, plus the costs of the remote drivers, but it maintains the illusion that Tesla is a growth company. [1] https://insideevs.com/news/760863/tesla-hiring-humans-to-con...

No way I get into a car that has a human driver who literally has no skin in the game.

They can always fire the driver if he/she causes a few fatalities.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#17

Post by UltraSane » Wed, Jun 04, 2025, 10:33 PM UTC

modeless wrote: post #14

Ataraxic wrote: post #8

Earlier quoted context omitted.

From the recent tesla acme test, where the tesla runs into a wall containing a picture of a road, they also tested rain and water blocking two mannequins. The lidar equipped vehicle managed to stop before hitting the mannequins while tesla's autopilot did not. The distance before automatic braking triggered was reduced iirc. Sure lidar is degraded by rain, but I think the conclusion that when it rains the whole syste…

That test was not using FSD. When retested on the latest version of FSD, the car stopped. https://www.youtube.com/watch?v=7cxTO8g47_k

Can you provide a link to this?

Re: One thing Tesla and Comma.ai overlooked in self-driving

#18

Post by modeless » Wed, Jun 04, 2025, 10:36 PM UTC

UltraSane wrote: post #15

modeless wrote: post #3

> it will probably not be very safe, otherwise nothing would stop Tesla from rolling out “the safe” autopilot to car owners today, unless you think that something will change in one month and we are getting some kind of breakthrough in software, so it seems like taxi will run today’s software. The last version of FSD to increment the major or minor version number was released last year. There haven't been any big upg…

The most recent release of FSD is no where near good enough to do what Waymo does.

I use it every day and you are right! But my point is with six months of improvements in the software at the rapid pace demonstrated last year, combined with a boost from bigger models running on more capable compute hardware, I think it's possible that it could be close. Certainly up to the point where Waymo was when they started public rides three years ago.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#19

Post by einpoklum » Wed, Jun 04, 2025, 10:36 PM UTC

henry2023 wrote: post #16

ifdefdebug wrote: post #9

Earlier quoted context omitted.

No way I get into a car that has a human driver who literally has no skin in the game.

They can always fire the driver if he/she causes a few fatalities.

Are you sure the driver won't have a poster up above his screen like this one? :

https://is.gd/Lw37Su

Re: One thing Tesla and Comma.ai overlooked in self-driving

#20

Post by Macha » Wed, Jun 04, 2025, 10:43 PM UTC

Animats wrote: post #10

wagwang wrote: post #4

Wait... the big insight here is that people live in cities? Brother, this has been talked to death in the "is level 4 good enough" conversation.

Right. In the US, once the top 12 taxi markets have been served, Waymo will have most of the industry.[1] That's all the cities with more than 1000 taxis. Serve the next 10 cities, and it's down to Saginaw, MI, with 41 taxis. In San Francisco, Waymo has passed Lyft and is gaining on Uber. [1] https://en.wikipedia.org/wiki/Taxis_of_the_United_States

Given these counts are all individually sourced to different articles (or not sourced at all in some cases), I suspect that wikipedia list is rather incomplete. For example, I suspect Birmingham, Alabama, to pick a city at random, has more than the 41 taxis of the 20th placed in Wikipedia's list.