Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A look under our trunk: what's in our compute

waymo.com

Post reply on HN
Print view View original

81–90 of 116 posts

Re: A look under our trunk: what's in our compute

#81

Post by xnx »

NicuCalcea wrote: post #75
cyberax wrote: post #43

Earlier quoted context omitted.

Yes? Why is this strange? Drivers emit a lot of CO2, just like every person in the US. A driverless car allows these emissions to be reallocated to other more productive areas.

You're counting emissions from a person's breathing as part of the vehicle's emissions?

People take their salaries and buy food, clothes, heating and air conditioning, cars and fuel, and plane tickets, and furniture. All of those things generate emissions in their production and/or transportation. This is why dollars is such a good proxy for co2 emissions.

Re: A look under our trunk: what's in our compute

#82

Post by DiscourseFan »

mschuster91 wrote: post #73
phoghed wrote: post #19

Earlier quoted context omitted.

Different 30 people in the US. I’d drop my car if Waymo was around and financially made sense. You’ll never catch me wasting my time on a bus though. We’ll never even be close to Europe or elsewhere wrt where public transit was even decades ago in those places. People here only ride the bus when there’s no other option for them. Waymo could make a bus too anyway, especially after they’ve been in an area a while and h…

> Waymo could make a bus too anyway Oh hell no. Cars are bad enough from the damage they can cause by kinetic energy, but a bus? Especially a bus where people don't wear seatbelts or, worse, are standing?

It would probably be safer than most bus drivers currently, so I don't see the issue. The discomfort around damage being caused by statistical anomalies vs individual error is not a rational one if the former is, in the aggregate, smaller than the latter. Individuals have stories, but societies operate more closely in line with the laws of physics: probabilistic at the small scale, deterministic more broadly. From that view, we are just replacing one statistical possibility with another, one which will result in less people dying; or, more importantly, less likelihood you or your loved ones will die because of a bus accident.

Re: A look under our trunk: what's in our compute

#83

Post by xnx »

m463 wrote: post #60
xnx wrote: post #5

Apologies for repeating myself, but it's hard to convey how far ahead Waymo is in every area: sensors, vehicles, training data, markets served, simulation, infrastructure, operations, regulatory. They're growing fast, but I do wonder what they see as the bottlenecks: municipal approval, training, manufacturing? It's really unfortunate that lobbying is slowing self-driving rollout in some of the biggest and most impor…

So are they? I know that tesla is behind, but they have so many simulation miles, several generations of their own hardware, and lots of people behind the wheel taking over. I've also seen the autonomous waymo vans, which seem nice too. Is there a "theoetical" ahead and a "practical" ahead? How do the finances look and does financial solvency count? will someone run out of money first? There have been two waymo artic…

> Is there a "theoetical" ahead and a "practical" ahead?

Yes. Waymo is doing something like 400K fully autonomous rides per week and Tesla is doing hundreds or low thousands of supervised rides.

Re: A look under our trunk: what's in our compute

#84

Post by xnx »

AndrewKemendo wrote: post #56
xnx wrote: post #5

Apologies for repeating myself, but it's hard to convey how far ahead Waymo is in every area: sensors, vehicles, training data, markets served, simulation, infrastructure, operations, regulatory. They're growing fast, but I do wonder what they see as the bottlenecks: municipal approval, training, manufacturing? It's really unfortunate that lobbying is slowing self-driving rollout in some of the biggest and most impor…

I’m glad you said it because as a 40 something nerd roboticist taking a Waymo is nothing short of a miracle to me. I mean drafting code on my phone with chatgpt in a Waymo is legitimately some Jestons shit and I’m not hearing otherwise. Amazing

And that's not even pointing out that also makes Waymo the most advanced robotics company.

Re: A look under our trunk: what's in our compute

#85

Post by ButlerianJihad »

mschuster91 wrote: post #73
phoghed wrote: post #19

Earlier quoted context omitted.

Different 30 people in the US. I’d drop my car if Waymo was around and financially made sense. You’ll never catch me wasting my time on a bus though. We’ll never even be close to Europe or elsewhere wrt where public transit was even decades ago in those places. People here only ride the bus when there’s no other option for them. Waymo could make a bus too anyway, especially after they’ve been in an area a while and h…

> Waymo could make a bus too anyway Oh hell no. Cars are bad enough from the damage they can cause by kinetic energy, but a bus? Especially a bus where people don't wear seatbelts or, worse, are standing?

> don't wear seatbelts

I just want to point out that the safety record of public transit drivers is typically impeccable. There is no safer vehicle out on the roads except buses and trains. I have been a rider of public transit for 50 years, minus about a decade when I owned a car, and in 3 different major cities, and literally I've never been aboard a bus that hit another vehicle, hit any obstacle, or had a reportable accident. Literally never happened. (I was aboard a light rail train that collided with an SUV, but that's a whole other ballgame.)

I've never felt unsafe without a seat belt on board a bus. Seat belts in cars may be a bit of a panacea, but I challenge you to find any metropolitan area where the buses have accidents, much less injury accidents, on a weekly basis.

Re: A look under our trunk: what's in our compute

#86

Post by ButlerianJihad »

novia wrote: post #71

> Redundant: Because there is no human to take over, proactive safety and redundancy are natively built in. Our compute is designed like two independent engines. While they normally operate as one unit running full parallel workloads, if one experiences a fault, the other seamlessly takes over. uhhh i've been in a situation where a remote human took over before it's rare but it can happen (or at least it could 1.5 ye…

Waymo has zero "remote human drivers". It is literally impossible to "drive" a Waymo by remote commands.

The Waymo agents are constantly monitoring the fleet for exceptions. When an exception is raised, a human reviews the footage and the situation. They can send commands to the car and suggestions about which way to branch or a course of action to choose. So ultimately, it is the Waymo Driver doing the actual driving.

If a Waymo gets really jammed up, they can indeed send in a human driver, who will appear in person, disengage the Waymo Driver automation, and drive the car normally. This has happened to me only once in 3,000 miles. The Waymo Driver, in my experience, is pretty good about keeping passengers informed when it's a bit delayed or confused by a complex situation. And the CSR will call in if they need to.

But no, it is a misconception that people have: Waymos cannot be "driven remotely" by any means.

Re: A look under our trunk: what's in our compute

#87

Post by ButlerianJihad »

munificent wrote: post #36

Interesting that the article uses "ML" everywhere instead of "AI". I'm sure some marketing person chose that term deliberately, and I wonder what went into that decision.

First of all, machine learning is a distinct category of A.I., so they are being more specific about their methods. If we get right down to it, "A.I." is, and has always been, a nearly meaningless buzzword that means whatever you want it to mean. In 2026 it is highly stigmatized by haters, anyway. So I appreciate that a techblog endeavors to explain exactly what sort of A.I. they're employing in these systems.

Re: A look under our trunk: what's in our compute

#88

Post by NicuCalcea »

xnx wrote: post #81
NicuCalcea wrote: post #75

Earlier quoted context omitted.

You're counting emissions from a person's breathing as part of the vehicle's emissions?

People take their salaries and buy food, clothes, heating and air conditioning, cars and fuel, and plane tickets, and furniture. All of those things generate emissions in their production and/or transportation. This is why dollars is such a good proxy for co2 emissions.

Surely a person driving 18 people will emit less than each person driving themselves? I can see how dollars can be a proxy for macroeconomic stuff, for example by comparing countries, but it doesn't make much sense for comparing individual units of transport.

I don't understand why we have to do napkin maths to guess what's more polluting. There is countless research showing buses emit less than individual cars per passenger per km, self-driving or not.

And there's no reason buses can't be self-driving as well, if anything, them driving slower, on dedicated lanes and on fixed routes would make it much easier.

Re: A look under our trunk: what's in our compute

#89

Post by MostlyStable »

nxc18 wrote: post #65
MostlyStable wrote: post #64

Earlier quoted context omitted.

I find the juxtaposition of "can't stop bragging" and "is not a good marketing term" to be fascinating. Have you considered that they know these recent reports aren't a good look for them but that they feel compelled (either because they know it would be far worse when it inevitably got out on it's own or maybe even from some moral sense) to report somewhat truthfully?

Two different audiences. Unfortunately for them, messages cross streams. I think it is a good message if you’re trying to scare people into giving you money (or else they lose out on unlimited profits down the road).

Who is this audience that thinks unasked-for and un-controllable felonies is a good message, or who wants to fund that?

Re: A look under our trunk: what's in our compute

#90

Post by Peanuts99 »

cyberax wrote: post #29
realo wrote: post #16

Earlier quoted context omitted.

30 people in 30 waymos vs 30 people in an electric bus? No contest.

An average bus load in the US is around 18 people. So if you just compute the raw bus emissions, then an electric bus is maybe about 2x more efficient than personal EVs. But wait, there's so much more! A single bus/subway ride costs about $20, which is a surprisingly consistent figure for various US cities. In NYC it's $17, for example. I'm talking about the _true_ cost, as in: "yearly operational budget divided by t…

A single bus journey is $20? That's insanely expensive.
Post reply on HN