Hacker News Reader

Comment by profunctor | Hacker News Reader

The reason is cost, LIDAR is expensive.

This information is out of date. LIDAR costs are 10x less than they were a decade ago, and still falling.

Turns out, when there's demand for LIDAR in this form factor, people invest in R&D to drive costs down and set up manufacturing facilities to achieve economies of scale. Wow, who could have predicted this‽

Cost is relative. LIDAR maybe be expensive relative to a camera or two but it’s very inexpensive compared to hiring a full time driver. Crashes aren’t particularly cheap either. Neither are insurance premiums.

Huawei has a self-driving system that uses three lidars, which cost $250 each (plus vision, radar, and ultrasound). It appears to work about as well as FSD. Here's the Out of Spec guys riding around on it in China for an hour:

https://www.youtube.com/watch?v=VuDSz06BT2g

Huawei received over $1 billion in grants from the Chinese government in 2023.

Western countries might not be smart enough to keep R&D because Wall Street sees it as a cost center.

Tesla spent about $10 billion on AI training hardware in 2024.[1] They just might not be getting as much bang for the buck, due to their paucity of sensors. Huawei get similar performance without Tesla's four billion miles of training data.

[1] https://www.tomshardware.com/tech-industry/artificial-intell...

You know what used to be expensive? Cameras. Then people started manufacturing them for mass market and cost when down.

You know what else used to be expensive? Structured light sensors. They cost $$$$ in 2009. Then Microsoft started manufacturing the Kinect for a mass market, and in 2010 price went down to $150.

You know what's happened to LIDAR in the past decade? You guessed it, costs have come massively down because car manufacturers started buying more, and costs will continue to come down as they reach mass market adoption.

The prohibitive cost for LIDAR coming down was always just a matter of time. A "visionary" like Musk should have been able to see that. Instead he thought he could outsmart everyone by using a technology that was not suited for the job, but he made the wrong bet.

but he made the wrong bet.

This should be expected when someone who is *not* an experienced engineer starts making engineering decisions.

It's not 2010 anymore. They will asymptotically reach approximately twice the price of a camera, since they need both a transmit and receive optical path. Right now the cheapest of the good LiDARs are around 3-4x that. So we're getting close, and we're already within the realm large-scale commercial viability.

That's ok, they're supposed to be. That's no excuse to rush a bad job.

The point of engineering is to make something that’s economically viable, not to slap together something that works. Making something that works is easy, making something that works and can be sold at scale is hard.

That's not engineering, that's industry. It's important to distinguish the two.

Engineering only exists within industry. Everything else is a hobby.

That's simply not true. Engineering can exist outside industry. "Stuff costs money" is not a governing aspect of these kinds of things.

FOSS is the obvious counterexample to your absurdly firm stance, but so are many artistic pursuits that use engineering techniques and principles, etc.

Industry includes FOSS and artistic endeavors, anything that’s done professionally.

My intent was to exclude research efforts, which is fundamentally different from engineering, which is a practical concern and not a “get it to just work” concern.

That's an interesting question, the question of whether engineering per se is strictly pragmatic. I personally think drawing a hard line between research and engineering is a misstep and relies too heavily on a bureaucratic kind of metaphysics.

If it would be easy there would already be a car costing a few million that few can afford but that has solved AD. But there isn't.

There is no market for such a thing. At that price point, you get a personal chauffeur. That’s what rich people do and he can do stuff that a self driving system never can.

And the rich people who don't want a chauffeur like driving the car. They will buy a $10M car no problem, but they want driving that car to be fun because that's what they were paying for. They don't want you to make the driving more automatic and less interesting.

Reply on news.ycombinator.com