There's multiple comments in this thread pointing to Chinese car manufacturers paying under $200 for their LIDAR hardware.

$200 is still a lot when a bunch of cameras cost maybe $20.

$200 to enable better FSD vs a decade of struggle to get FSD only partially working with $20 cameras. Which one do you think is more expensive overall?

The fact that we still do not have a significant number of cars with LIDAR on our streets somewhat proves which approach the auto industry considers viable for business.

I am much more curious about the next ten years. If we can bring down the cost of a LIDAR unit into parity with camera systems[1], I think I know the answer. But I thought that 10 years ago and it did not happen so I wonder what is the real roadblock to make LIDAR cheap.

[1] Which it won't replace, of course. What it will change is that it makes the LIDAR a regular component, not an exceptionally expensive component.

The fact that the only working self driving system uses LIDAR might say even more.

1) make it work

2) make it right

3) make it fast (or cheap in this case)

Elon thinks his genius intellect allows him to skip to #3.

> $200 is still a lot when a bunch of cameras cost maybe $20.

Anything except the lowest end car will cost $20K or more, so $200 is one percent of that price.

I mean, I'd rather be building a $30,200 robotaxi that works than a $30,020 robotaxi that doesn't.

You won't get rich with a $30,200 robotaxi, you won't even have a viable business. The game is the mass market and there the usual unit of currency is not cents, its tenth of cents.

Even if your robotaxi only manages 2000 rides that’s still down to just 10 cents a ride to cover the cost of hardware.

It’s nothing.

All taxis are variants of mass produced cars, that will not be different for robotaxis. The mass market is the enabler and there every tenth of cent counts.

Price without specs per radian is meaningless.

Investments into re-engineering production to bring down cost is done when there is a market large enough for said product.

True self-driving is still a baby that needs to grow and cannot even compete against an adult human with 30+ years of experience. As self driving actually forms to that level the market will grown.

Why do all of you think prices haven't come down? I can buy an AT128 from hesai for a few hundred dollars in volume. It's higher performance than any spinning lidar I could buy in 2017.

Elon's bet was one LIDAR against a bunch of cameras. A few hundred dollars is still way too much when you can get the cameras for a few tens.

In what universe is 'a few hundred dollars is way too much' for implementing full self-driving on an autonomous vehicle that moves like, and at the speeds of and in the spaces of, an automobile?

A two to four ton vehicle that can accelerate like a Ferrari and go over 100 mph, fully self-driving, and 'a few hundred dollars is way too much'.

Disagree. Even as they are dialing back the claims, which may or may not affect how people use the vehicles. These things respond too quickly for flaky senses based on human sensoriums.

You may have interpreted that and that is not what I said.

Once a product starts to sell after initial design, time is take to reduce the development cost. Try to reuse parts or replace part A with B. A machine from early 2018 can be little different than ones going out the door late 2018. _Kaizen_ was coined for this.

My point of view of when mass reduction in cost will be when self-driving is cost effect secondary feature on all Toyota vehicles. I see that as the litmus test for knowing that self-driving has reached true utility.

Also well designed vehicles would need a multi-sensor system to operate in self-driving mode. A human operating a car is using multi-sensor intake. Lack of multi-sensor in humans prevent them from operating a vehicle. Blind people need a secondary sensory input like walking stick. Vehicles need a multi-sensor system to prevent harming, mutilating, and killing the passengers and pedestrians.