Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Elon: FSD Beta tweets

This site may earn commission on affiliate links.
You are misunderstanding HD maps.
That’s just regular HD Map which acts like a guide.
Oh c'mmon, I know how to read.

The article I read specifically talked about some special things they do for parking lots. Including alternate paths they draw in case of blocks. When I have time I'll find the article.

Do they draw those alternate paths for all the roads, too ?

If so, Waymo is worse than I had thought.
 
Last edited:
You are misunderstanding HD maps. HD maps are annotated with information like one way lanes in parking lots so that the car won't accidentally drive the wrong way in a parking lot. That is not the same as actually hardcoding the entire route to follow. The car is free to pick whatever routes it wants, it just has map info to help it obey parking lot rules like direction of travel down lanes. And the car has an on board Router that determines the best route and a Planner that can change the route on the go as needed if the route needs to change. So the entire route is not hardcoded.

Understood. I thought “hd maps” also referred to having high precision location data of stuff. Lanes, curbs, maybe other obstacles. That seems like it would be a challenge to compute from vision only - especially in bad weather, or when viewing is obstructed. Anyway just curious.
 
Understood. I thought “hd maps” also referred to having high precision location data of stuff. Lanes, curbs, maybe other obstacles. That seems like it would be a challenge to compute from vision only - especially in bad weather, or when viewing is obstructed. Anyway just curious.
HD maps can certainly include those things. All you need to know for this forum is that Tesla does not and will never use HD maps. As Tesla adds more detail to their maps the definition of HD map will change to reflect that. :p
 
Understood. I thought “hd maps” also referred to having high precision location data of stuff. Lanes, curbs, maybe other obstacles. That seems like it would be a challenge to compute from vision only - especially in bad weather, or when viewing is obstructed. Anyway just curious.

HD maps can include a lot. Each company can include different things. I don't think all HD maps are the same.

Yes, the idea of HD maps is to include info that might be difficult or unreliable with vision-only so that the AV can drive with more confidence.
 
The only way to keep the Tesla hype high these days is bashing competitors

Hah, so true. The Waymo whataboutism on this forum is strong, which is kinda funny because no matter what Waymo does, it has zero measurable impact on what Tesla is doing. Just because one company failed doesn't mean that the other competitor is a winner.
 
Hah, so true. The Waymo whataboutism on this forum is strong, which is kinda funny because no matter what Waymo does, it has zero measurable impact on what Tesla is doing. Just because one company failed doesn't mean that the other competitor is a winner.
Yes. I think that what waymo is doing is irrelevant to anyone on these forums. No one here wants to ride in a taxi. They want to ride in their own car. What waymo is doing solves none of our problems and therefore it is mainly irrelevant.
 
Care sharing what you disagree with @diplomat33

Sure. I disagree that Waymo is irrelevant. A company that is a leader in autonomous driving, contributing innovations in machine learning and deploying robotaxis, is not irrelevant. They are certainly relevant in the AV field. I think it is rather self-centered and closed-minded to suggest that Waymo is irrelevant just because they don't offer something YOU want. They might not offer something that YOU want, but they offer something that OTHERS want. There are lots of people who do care about riding in a robotaxi. And the wants and needs of others matter too. And the impact Waymo is having on the AV industry matters. So Waymo is very relevant even if it is not relevant to YOU.
 
Yes. I think that what waymo is doing is irrelevant to anyone on these forums. No one here wants to ride in a taxi. They want to ride in their own car. What waymo is doing solves none of our problems and therefore it is mainly irrelevant.
Waymo is irrelevant in the same way that the original Tesla Roadster was irrelevant to those who wanted a cheaper four door EV that could be used as a primary vehicle.

I'm not betting that Waymo specifically will survive and sell tech for consumer vehicles but it seems likely that the technology used in robotaxis will eventually filter down to consumer vehicles. Obviously if you believe that Tesla has already solved the problem then everyone else is irrelevant given Tesla's massive cost advantage.
 
Understood. I thought “hd maps” also referred to having high precision location data of stuff. Lanes, curbs, maybe other obstacles. That seems like it would be a challenge to compute from vision only - especially in bad weather, or when viewing is obstructed. Anyway just curious.

HD Maps also require high-precision localization of the car itself to be useful, which is what the LIDAR is primarily used for on Waymos and others. LIDAR and HD Maps go hand in hand together. Ultimately I expect they'll get rid of both in the drive to become profitable, but for now the whole tech stack is built around it.
 
Sure. I disagree that Waymo is irrelevant. A company that is a leader in autonomous driving, contributing innovations in machine learning and deploying robotaxis, is not irrelevant. They are certainly relevant in the AV field. I think it is rather self-centered and closed-minded to suggest that Waymo is irrelevant just because they don't offer something YOU want. They might not offer something that YOU want, but they offer something that OTHERS want. There are lots of people who do care about riding in a robotaxi. And the wants and needs of others matter too. And the impact Waymo is having on the AV industry matters. So Waymo is very relevant even if it is not relevant to YOU.

Waymo today has the most solid and complete tech stack out of everyone, but I could imagine them becoming irrelevant just because the business is building the wrong product. It's like making a version of Uber except without the benefit of network effects and the addition of massive capital costs of buying and maintaining your own customized vehicles. And all they get to offer is a rideshare service that could eventually become less costly than human-powered alternatives (but also probably more cautious and slower). So people can choose to save a dollar and take a slow Waymo, or pay a bit more and take an Uber whose driver will roll stop signs and be more aggressive, and never need a remote operator to phone in unexpectedly. It's like UberX vs UberPool, pay a bit less for a slower trip.
 
  • Helpful
Reactions: diplomat33
Sure. I disagree that Waymo is irrelevant. A company that is a leader in autonomous driving, contributing innovations in machine learning and deploying robotaxis, is not irrelevant.

but, they will never ever develop fsd for a car you can buy. Only for taxies. No one here wants to ride in taxis. They want to ride in their own car, hence they are irrelevant.


They are certainly relevant in the AV field. I think it is rather self-centered and closed-minded to suggest that Waymo is irrelevant just because they don't offer something YOU want.

or anyone who owns a Tesla wants.

. And the impact Waymo is having on the AV industry matters.

What impact?
 
Waymo today has the most solid and complete tech stack out of everyone, but I could imagine them becoming irrelevant just because the business is building the wrong product. It's like making a version of Uber except without the benefit of network effects and the addition of massive capital costs of buying and maintaining your own customized vehicles. And all they get to offer is a rideshare service that could eventually become less costly than human-powered alternatives (but also probably more cautious and slower). So people can choose to save a dollar and take a slow Waymo, or pay a bit more and take an Uber whose driver will roll stop signs and be more aggressive, and never need a remote operator to phone in unexpectedly. It's like UberX vs UberPool, pay a bit less for a slower trip.

Fair enough. Certainly, Waymo could become irrelevant if they are not able to compete with human driven ride-hailing. I'm just arguing that they are not irrelevant now.

but, they will never ever develop fsd for a car you can buy.

You can't say that. We don't know that Waymo will never do a consumer car. Obviously, right now, the FSD is too expensive and not solved yet so it can't be put on a consumer car yet. But that is likely to change in the future. It is likely that the cost will eventually be cheap enough and once Waymo solves FSD, there would be nothing stopping them from putting it on a car you can buy.

Only for taxies. No one here wants to ride in taxis. They want to ride in their own car, hence they are irrelevant.

You are being rather presumptuous. You don't speak for everybody. Some people on this forum probably would ride in a robotaxi. I know I would.

or anyone who owns a Tesla wants.

Not everybody owns a Tesla. This is what I mean by being self-centered and closed-minded. Tesla owners are not the only ones that matter.

And again, irrelevant to YOU maybe but you are NOT the only person that matters. Waymo is certainly not irrelevant to the people in Chandler and SF that ride in a Waymo robotaxi every day. In fact, there are people with mobility issues that are very thankful that Waymo exists. Waymo is not irrelevant to them.

What impact?

A huge impact. Waymo/Google has pioneered a lot of the machine learning that others, including Tesla, are now using in their FSD. We would not be where we are today with FSD, if it wasn't for the Google self-driving project which later became Waymo.
 
Last edited: