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

Autonomy Investor Day - April 22 at 2pm ET

This site may earn commission on affiliate links.
Well, you paraphrased Elon's claim that if you had a great simulator, you'd already have solved the self-driving problem. I think that's complete nonsense.
Every single company active in this field uses both.
And yet, Tesla has not demonstrated that they can replicate what Waymo does (or Cruise, for that matter). Now we have Elon hyping up things like traffic light recognition as the next big step toward "feature completeness", even though others have been doing it for years. I think there are good reasons to be sceptical about their claims. Mind you, as a Model 3 owner I hope I'll be proven wrong.

Others are using more sensors to ramp up "progress" more quickly. But does it matter yet? Have they obtained L5 in useful conditions? Not really, though you could argue Waymo has in specific locations they are deployed.

No one argues on current state, but on predicting who will make it to a really deployable L4/L5 first. I am only arguing Tesla has a true benefit in their data access to achieve that goal. Not that it is sufficient on its own, or that others don't have a lead.
 
  • Like
  • Helpful
Reactions: Nauti-Boy and CarlK
Others are using more sensors to ramp up "progress" more quickly. But does it matter yet? Have they obtained L5 in useful conditions?
Nobody has achieved level 5, but Waymo has had level 4 since at least 2017 (restricted geographic areas, as you mentioned). And it is still unclear if Tesla's low-cost sensor suite will ever be sufficient to get there.
 
  • Like
Reactions: rnortman
when i said that autonomy day was done solely to raise the stock price and to directly increase their valuation from investors i was ridiculed. Yet here we are a week later.

With cash dwindling, Tesla seeks to raise $2 billion in debt and equity

5YcBC47.png


I'm tired of being right all the time.
 
when i said that autonomy day was done solely to raise the stock price and to directly increase their valuation from investors i was ridiculed.

But what you're missing is that increasing the valuation isn't the end goal! Elon is a selfless savior of mankind! He's just trying to slow global warming enough to give him time to move us all to Mars. Or at least those of us who can afford the ticket! And he needs to increase the valuation so he can raise more money so he can replace more combustion engines so he can build more rocket engines so he can move people anywhere on the planet without needing cars at all so he can afford to build rockets to move people to Mars.

It's all so brilliant.
 
You clearly don't understand how beneficial a good simulator is. You can take one instance you see in the Real world and create thousands of variations of it in a simulator. The problem is NOT simulating individual tail cases, the problem is simulating the way humans drive as a whole.

For example some of the pros of simulation:
  • It's cheap -- no need to even have a car, let alone a road or safety drivers.
  • It's fast -- you can drive a million miles in simulator in a night.
  • It's safe -- you can test dangerous situations you could never try in the real world, like pedestrians running in front of you or being in the middle of accidents
  • It's complex -- you can drive all your simulated miles in interesting and challenging conditions, not wasting time on simple and boring ordinary roads.
  • It's repeatable -- you can do the same test again and again.
  • It's variable -- you can try millions of tiny variations of any situation. You can try every type of weather and lighting. You can try every type of road and traffic.
  • It's strange -- you can test extremely rare situations, like what your car might do in a tsunami, earthquake, fire, flood, tornado or dust storm.
  • It's software -- you can create complex failures of your simulated hardware, mixed in with every situation, and know how you deal with them.
Just an FYI. Tesla is running simulations as well. I think they downplay that channel of work. Karpathy wrote about the new chip reducing simulation time from , I believe about 3 minutes to seconds with the new chip vs the Nvidia chip they are replacing.
 
Just an FYI. Tesla is running simulations as well. I think they downplay that channel of work. Karpathy wrote about the new chip reducing simulation time from , I believe about 3 minutes to seconds with the new chip vs the Nvidia chip they are replacing.

Do you have link of that? Elon clearly said in the investor event that Tesla does not do simulation. He said if you can set up the correct simulation, meaning it covers all tail cases, you have already solved the self driving problem. That makes tons of sense to me.
 
Do you have link of that? Elon clearly said in the investor event that Tesla does not do simulation. He said if you can set up the correct simulation, meaning it covers all tail cases, you have already solved the self driving problem. That makes tons of sense to me.
The presentation showed a simulation from Tesla (has dual dump truck in right lane). Additionally, Tesla had job openings for Unreal programmers.
 
Do you have link of that? Elon clearly said in the investor event that Tesla does not do simulation. He said if you can set up the correct simulation, meaning it covers all tail cases, you have already solved the self driving problem. That makes tons of sense to me.
Elon said simulation is not a substitute for real world data. They do simulation to test.
 
The presentation showed a simulation from Tesla (has dual dump truck in right lane). Additionally, Tesla had job openings for Unreal programmers.

Elon said simulation is not a substitute for real world data. They do simulation to test.

Thank you guys for the clarification. Tesla still only uses real world data to train neural net which is my understanding from what Elon said. Missed the part that Tesla uses simulation to do test.