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

Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

This site may earn commission on affiliate links.
My thinking is that human driver inputs accelerate neural net development. No correction, no learning. This is not the entire game but a part of it as I view it.
I believe you’re right. Didn’t Tesla say a long time ago that it’s what didn’t work that they want to see not what did work. It was the disengagements that got sent back to the Mother Ship.
 
you would think cutting off the oxygen would be a better approach to electrical fires.. i propose a big fireproof blanket
I remember someone bringing this up before but I cant find it directly in this thread, so here is an external source:
Q: Can lithium-ion batteries burn without oxygen?
A: The lithium-ion battery or Li-ion battery fire does not need any oxygen to burn at all, because it can do that without any oxygen at all. The nature of the fire is recognized as completely chemical.

But I would stick to what @Artful Dodger linked above and do what Tesla recommends. Lots of water 🌊🌊
 
I went to Lhasa in the remotes of Tibet and saw a Tesla EV charger .... bummer ;) /jk

wood cutters in remote areas of the world use chain saws (& not axes) (saw it myself) ... so gas must be getting hauled via mules ...

World Is Flat ...
Indeed Lhasa does have that St Regis too and was the last major stop on the route to Mount Everest Base camp by those intrepid Chinese adventurers.
I still cannot really believe that even after being there a decade ago. The world is shrinking!
 
I believe you’re right. Didn’t Tesla say a long time ago that it’s what didn’t work that they want to see not what did work. It was the disengagements that got sent back to the Mother Ship.
But if you intervene, how do you know FSD didn't work? That's my point.
 
I remember someone bringing this up before but I cant find it directly in this thread, so here is an external source:
Q: Can lithium-ion batteries burn without oxygen?
A: The lithium-ion battery or Li-ion battery fire does not need any oxygen to burn at all, because it can do that without any oxygen at all. The nature of the fire is recognized as completely chemical.

But I would stick to what @Artful Dodger linked above and do what Tesla recommends. Lots of water 🌊🌊
got it... i missed those discussions ... is there a chemistry based suppressant ? foam or powder that could be developed ...
 
Do battery fires need oxygen?

The lithium-ion battery or Li-ion battery fire does not need any oxygen to burn at all, because it can do that without any oxygen at all. The nature of the fire is recognized as completely chemical, so there is no role of oxygen in it that can start this fire.

My understanding is that the electrolyte solvents used in a Li-Ion cell release both oxygen and hydrogen gas during a thermal runaway event. More here:

[PDF] How Electrolytes Influence Battery Safety www.electrochem.org › interface › sum › sum12 › sum12_p045_049
 
It's a quarterly "Triple Witching" options expiration day. Big options writers with the ability to temporarily manipulate share prices (hedge funds and market makers) may want to keep today's TSLA closing price near strike $300 where options' open interest and volume are heavy. When the share price shot above a few minutes ago, they appear to have taken action to cap it. The rest of this final hour could be interesting.
 
Last edited:
But if you intervene, how do you know FSD didn't work? That's my point.
Often interventions are after FSD has already screwed up. Driving in a bike lane, making an illegal turn, stopping in a dangerous place, etc etc.

Just because FSD hasn’t caused an accident doesn’t mean it is doing the right thing. Considering there have been no significant accidents while FSD is in charge suggests they have the biggest problems solved. You know FSD didn’t work if its stopped in a traffic lane where it doesn’t belong. You know when it turns into a lane opposite traffic, etc etc.
 
But if you intervene, how do you know FSD didn't work? That's my point.
One possible way to look at is that with no correction, FSD behavior is reinforced as ideal when it may not be ideal. I am sure there is more to it but this is part of it.

There must be degrees of correction that appear as thousands of corrections are blended somehow. Lots to it for sure.
 
But how would they know if you don't intervene and FSD works, albeit in a manner you wouldn't have used?

Neural nets work on statistics. Knowbody nose anything.

You are the Driver. If FSD doesn't drive the way you would, take over.

What would you do if a 14-yr old was driving stupidly in your new, shiny, expensive car? That's right: you'd take over.

Let Tesla and the Autopilot team worry about "knowing", while you take care of your property.
 
So if there is a fire in the US in other electrical equipment - like a substation for instance - do the Fire Departments use water only to extinguish those fires too?
Electrical equipment fires and battery fires are two different animals and require different strategies.
 
Often interventions are after FSD has already screwed up. Driving in a bike lane, making an illegal turn, stopping in a dangerous place, etc etc.

Just because FSD hasn’t caused an accident doesn’t mean it is doing the right thing. Considering there have been no significant accidents while FSD is in charge suggests they have the biggest problems solved. You know FSD didn’t work if its stopped in a traffic lane where it doesn’t belong. You know when it turns into a lane opposite traffic, etc etc.
True dat. I'm talking about instances where it doesn't stop or accelerate as quickly as you might. Intervening before it "might" screw up.

As I haven't been in the driver's seat, I have no experience, but in most of the videos I've seen recently, the driver frequently intervenes prior to one truly being necessary.

OK, I'll shut up now and watch the SP rise...
 
Tesla has always recommended copious amounts of water.

Perhaps an EV specific solution can be developed and become widespread.

EDIT:

Do battery fires need oxygen?


The lithium-ion battery or Li-ion battery fire does not need any oxygen to burn at all, because it can do that without any oxygen at all. The nature of the fire is recognized as completely chemical, so there is no role of oxygen in it that can start this fire.
So it might be best to have a box, load the car in it, and pour in liquid oxygen.
 
  • Funny
Reactions: wipster
I really wish Tesla would allow massive amount of crowd sourcing to help them with FSD beta. Have the fleet fix the map data for them by having us annotating miss stop signs or wrong speed limit (and have Tesla owners put in the speed limit that's right). Give us a few clickable feed back button every time we click on the camera button such as "incorrect lane switching", "missed turn", "median not wide enough", "phantom braking"...and whatever they choose. I feel like it'll save them so much time vs trying to figure out what the problem was using a small clip.

Even if they don't want us messing around with map data or report data, just having the data helps tremendously. Instead of having "disengagements/mile " metrics, now Tesla can see "oh now there's more phantom braking with this update, or now there's more missed turns...These type of macro data is extremely helpful to gauge what they should be working on.

Leverage the power of what made programs like Waze and Wikipedia such a powerhouse.