Tesla’s FSD Beta 10.3 Coming This Friday 10-22-2021

[Knightshade]

I predict that when the IIHS publishes fatality data it will show that Teslas are not the safest modern cars

I'd be willing to bet money that's not accurate.

NHTSA testing showed the 3 cars with lowest probability of injury ever tested were the Model 3, Model S, and Model X a couple years ago. I'd be shocked if the Y isn't #4 in the top 4 at this point.

Tesla advertised this fact (to the degree Tesla "advertises" anything) - and NHTSA got mad because they don't officially rank cars that way (5 star is the top rating, so all 5 starts are "equally" good to them) but if you actually check the raw data it confirms Teslas claims

 

[Bladerskb]

The reason you believe Tesla will be able to upgrade existing FSD vehicles is based on your own assumption of what the upgrade will consist of. You believe that it will only require upgrades to things intended to be upgraded like the computer or the sensor heads.

@Knightshade is a lost cause. Here you have a guy who NOW believe that Tesla were misleading when they said their 10 TOPs HW2 was enough and then their HW 2.5 and then their 144 TOPS HW3. But now believes you need HW4 although in 2019 Musk was saying they could still do it on the 10 TOPs HW2.

Of-course back then we who were saying this simple reason and logic were called fud and shorts.
While the fans who said Nvidia were lying and just trying to market their chips when they said L4 or L5 required more TOPs were right.

Now all of a sudden you are saying what we were saying?

Yet this same guy completely believes that a 8 blurry low resolution 1.2 megapixels cameras, with short range, compromised views, easily crippled during mild rain/snow (Rear view cam), multiple blind-spots and no air/water cleaning solution is the way to go and is all you need to achieve L5 better than humans because Elon told him so. The same Elon who said that 10 TOPs was all you needed. This is without even getting to occlusion during inclement weather (fog, mist, dust, etc).

 

[Bladerskb]

I'd be willing to bet money that's not accurate.

NHTSA testing showed the 3 cars with lowest probability of injury ever tested were the Model 3, Model S, and Model X a couple years ago. I'd be shocked if the Y isn't #4 in the top 4 at this point.

Tesla advertised this fact (to the degree Tesla "advertises" anything) - and NHTSA got mad because they don't officially rank cars that way (5 star is the top rating, so all 5 starts are "equally" good to them) but if you actually check the raw data it confirms Teslas claims

Lol those tests are nonsense. As per for course with Tesla fans. When the facts are not lining up to their narrative they dismiss it.
Yet will take a statistics that consists of cars that are average 15-17 years old versus Tesla which are 3 years old and claim Tesla are 10x safer than any car on the road.

I guess First grade logic is hard.

 

[Knightshade]

@Knightshade is a lost cause. Here you have a guy who NOW believe that Tesla were misleading when they said their 10 TOPs HW2 was enough and then their HW 2.5 and then their 144 TOPS HW3. But now believes you need HW4 although in 2019 Musk was saying they could still do it on the 10 TOPs HW2.

.... what?

I've said from day 1 nobody knows how much compute you need for FSD because nobody has done it yet.

As long as Tesla keeps upgrading the computer for free as needed- I don't care either.

Why do you?

And why do you keep making up things I never argued?

Of-course back then we who were saying this simple reason and logic were called fud and shorts.
While the fans who said Nvidia were lying and just trying to market their chips when they said L4 or L5 required more TOPs were right.

Now all of a sudden you are saying what we were saying?

No, you're making up strawmen I was never part of.

Again.

Yet this same guy completely believes that a 8 blurry low resolution 1.2 megapixels cameras, with short range, compromised views, easily crippled during mild rain/snow (Rear view cam), multiple blind-spots and no air/water cleaning solution is the way to go and is all you need to achieve L5 better than humans because Elon told him so. The same Elon who said that 10 TOPs was all you needed. This is without even getting to occlusion during inclement weather (fog, mist, dust, etc).

Again, making up nonsense I never claimed.

On the contrary I've remarked on the weather concerns (as evidenced by even NoA reducing features during moderate rain) for years.


Do have anything actually useful to add based on peoples actual words instead of nonsense you put in their mouths?

 

[EVNow]

When the facts are not lining up to their narrative they dismiss it.

Isn't that what you do ?

BTW, Troll - why are you here ?

When are you going to mention your conflict of interests in the signature or do you not have even an ounce of integrity ?

 

[Bladerskb]

.... what?

I've said from day 1 nobody knows how much compute you need for FSD because nobody has done it yet.

Exactly. That's complete nonsense and its something someone who is completely clueless about AV tech would say. It was wrong in 2016 and also wrong in 2021. Its a convenient way of lying. Some people outright lied and said Nvidia was marketing and others like you just confessed asserted their ignorance of technology and AV tech. This is my point you were apart of the clueless people and are now asserting that its not enough.

As long as Tesla keeps upgrading the computer for free as needed- I don't care either.

Why do you?

And why do you keep making up things I never argued?

I didn't make up anything, you literally just admitted you asserted something equivalent to flat earth theory in AV tech.

No, you're making up strawmen I was never part of.

Again.

Never said you were part of the people who said Nvidia were lying. I just said there were people who did.

Again, making up nonsense I never claimed.

On the contrary I've remarked on the weather concerns (as evidenced by even NoA reducing features during moderate rain) for years.


Do have anything actually useful to add based on peoples actual words instead of nonsense you put in their mouths?

You completely believe that a compute upgrade is all it takes. This is literally the issue @S4WRXTTCS had with you.
Talking to you is hopeless.

 

[Knightshade]

Exactly. That's complete nonsense and its something someone who is completely clueless about AV tech would say.

Dude.

Are you drunk?

As I've said for a long time-nobody knows how much compute is needed.

Your claim I am "only now" admitting HW2 was insufficient is not a factually true claim

I never claimed it was.

Because again-nobody knows how much compute is needed.

Including the experts.

And especially including you.

It was wrong in 2016 and also wrong in 2021. Its a convenient way of lying.

So far the only people caught lying here are....you.

Having claimed a ton of stuff going back to 2016 (when I wasn't even a member here) and then some more stuff more recently I also never said.

I didn't make up anything, you literally just admitted you asserted something equivalent to flat earth theory in AV tech.[/QUOTE[

Yes, you made nearly everything in your previous post up.

Lol those tests are nonsense.

How, specifically, are NHTSA crash safety tests "nonsense"?

I guess First grade logic is hard.

It sure seems to be for you.


I never said any of the garbage you claimed.

I wasn't even here when you claimed much of it was said.

Never said you were part of the people who said Nvidia were lying. I just said there were people who did.

You said this about me- which is a lie


"this same guy completely believes that a 8 blurry low resolution 1.2 megapixels cameras, with short range, compromised views, easily crippled during mild rain/snow (Rear view cam), multiple blind-spots and no air/water cleaning solution is the way to go and is all you need to achieve L5 better than humans "

You completely believe that a compute upgrade is all it takes.

This is another example of you lying.

I've posted multiple times I think a more forward-mounted side-facing cam is needed to avoid the "creeping" issue for example.


Do you ever not lie and make up crap people didn't say?

 

[cadreeeem]

Safety Score Beta | Tesla Support

The Safety Score Beta is an assessment of your driving behavior based on several metrics called Safety Factors. These are combined to estimate the likelihood that your driving could result in a future collision. We combine your daily Safety Scores (up to 30 days) to calculate the aggregated...

www.tesla.com

If you follow 3-4 seconds behind the car in front of you, it won't do anything to your safety score regarding safe following distance. You need to counteract the negative score by following two seconds behind the car. You have to be greater than 1 second, and less than 3 second at 50+ miles per hour.

Not true. Trying to stay 1-3 seconds distance at 50mph+ will truly get you dinged for close following! I improved my score from 96 to 100 so I speak with experience.

 

[DreamBurn]

Not true. Trying to stay 1-3 seconds distance at 50mph+ will truly get you dinged for close following! I improved my score from 96 to 100 so I speak with experience.

Thank you Mr. Experience; however, your experience is inaccurate.

Safety Score Beta | Tesla Support

The Safety Score Beta is an assessment of your driving behavior based on several metrics called Safety Factors. These are combined to estimate the likelihood that your driving could result in a future collision. We combine your daily Safety Scores (up to 30 days) to calculate the aggregated...

www.tesla.com

And I quote: "Unsafe following is the proportion of time where your vehicle’s headway is less than 1.0 seconds relative to the time that your vehicle’s headway is less than 3.0 seconds. Unsafe following is only measured when your vehicle is traveling at least 50 mph and is incorporated into the Safety Score formula as a percentage."

Following less than 1 second behind at 50+ is bad. Following greater than 3 seconds behind at 50+ does not improve (lower) the percentage of time you are "unsafe following" it will only increase the number of miles you have travelled. So, following at 3+ seconds will improve your safety score by increasing the number of miles to apply your good daily score, but do NOTHING to lower the "Unsafe Following" percentage metric if you happened to get "dinged" a little bit during your drive.

If you get home after a close merge on the freeway, or being cut off on the highway, and realize you earned a small percentage of "Unsafe Following"—all you have to do is take the car out and follow somebody at 2 seconds (1 to 3 seconds is the window) at 50+ mph for a little while in order to reduce the percentage on the "unsafe following" metric to zero. If you got dinged and earned a percentage of "Unsafe Following" and you follow somebody with 4 seconds of headway above 50mph, it will do absolutely nothing to that percentage. Read their documentation.
Congratulations on your score.

 

[cadreeeem]

Thank you Mr. Experience; however, your experience is inaccurate.

Safety Score Beta | Tesla Support

The Safety Score Beta is an assessment of your driving behavior based on several metrics called Safety Factors. These are combined to estimate the likelihood that your driving could result in a future collision. We combine your daily Safety Scores (up to 30 days) to calculate the aggregated...

www.tesla.com

And I quote: "Unsafe following is the proportion of time where your vehicle’s headway is less than 1.0 seconds relative to the time that your vehicle’s headway is less than 3.0 seconds. Unsafe following is only measured when your vehicle is traveling at least 50 mph and is incorporated into the Safety Score formula as a percentage."

Following less than 1 second behind at 50+ is bad. Following greater than 3 seconds behind at 50+ does not improve (lower) the percentage of time you are "unsafe following" it will only increase the number of miles you have travelled. So, following at 3+ seconds will improve your safety score by increasing the number of miles to apply your good daily score, but do NOTHING to lower the "Unsafe Following" percentage metric if you happened to get "dinged" a little bit during your drive.

If you get home after a close merge on the freeway, or being cut off on the highway, and realize you earned a small percentage of "Unsafe Following"—all you have to do is take the car out and follow somebody at 2 seconds (1 to 3 seconds is the window) at 50+ mph for a little while in order to reduce the percentage on the "unsafe following" metric to zero. If you got dinged and earned a percentage of "Unsafe Following" and you follow somebody with 4 seconds of headway above 50mph, it will do absolutely nothing to that percentage. Read their documentation.
Congratulations on your score.

Good luck