-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable
- Thread starter AudubonB
- Start date
ZachF
Active Member
They will probably add another $2k for using a dealership.
StealthP3D
Well-Known Member
Don't be silly.
FSD will only follow the directions of the traffic cop if it is safe to do so. And, yes, anyone dressed like a cop can direct traffic.
FSD will not ask the cop for proof of identity. Is that what you do before you follow the policeman's hand signals? I just check other traffic to see if it's safe to do so.
I seriously wonder about people who raise such transparent objections to FSD. It's like they are rooting against it's success so fervently that they can't even see how silly their objections are.
Not sure if this has been shared yet.
www.teslarati.com
As an aside:
Came back from a road trip vacation, feels good to have some form of life coming back post-COVID. Took my first long road trip in the M3P and I never want to go back to a non-electric vehicle nor a non-Tesla vehicle. Such an easy and **fun** drive for a road trip. I feel more energized after driving than exhausted somehow!
Tesla Model Y body castings begin to stack up at Giga Texas
Tesla Model Y castings are beginning to stack up in impressive numbers at the automaker’s Giga Texas plant just outside of Austin. A Snapchat video put up by someone on-site shows perhaps thousands of Model Y castings on the property as Tesla begins to work toward the initial production of the...
www.teslarati.com
As an aside:
Came back from a road trip vacation, feels good to have some form of life coming back post-COVID. Took my first long road trip in the M3P and I never want to go back to a non-electric vehicle nor a non-Tesla vehicle. Such an easy and **fun** drive for a road trip. I feel more energized after driving than exhausted somehow!
AimStellar
Member
To clarify Kim did state recalibration helped which can be seen in her second video. She has also driven with a FSD beta driver while out in SF (Whole Mars Catalog I believe) and reported there was a major improvement in how well FSD handled SF roads compared to the cow pasture roadway found in Newport RI. This is what prompted Elon's comment that FSD training currently favors the SF area. I find this encouraging as the NN will help address many of the differences in how well FSD drives in different locations. Would also be nice to see people give Kim a break especially after she went thru a lengthy explanation of why so many disengagements.
Last edited:
Mike Smith
Active Member
It looks like the Big 3's plan is to make almost exclusively hybrids with the minimum sized battery required. The 7kwh minimum rises to 10 kwh for 2023 to 2031, but with declining costs that battery will still cost well under $1000. Customers will get a $9000 rebate, so if they sell 6 million of these hybrids per year it will cost the government over half a trillion dollars in subsidies.
6m X $9k X 10yrs = $540B
Diabolical but clever - it solves their cell supply problem and gives them (in their mind) a possibility of competing against Tesla.
Oh I do hope they make this data from the dirty dozen public before Christmas! This is good I think for playing on a level field, but watch for the cheaters.
Knightshade
Well-Known Member
Given the inferior, often vastly inferior, telematics in legacy auto cars- I wonder how much of this info they actually have.
I am doubtful data from the dirty dozen will help Tesla's case. Those other "driver assist" systems are so much weaker, that owners simply do not use them (they are practically useless as they can't even keep the car in the lane through a mild curve -- see Sandy Munro video for example). With negligible low amount of use, they probably do not have any crashes (due to low statistical probability of occurrence), therefore it will not make Tesla look good in comparison.
Good point. I know our Telluride has a similar "autopilot" type feature that requires hands on the wheel like Tesla every minute or so and I can tell you that if I just let it go that it would probably crash just on a simple bend in the highway. Probably would hit a parked car that was partly in the lane as well but I haven't tried that yet to see what would happen, lol. I like the feature but definitely still have to keep my eyes on the road.
Todd Burch
14-Year Member
She mentioned about 8 minutes into the video that "so far this drive has been a lot better than yesterday's"...but at that point she had mostly been in traffic and hadn't encountered any challenges for FSD. After that point, she had some big issues come up. Can't really go by what she said before the drive got interesting.
She titled the video "FSD Beta V10 - UNEDITED drive two on test loop, after recalibration and reset. Not much better for me." In fact, not only was there no noticeable improvement, there were more "dangerous" situations (e.g. car trying to pull out directly into moving traffic) than in the first video. Any difference can really just be attributed to minor environmental differences in the two drives.
It's clear this wasn't a calibration issue--it's flat out just not performing very well for her. Some things are understandable (the streets there are narrow, wonky, and there's poor visibility). Other things are head scratchers, like why (despite good visibility at an intersection) her car tried to pull out directly in front of moving cross traffic.
Regarding her statement about Omar, she said that she has driven with Whole Mars in the past in San Francisco (before FSD 10) and was talking about how FSD performance in San Francisco is much better than in her town of Newport, RI. She was referring to Elon's Tweet that FSD is somewhat "overfit" for San Francisco and performs better there.
Todd Burch
14-Year Member
Hopefully the data that's provided is quantified along with the number of miles driven with each of the systems engaged. But when you think about it, it's likely that none of the other automakers even HAVE this telemetry, so you're probably right. I mean really...Toyota, how many miles were driven with your TACC engaged? How would they know?
Rob just went off on Twitter.
ZachF
Active Member
Boston in the winter will definitely be Nightmare/hardcore mode for FSD.
It has to be presented statistically, but that's a give-away so as you say we'll be looking at total numbers of disengagements in a given period. However, collisions with emergency vehicles might show some truth as there's only 11 there for Tesla.
insaneoctane
Well-Known Member
Exactly my thoughts....no way they have this data....maybe on some punch cards in a closet?
insaneoctane
Well-Known Member
I have to disagree....my belief is for any examples of collision that they might come up with, how will they be able to show they did more than Tesla to try and avoid?
This is both a sensible step and probably positive. It introduces comparison with the performance of other ADAS systems. It seems likely that this statistical information will not be available from some (if not most) of the automakers. The elements regarding driver engagement and detection of first responder etc also allows comparison.
It would also seem to reduce the likelihood of any recommendations which result being focussed purely at Tesla.
navguy12
Active Member
Based on a debrief from my mobile repair chap last year:
Using the recalibration button wipes out all past alignment activities and begins a fresh “coarse alignment” that requires manual driving inputs until the parameters of the “coarse“ alignment allow TACC/Autosteer to work. Then, over the next several 100 kms of freeway driving, the cameras continue to calibrate.
At the time I had this conversation, my mobile repair chap gave me the percentage figures of what alignment state each camera was in.
IIRC, one of the cameras was 75ish% and the rest were between 91 and 96…so the alignment requires more than just getting the car to allow one to use TACC/Autostee.
Similar threads
- Locked
- Locked
- Locked
