AlanSubie4Life
Efficiency Obsessed Member
Oh no! I meant to mention to set FCW to Early in the car, too. Oh well.only to find out that I had another phantom FCW and unsafe following during this drive
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Safety Score
- Thread starter ScottNY
- Start date
pitabun
Member
I did the impossible, drove from Irvine to LAX & back to pick up my wife while maintaining a 100 for the day. Unfortunately it didn't move my score, still sitting at 98.
HighZ
Member
If you do multiple drives in one day and get the ding on one drive, is it possible that the daily score is averaged using the individual drives and events like the total score, so the daily total of events doesn't have to match the daily score? That is the case on the total score.
Silicon Desert
Well-Known Member
jebinc
Well-Known Member
@n.one.one Thanks for this! I loaded your spreadsheet with my data and compared the result to my simplistic one and got the same result. No need to input all the event data. Attached is a view of my uncomplicated spreadsheet, both with and without formula view.Here is mine for your review. It's gone through many revisions and has been helping me for several days.
- This is set to read only so you'll need to copy the range of the active items and paste it in your own sheet
- There are hidden columns with some of the working details so be careful what you change or move
Tesla Safety Score - Tracking & Planning v10
Safety Score This is set to read only. To make your own sheet go to the file menu, 'Make a Copy' and open in Google Sheets (not in Excel). No share requests please. There are hidden columns with some of the working formulas so be careful what you change or move. <a href="https://teslamotorsclu...docs.google.com
Last edited:
Silicon Desert
Well-Known Member
Right, yet don't assume you will be in the first group to get the beta for driving only 70 miles and maybe you know the reason which Telsa does not not make public.
EVNow
Well-Known Member
We don't know. To play it safe I'm driving about 10 miles a day. That is more than I normally drive now-a-days.
ps : It really depends on Tesla's objective here. Is it to get good testing data or is it to expand beta to pacify irate customers .... I can understand if they have a low mileage cut off (10 miles / 50 miles / 100 miles ?) saying there simply isn't enough data. But if it was up to me, I'd avoid giving preference for city FSD beta based on highway miles driven - so won't give preference to higher mileage.
In a tweet in March, Elon said they want to greatly expand beta (it was 2k then) first by 2x (which I think they did) and then by 10x. My guess is they will include some 20k in this rollout. No idea what they will do if a lot more than 20k people have decent safety scores.
Last edited:
Everything is guess work right now. To your point, I wouldn't assume that 100+ miles a day on the highway with a 100% Safety Score would get someone into the beta either. Since the beta is for FSD on City Streets.Right, yet don't assume you will be in the first group to get the beta for driving only 70 miles and maybe you know the reason which Telsa does not not make public.
EVNow
Well-Known Member
Speculation stated as fact. If Tesla hasn't made it public - nobody but some Tesla insiders would know. You surely wouldn't.
are there a lot of irate customers about the safety score?
i’m mixed. i’ll preface this by saying i don’t really use FSD, most of my miles are local backcountry roads and residential side streets. i use fsd on highway sometimes. i included eap/fsd with my 3P purchase in 2018. have car 3 years (almost to day) and just over 17k miles.
like in life, you can just be unlucky and run into the wrong idiot at the wrong time, and get a poor safety score because of it. i got a couple forward collision warnings in my short travels that really couldn’t be avoided. i don’t have any forced ap terminations.
that said, i get a lot of messages stating i can’t use ap bc camera blocked - different cameras sometimes.
(i made appt to see what was up with that)
i know EM said the hardware (cameras/sensors - i already had nvidia chip replaced with tesla’s) from 2018 will be sufficient enough to integrate with the new software…
either way, im surely not getting beta with my score of 91. and i don’t really care. i’m not a huge social media presence so i have nobody to brag to about it and nobody’s going to see data if i was to post it.
that said. i paid for it 3 years ago. should i or others not be eligible at all for the post-beta production version, then what happens?
HighZ
Member
Actually, I don't. I think Tesla has backed itself into the perception that FSD Beta has to reach this unattainable performance for it to be released when in reality, we have been using versions of beta on streets for quite a while. Many times, I have to take control to save the car and myself, and the same will be true in certain edge cases with this new beta.
I really wish that they would have been working on incrementally improving our version of city streets FSD over the last year rather than wait for this one big update to drop. Maybe they have incorporated some new logic, as I think city streets today is smoother especially over the last couple of updates at places where it used to be an issue.
I think that releasing to a relatively smaller group - maybe a few thousand to begin with and then increase to 20 thousand + over the next 6-8 weeks - will give them (and us) more confidence that it isn't the Armageddon that some predict.
EVNow
Well-Known Member
There are about 2,000 testers now (including employees). What is small (about 200) are non-employees. It would be very common to go from that 2k to 20k testers. Anyway, they will do a slow 1k/day roll out and can stop at any time if there are unanticipated problems ...
EVNow
Well-Known Member
In case it was confusing - my reference was to customers who paid for FSD a long time back.
As I said, depends on the objective. The tweets made it sound like everyone who has
- Bought FSD
- Has a good score
would get FSD beta. People with lower scores would get later, as the rollout is 1k/day.
We don't know what is a good score (someone mentioned 80, haven't seen any tweet that specified it), but, I'd say surely anything 90%+ would surely qualify as a good score. So, you are likely to get it.
jebinc
Well-Known Member
Your future participation is requested here!
teslamotorsclub.com
Wiki - MASTER THREAD: Actual FSD Beta downloads and experiences
22OCT21 Update: We are anticipating the release of v10.3 after midnight (PDT) tonight! Please keep your data current in the attached tracking spreadsheet ("S/S"). Please post your thoughts on v10.3, as either a new user, or one coming from v10.2. What improvements (or setbacks) have you...
teslamotorsclub.com
The false forward collision warnings that I get when a vehicle has turned off of the road in front of me are annoying and really the only dings I get. Reason why the warnings were turned off the day I got the car. pretty much all false warnings.
it kinda sucks to not drive much and be careful and still be under 90 - 3 days with 1 warning that puts the day to 43 or so combined w/ 100s ...driving normally I'd probably be under 2...for real ( I like >1 g on cornering and sometimes I accelerate towards vehicles that are still > 300ft away). Autocross this weekend so if that changes my score I'll be driving 'normal' next week
it kinda sucks to not drive much and be careful and still be under 90 - 3 days with 1 warning that puts the day to 43 or so combined w/ 100s ...driving normally I'd probably be under 2...for real ( I like >1 g on cornering and sometimes I accelerate towards vehicles that are still > 300ft away). Autocross this weekend so if that changes my score I'll be driving 'normal' next week
AlanSubie4Life
Efficiency Obsessed Member
No.
For example, on the second day, I had three trips, 1.2 mile, 1.2 mile, 8 miles. First one was a 96.07 due to 1.9% hard braking (it was a short steep downhill run and I very lightly touched the brakes, sad!). The others were 100. That worked out to a 99 score, with 0.5% hard braking on the daily summary.
That 99 can't be arrived at with mileage-weighted averaging of the individual drives, even if you round the scores first. And the 0.5% is not consistent with mileage weighting of the events.
Furthermore, in practice, we know that if you want to "fix" a hard braking ding, you can't simply drive a long way and get a 100 score, without doing any braking! You have to do some moderate braking.
If they were averaging the individual drives and events like they do for the overall score, you wouldn't have to actually do braking/cornering/following to "fix" a "bad trip."
Events have to be fixed same day with minimum possible mileage, using positive events (obvious exceptions are FCW and AP but those are thankfully usually not an issue). To the extent they can't be fixed, they have to be made up with clean miles on other days.
Last edited:
I think it goes off when the car in front is slowing down and you’re speeding up. Plus some calculation about how soon that will be a problem. I think it’s missing some small delay to see if the driver is going to react or not. Normally I don’t mind the extra alert because I can pretty quickly mentally dismiss it. But right now its a problem since it goes to the score and all …The false forward collision warnings that I get when a vehicle has turned off of the road in front of me are annoying and really the only dings I get. Reason why the warnings were turned off the day I got the car. pretty much all false warnings.
it kinda sucks to not drive much and be careful and still be under 90 - 3 days with 1 warning that puts the day to 43 or so combined w/ 100s ...driving normally I'd probably be under 2...for real ( I like >1 g on cornering and sometimes I accelerate towards vehicles that are still > 300ft away). Autocross this weekend so if that changes my score I'll be driving 'normal' next week
If you’re like way far back that seems to avoid it. But that can be difficult to do in city driving.
TripleG3
Member
Exactly this. I only had a hard breaking incident on the first day, before I knew anything about this. Since then, the city driving has always been 100%, whereas when I'm on the freeway, I've gotten dinged for unsafe follow distance. The first time was about a week ago, when I thought AP was being unsafe with a lane change. So, I disengaged AP and slowed down. I later saw that it dinged me for that. I've been following this thread and have tried to keep all the suggestions in mind while driving. Yesterday when driving on the freeway, I kept my distance from all other cars while entering the freeway and then engaged AP. I didn't turn off AP until I had exited and there were no cars in front of me on the exit. The only time I was driving above 50 was on the freeway and the only time any cars were in front of me was when AP was on and I got dinged for unsafe following. So, to me, my first incident could have been because I disengaged AP (and it probably was) but this most recent drive doesn't make any sense. I MUST be getting dinged while AP is on, because there was literally no other opportunity for unsafe follow distance.
I'm out. I went ahead and used the spreadsheet that @HighZ posted. Thanks BTW.
Results: From the ONE FCW that I had on day two, those 2-3 seconds, I've made my way back up to 95 in the past 9-10 days (after having dropped to 53 that day) but it would take 162 miles perfect over the next few days to get back up to just 98 and 758 MILES to get back to 99.5 score.. so no way no how any of that is going to be happening or is worthwhile IMHO.
Best of luck to all others.
Results: From the ONE FCW that I had on day two, those 2-3 seconds, I've made my way back up to 95 in the past 9-10 days (after having dropped to 53 that day) but it would take 162 miles perfect over the next few days to get back up to just 98 and 758 MILES to get back to 99.5 score.. so no way no how any of that is going to be happening or is worthwhile IMHO.
Best of luck to all others.
Similar threads
