Safety Score

[TIppy]

Going downhill causes the accelerometer to measure deceleration, not uphill. Think about it. If you’re at a stop on a steep decline, gravity is pulling you forward against your seatbelt, exactly like braking would on level ground. On a steep incline, gravity is pushing you back in the seat just like acceleration.

This is true, but I don't think the car uses the y-axis accelerometer to measure vehicle acceleration (change in the car's speed). They have an imu that measures vehicle dynamics. Otherwise they would have difficulty bringing the car to a a a stop consistently as the slope of the road changed. If you were on a 20 percent downhill grade, the y-axis accelerometer would read 0.0 gs if you weren't impeding the car's acceleration with the brakes (ignoring drag for the moment), but the car's speed would be increasing at 0.2gs, like when you are in freefall toward the earth. Your speed is increasing, but you don't feel the acceleration (see vomit comet). Now you put the car in drive, but don't apply the accelerator. The car will start to apply regen, but that is limited to 0.2g. The car stops accelerating so the speed remains constant, and the y-axis accelerometer reads -0.2g. If the car only used the y-axis accelerometer, it would assume it was slowing down, but it wouldn't until the bottom of the hill.

 

[follow.johnny]

For those of us who have a perfect 100 score, can we keep it up for another 7 days?! ugh LOL first world problems

 

[kbM3]

Whoa, take a chill pill. Perhaps the guy was cruising through the desert where nothing ever happens. Which would be perfectly fine. You never know.

I would take a chill pill if it was only his life being endangered. It’s the fact that he’s endangering others that I don’t appreciate.

Yes of course. Nobody else EVER drives in the desert. It’s guaranteed 100% safe.

 

[helvio]

I would take a chill pill if it was only his life being endangered. It’s the fact that he’s endangering others that I don’t appreciate.

Yes of course. Nobody else EVER drives in the desert. It’s guaranteed 100% safe.

I would only suggest that you get all the variables filled in before you make any judgements. You know, assuming makes...

 

[jebinc]

I’m running a 99, but need 500 more miles at 100 to get my weighted average up to 99.5.

 

[ChimpledPot]

I would take a chill pill if it was only his life being endangered. It’s the fact that he’s endangering others that I don’t appreciate.

Yes of course. Nobody else EVER drives in the desert. It’s guaranteed 100% safe.

This is a disgusting and disingenuous way to argue.

 

[_jal_]

As explained by many here, It IS hard braking time / overall braking time. The problem is time on AP doesn’t count. Time on AP acts like an amplifier in the sense that it weights either a good score, or a bad score more heavily in your overall.

E.g. 1 s of hard braking with no other braking, followed by a 500 mile AP Trip would give you a horrible score that would wreck your week.

conversely 1 mile of perfect driving followed by a 500 mile AP trip would give you 501 miles worth of a 100 score for that day.

I understand that. My beef is that in aggregate the hit those braking events cause is amplified by the day's driving when calculating the overall score (the one you see on the first screen of the safety report). What I'm getting at is that all the braking events for all the days I drive should be aggregated and that should be used when calculating the overall score. Like I posted somewhere, if I use the figures shown on the safety score screen and put those in the simulator, then I get a 98, but my overall score is a 94 (actually 95 today) because they are weighing the individual days' scores by miles driven that day.

If the goal is to identify drivers who are likely to brake hard a lot, then you should look at all the time they brake regardless of how many miles they drive in a particular day. When they weigh the day's contribution to the overall score they are making miles driven a independent variable which it isn't when determining the hard braking %

 

[quikshifter]

Have you met the requirement? For your car: 2021.32.22, and for your phone: 4.1.0?

yes to both, no safety score. edit, i see it now.... 26% unsafe following distance lol

 

[drb116]

If the goal is to identify drivers who are likely to brake hard a lot, then you should look at all the time they brake regardless of how many miles they drive in a particular day. When they weigh the day's contribution to the overall score they are making miles driven a independent variable which it isn't when determining the hard braking %

I am not sure the point is to identify drivers that are likely to break hard, but rather drivers who drive in situations where you need to brake hard. If the FSD is going to handle the driving, the situation is more important than the actual driver.

So yes, the way it is set up amplified a bad day, but I think that is the point. If you work from home 4 days a week, but that one day you drive in rush hour traffic, then I think they want that one day to ding you more because the FSD may need to deal with that one day.

Honestly, I don't drive in rush hours or in cities, so I don't have hard breaking. It doesn't mean I am a better driver than someone who does have hard breaking, but it means the FSD beta will probably initially perform better on my driving routes than someone else's.

 

[EVNow]

Honestly, I don't drive in rush hours or in cities, so I don't have hard breaking. It doesn't mean I am a better driver than someone who does have hard breaking, but it means the FSD beta will probably initially perform better on my driving routes than someone else's.

This.

Currently they are not grading drivers - by figuring out where it would be safer to deploy FSD Beta early access.

But, IMO, unintentionally.

 

[brkaus]

Vision car vs radar car, I assume.

Mine is radar. The ultrasonics did beep a few times, but 100% score.

I ordered the CFA sandwich, fries, and 12 count nuggets if that matters

 

[FalconFour]

It needed to be more elastic and smooth as it was unnaturally jerky and made passengers car sick. This was frequent complaint by passengers who then didn't want the driver to use AP at all.

this is just really messed up. For those of us that are driving and monitoring, not the backseat drivers, this new behavior is awful.

Solution would be to have an option of some kind for this, or better on-screen indication of what it's seeing and thinking. Sacrificing driver's comfort for passenger's comfort - without any option otherwise - is a really dumb move. If it were an option, I'd probably pick the middle ground closer to "rigid".

 

[HighZ]

If you have ever looked away from the road for 10 seconds, not only should you NEVER get in the FSD Beta, you should have AP removed from your vehicle.

That is quite a statement coming from someone that doesn't know where or how I drive. I am a very attentive driver but drive in western Illinois where you might not see another car on the freeway front or back at times. I value my life and those of others very highly, especially on AP. My point was that the camera is functioning to provide some alerts. I don't know the time, maybe it's 5 seconds, but there are new alerts, even if I have a peripheral view of the road.

 

[brianman]

If I happen to be able to get my score back up to 100 (actually back to 99.5), I’m planning to only drive my car selectively for the remainder of the test period. Probably just a few low risk miles in a controlled environment on each day if I do reach this goal.

But it’s not clear that I’ll actually be able to get back to that level. Currently at 99.3 or so.

We have two vehicles. For the second vehicle, day 1 was a 98. Yesterday, the average rose from (day 5) 99 to (day 6) 100. Confirmed that "100 average" (reported in the app) does not require "100 every day". FYI.

Just another data point.

 

[brianman]

My grandmother is in her mid-80s and she will scare the urine from your bladder. Not because she's ignorantly dangerous but because she's aggressive like an Italian driver in Rome after espresso in rush hour. People that drive slow for whatever reason are either inexperienced or have a disability.

Or testing new firmware, or gaming a score, or hauling something, or have vehicle/mechanical issues, or recently got a ticket, or ...

 

[brianman]

Think of tending FSD as a game, but with deadly consequences for failure. ....

Ender's Game comes to mind. Perhaps we should think of this "driving with awareness of the score" as a preview of FSD testing.

 

[FalconFour]

Or testing new firmware

don't have to drive slow/under the limit

or gaming a score

don't have to drive slow/under the limit

or hauling something

okay you got that one but come on, it's obvious

or have vehicle/mechanical issues

get off the freeway or use hazard flashers

or recently got a ticket

even more to the point: don't have to drive slow/under the limit.

There's one speed that should apply to all the freeway: the limit (or in CA: limit + 3, ugh lol). Right lane: you go that speed. Next lane over: +3. If the traffic in the right/center lane isn't going that speed, you move to the left, go +3 over that, move back over when it's clear. And the HOV lane is there in case all those lanes are being incredibly dumb/slow/overloaded/stupid so you get a special leftmost lane you can use *when needed* and move the hell over when it's clear again.

If everyone just did that (or if you really want, shift everything so the max speed is truly "the speed limit" but that'd require universal tectonic shifts in society), imagine how much smoother traffic would be.

 

[jebinc]

Mine is radar. The ultrasonics did beep a few times, but 100% score.

I ordered the CFA sandwich, fries, and 12 count nuggets if that matters

Mine's a Vision; it blows.

 

[phantasms]

Hi! Sorry, late to the party here can't can't read 49 pages.

Are the days weighted depending upon how many miles are driven in each day?

I was running a solid 98 for a few days but then I had 2 days where I got hit with the forward collision warning [which I never saw!] while I was dropping my kid off at school. It's a real head scratcher. I was driving completely easy and well aware of all the scoring factors. Never was there ever any risk or anyone turning in front of me while I kept going, and yet, forward collision warning. So, now I'm using my wife's Y to drive my kid to school.

My score is currently a 94 and I'd like to know if I went for a safe/clean really long drive if it would weight heavily in helping my overall score or simply just have an impact on that particular day. Thank you!

FWIW it's a 2021 P3 with vision.

Do the parking sensors count as forward collision warning? Sigh.

Best,
Gene

 

[Twiglett]

Another short 6 mile drive - this time with a "safety" score of 0 - yes, ZERO, thanks to four false positive forward collision warnings. One after another from the "safety" system seeing oncoming cars (in the correct lane) and triggering false FCW.

Now AP with all its inconsistencies and inaccuracies is now rating real drivers - its like asking a five year old to grade your dissertation.