I Paid $6000 For Tesla FSD & Now Have The Privilege Of Taking Tesla’s Totally Unfair Safety Test





Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

To be fair to Tesla, I am in crazy love with my Model 3, which I have owned for two years and driven 50,000 miles. I paid $6,000 for Full Self Driving  (“Tesla FSD”) when I bought the car. I have obsessively used every bit of Tesla automation, and for the most part, I think it is fantastic. I just drove my Model 3 1500 miles from Northern Wisconsin to Utah in three long 500 mile days. I’m 81 and Tesla’s steering assist made the long days very much less stressful.

Furthermore, I am dying to get what Tesla calls “Feature Complete” Full Self Driving, which allows the car to navigate from point A to point B on city streets without intervention under ideal conditions.

I have been obsessively following the reports of some of the ~2000 testers of what is now typically referred to as “FSD Beta” for the past year as Elon Musk has been promising the general release of the software, which has basically been the case for over two years. About a month ago, Elon said that those who had paid for the software would get it next Friday. Well, next Friday came and went and Elon said that he had really meant that on that Friday we would get a button that would allow us to download FSD Beta software. When the time came, I got the software release 2021.32.22, which had the button that I pushed immediately.

But no, we didn’t get the Tesla FSD Beta software — we got the privilege of taking a Tesla automatic “Safety Test” to determine if we were worthy to get the software. We didn’t get any rules or criteria. We found out that we would be getting a daily score and those who scored high enough would get the software after a week. Again, no clear rules or criteria. My score the first day wasn’t great, but as time went on, my scores were in the 95 to 100 range. After a few days, I was getting some 99s, and after a week, I frequently got perfect scores of 100. My average was 98. I also realized that if you only drove 10 or so miles per day and concentrated on no braking more than regenerative braking provides, no turns faster than the speed given on the turn signs, and no close following, you could easily get a score of 100. However, if you actually had to use your car for transportation and had to drive 50 to 300 miles per day, it was really hard to get a perfect score.

After I had been taking the test for a week and had an average score of 98, we were informed that only those with a perfect 100 score would be getting the download. I was bitterly disappointed. I have a PhD and have taken a lot of tests in college, and as I recall, a 98 was an A+. Really, an A+ wasn’t good enough. My chief editor, Zach Shahan, did get a perfect score on the Tesla FSD driving test, and got the download. I enviously read his reports on using the software.

I was hoping that if I got enough perfect 100 scores, maybe I could raise my score to 99 and be among the second group to get the software. However, then I did the calculation, and because of a bad first day score and a number of 95s–98s, I would need 20 days of 100s to raise my score to 99. That wasn’t going to happen. There are now 30 daily reports showing up on my App. See below. It looks like the first days have started to drop off. I have now had a perfect score of 100 on 14 different days and a 99 on six different days, so under ideal conditions, I do know how to get a high score. But apparently that is not good enough. You will note that I got 100 on a few days when my mileage was in the 60 to 80 range, but most high scores were on low-mileage days.

Tesla Model 3 lineup at a Tesla Supercharger in Eau Claire, WI. Photo by Fritz Hasler/CleanTechnica,

Now we come to the backbreaker. My wife and I had to make the long 1500 mile trek from Northern Wisconsin to Utah in three ~500 mile days. We were barely 50 miles en route on October 23rd when three deer charged across the highway and I had to brake hard to miss them. I was dinged 1.7% for Hard Braking. What was I supposed to do? Run into the deer doing only regenerative braking? This is what I mean by the test being totally unfair. Then we had to drive across the entire St. Paul/Minneapolis metropolitan area at rush hour. Surprise — I got dinged 20% for Unsafe Following. I challenge anyone to travel across a large metropolitan area at rush hour with drivers cutting in front of you without getting dinged for Unsafe Following. My rating was 96 for a 462 mile day — again, totally unfair.

Tesla Model 3 with two e-bikes on rack in South Dakota. Photo by Fritz Hasler/CleanTechnica.

Miraculously, on October 24, I managed to drive the 596 miles from Worthington, Minnesota, to Lusk, Wyoming, with a score of 99. However, if day #1 was a backbreaker, day #3 was a 596 mile total disaster. I was driving through Casper, Wyoming, when I came around a blind corner to see a red stoplight at the bottom of a short steep hill. I had to brake really hard and got dinged 4.8% for Hard Braking. What was I supposed to do, run the red light? I considered it, but there was too much traffic. Later, I had a brain lapse and decided to pass a car on a totally straight section of two-lane level road between Casper and Rawlins, Wyoming. I got dinged 4.5% for Unsafe Following. Absolute total disaster — I got an 88 for the day and I can even kiss my 98% rating goodbye now. I am currently rated at 96% and doubt if I can raise it again in the foreseeable future.

In the meantime, those with scores of 99 also got the FSD Beta software. Will I ever get it? I am not holding my breath. My only consolation: Zach Shahan is very disappointed in the way that Tesla FSD Beta software handles a number of situations. However, I would be very pleased if the FSD Beta software would get me over the river and through the woods to grandmother’s house on country roads without intervention.

We are now back in Utah right near 6-lane State Street and 12-lane I-15 with heavy traffic all day long. It took me a day to get used to driving in traffic again, so I got a 94 the first day in 27 miles, but today I am back up to a score of 100 in 18 miles.

Bottom line: the Tesla FSD Safety Test works great for drivers driving 10 to 20 miles per day in light traffic with 100% concentration on not breaking any rules, once you learn the rules. It is totally unfair for drivers driving high daily mileage and encountering unavoidable situations like avoiding hitting a deer, rush hour traffic, an unusual traffic light after a blind turn at the bottom of a short steep hill, or passing a slow-moving car on a two-lane road.

My Tesla FSD Driving Test Daily Safety Scores

My average was 98 until I made the long trek from Wisconsin to Utah. I had to brake hard to miss a deer dashing across the road in front of me. I had to cross a major metropolitan area at rush hour. Finally, I had to brake very hard to stop at a stop light at the bottom of a very short steep hill. These events brought my average down to 96.

Date — Score — Miles Driven

Sep 26 – Scores for this date and previous are no longer reported

Sep 27 – 97 – 45

Sep 28 – 95 – 28

Sep 29 – 99 – 35

Sep 30 – 99 – 64

Oct 01 – 99 – 46

Oct 02 – 96 – 41

Oct 03 – 100 – 10

Oct 04 – 100 – 67

Oct 05 – 96 – 88

Oct 06 – 100 – 30

Oct 07 – 98 – 312

Oct 08 – 100 – 48

Oct 09 – 100 – 9

Oct 10 – 97 – 63

Oct 11 – 100 – 39

Oct 12 – 98 – 325

Oct 13 – 100 – 81

Oct 14 – 100 – 12

Oct 15 – 99 – 85

Oct 16 – 100 – 15

Oct 17 – 100 – 63

Oct 18 – 100 – 11

Oct 19 – 99 – 32

Oct 20 – 100 – 60

Oct 21 – 100 – 48
(Average to this point was 98.)

Oct 22 – 96 – 462 

Oct 23 – 99 – 596

Oct 24 – 88 – 558
(After this disaster, my average dropped to 96.)

Oct 25 – 94 – 27

Oct 26 – 100 – 18


Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
Sign up for our daily newsletter for 15 new cleantech stories a day. Or sign up for our weekly one if daily is too frequent.
Advertisement
 
CleanTechnica uses affiliate links. See our policy here.

CleanTechnica's Comment Policy


Arthur Frederick (Fritz) Hasler

Arthur Frederick (Fritz) Hasler, PhD, former leader of NASA Goddard Space Flight Center Scientific Visualization & Analysis Laboratory (creator of this iconic image), and avid CleanTechnica reader. Also: Research Meteorologist (Emeritus) at NASA GSFC, Adjunct Professor at Viterbo University On-Line Studies, PSIA L2 Certified Alpine Ski Instructor at Brighton Utah Ski School.

Arthur Frederick (Fritz) Hasler has 136 posts and counting. See all posts by Arthur Frederick (Fritz) Hasler