Rated Miles stepping down with firmware updates?

[ZeroDarkSilver]

When I took delivery of my P90D, a 90% charge was 252 Rated Miles. This was firmware 2.20.45.

After the upgrade to 2.24.30, the 90% charge dropped instantaneously to 246 Rated Miles.

After the upgrade to 2.36.31 (the minor update immediately before 8.0), the 90% charge dropped again to 242 Rated Miles.

The car is only a few months old; it seems unlikely it has actually lost 10 miles of range already. And these step changes only happen immediately after firmware updates. It feels like Tesla is changing the RM calculation in the firmware.

Any ideas on what's happening here?

 

[Patrick W]

FWIW, my 85 (no P or D) is about a year and a half old. When I first started charging to 90% right after I got the car it charged to the upper 230s.

Now, with the car recently updated to 2.36.108 it's still charging to the upper 230s.

 

[rfmurphy81]

Pulling a post of mine from What's your 90%? as it seems relevant here:

I noticed something interesting since upgrading to 2.36.25 (and still the case on 2.36.31). My 90% rated range always stopped at 219.91 (showing 220 rated miles on the IC). Since 2.36.25, it's now stopped at 216.58 (showing 217 rated miles on the IC). My first 90% I could find on TeslaFi was from 7/5 when I was on 2.20.30 which shows 219.91. I haven't read a majority of this thread so I don't know if this happens often, but it seems like they changed the calculation in 2.36.25 and beyond. I'm curious if 2.36.108 has changed the rated miles. Has anyone who has 8.0 noticed a difference in their rated miles since upgrading?

I have a feeling like Tesla has been modifying their magic formula for rated miles to probably be more realistic. I had noticed that most times, when I charged to 100%, it seemed like the first 5-10% went by faster than the rest of the battery, but that's purely speculative.

Since that post, I've upgraded to 2.36.108 and have completed three 90% charges. TeslaFi shows all three as 216.95 rated miles, so it definitely seems like the formulas are tweaked in each release. I'd be curious to see if there is a correlation between charges at specific SOC limits against firmware versions.