The biggest problem with the track is that it is so much faster than reality. Even using the new version 2.1, I did 1:24.815 at 36 °C air / 51 °C track (the highest available values in vanilla AC) with the SDD Ferrari F1-75 (which is VERY close to real times in Melbourne 2022, for example, because my WR there (1:16.0) was done at 10 °C air instead of the much higher ~25 °C air IRL).
Although Pyyer wrote about a skin for non-CSP users in v2.1, the banners still flicker the same. But it has no effect on my times, so no big issue there.
The real Ferrari F1-75 accelerates MUCH better (during the pole lap, Charles has 269 km/h at the last DRS line from his minimum speed of 65 km/h in T17, I only manage 257 km/h from 70 km/h - I tried accelerating from T17 several times and the best was just 259 km/h at the DRS line), reaches the SAME top speed (seriously - Charles, Carlos, and me, all 336 km/h), and brake MUCH better (during the pole lap, Charles brakes for T17 at ~115 meters, me ~140 meters ). It is, therefore, extremely strange that my overall lap time is almost 4 seconds better at similar track temp (IRL, the Q track temp was ~52 °C, according to formula1.com Q lap-by-lap), despite being slower than Charles in certain corners!
Pole lap 1:28.796 with telemetry
https://streamable.com/g3j96x
Max Q3 1:28.991 with telemetry
Ferrari/Mercedes/Red Bull Q telemetry
My WR 1:24.815 at 51 °C track with Miami v2.1
Anyway, thanks to Pyyer for the effort put into this!