Did some data collection so I could try out the Kruskal-Wallis test for some insights. This particular session was meant to see how different the setups needed to be before you would start getting noteworthy differences in lap time. SciPy was going to be doing most of the heavy lifting, which is so much faster than trying to do it all by hand.
First goal was to collect enough data to reject H0 (There is no statistically significant difference in medians). To do that I needed get a P-Value under 0.05.
I checked p-values after every pair of ten laps.
There are some features I want to add to this file, but uh... baby steps.
Combo
Car: Audi VGT
Tire: Racing Medium
Circuit: Blue Moon Bay - Infield B Reverse
Time of Day: 10:00 AM
Laps Completed: 154
Wind: 9.8 ft/s, South-West
Setups and Laptimes
The only thing I wanted to play with here was the natural frequency. Adjusting different downforce levels/ratios will be something for later.
All setups use 1650 Lv. Downforce at ~44F:56R.
Setup Name | Natural Frequency (F) | Natural Frequency (R) | Mean | Median | Comments |
Default | 4.00 Hz | 4.00 Hz | 43.098s | 43.067 | |
Test-1650.1 | 4.23 Hz | 4.13 Hz | N/A | N/A | (Lap-times replaced due to increasing P-Value; NDefault = 30*, NTest = 30) |
Test-1650.2 | 4.43 Hz | 4.29 Hz | 43.153s (+0.055s) | 43.194s (+0.127s) | (NDefault = 63*, NTest = 61) |
* in the interest of time, I reused the data I had for the default setup.
...Had some interesting times on the board.
Pretty tired now! T
--
Finally got around to finishing the X2019 Nations and Jimny cafe menus, along with some road car events on Yamagiwa & Tsukuba. Did these before I started the car testing.