Closed zanaptak closed 4 years ago
Thanks for the feedback, this is what I couldn't figure out when trying to crack the formula :) I will get back to it later, and update it.
Not that it matters I'm sure due to number of players, but for simplicity I removed the zero-point cutoff in the formula above, it was based on some speculation that max of 1 million players get points, but the dev tweets are only mentioning one million as max possible campaign score so I'm not sure a cutoff on the low end exists.
Sorry for digging this back up again, but I'm now allowed to disclose what I was told by one of Nadeo employees, since the Open Grand League is using it as well. This is moreso for documentation purposes.
@zanaptak's formula is indeed correct, but there is another expression of it. Here's the pseudocode
fun calculate_sp(position):
tier = ceil(log10(position)) # make sure that there are no floating point errors here that might cause ceil to misbehave
if tier == 1:
return round(40000 / position) # make sure you do floating point division here
A = 4000 * pow(5, tier-1)
B = 3600 / pow(2, tier-1) # make sure you do floating point division here
return round(A / position + B)
And here's a spreadsheet to show how it works: https://docs.google.com/spreadsheets/d/1xMf8wuIHbvZGioItZ8R36BXUSDloT356FP0DjIXSXVE/edit#gid=1258500882
Would you consider updating the formula for this? Currently the calculator points appear to be inflated compared to in-game data.
Based on in-game data and some Excel exploration I derived the following. It appears to be accurate to the exact point value on multiple campaigns I checked in-game (Fall 2020 and some club campaigns). Pseudocode:
The misunderstanding appears to be that the 10^x values mentioned in your referenced tweet are not actually points on a smooth global curve but rather represent rank tier boundaries, and within each tier there are isolated point curves with an early point dropoff.
Edit: Removed zero-point tier from formula.