Crowdsourcing Data for Rates

[rushyrulz]

Re: Crowdsourcing Data for Rates

I should play a bunch of outlier files like Fei Longer and Waterfall to see how you deal with the major inconsistency.

[Rapta]

Re: Crowdsourcing Data for Rates

Shouldn't you do something like lock all legitimate scores so no data is messed with or deleted?

[Trumpet63]

Re: Crowdsourcing Data for Rates

Shouldn't you do something like lock all legitimate scores so no data is messed with or deleted?

I can do that, definitely. And Google sheets automatically records revision history, so I could revert to an earlier version at any time.

[noname219]

Re: Crowdsourcing Data for Rates

I was trying to come up with something similar to this after rates were being introduced. Basically, I wanted to find a formula that makes all songs feel like you're playing on the same difficulty (using rates). For example, you could play Fei Longer like a 77 or Eradication like a 35. My biggest problem was that not all songs scaled very well (just like OWA explained) so every song should have its own coefficient.

[Trumpet63]

Re: Crowdsourcing Data for Rates

moar plz?

[Trumpet63]

Re: Crowdsourcing Data for Rates

Um, new rule, I need everyone to give me data for at least two songs of different difficulty.