View Single Post
Old 05-12-2011, 12:21 PM   #3679
qqwref
stepmania archaeologist
Retired StaffFFR Simfile AuthorFFR Veteran
 
qqwref's Avatar
 
Join Date: Aug 2005
Age: 34
Posts: 4,090
Default Re: Ingame Song Information

Quote:
Originally Posted by jimerax View Post
If we determine difficulty rating by people's scores (called statistical method), it will be take so long to get relatively accurate difficulties. I say at least n=5,000-10,000 is required.

I think it should be considered for difficulties as one of factors (along with technical factors like bpms, framers, tps) though, it's just difficult to compare the parameters between songs due to the number of players, release date, etc.
Come on, you guys... I've already thought about stuff like this, that's why my proposal is different. This will not work and I don't know how many times I have to say this. If you randomly compare technical factors you will end up with a number that represents something roughly related to difficulty, but certainly doesn't represent what you want - a measure of how much effort and skill a typical real, human FFR player will need to get a certain score on the file.

Here is the biggest problem: even if you could make a formula like that, you would never be able to calibrate it. Exactly how hard is a 207 bpm anchoring pattern like 33[34]44, for an average player? Should the difficulty of a file like Crowdpleaser depend on the length of the file? How about a file like EHHS? (impossible to know, no, and yes) This is pure pseudoscience, and it's just going to be used as a silly justification of someone's opinions. You're never going to get a useful difficulty measure by looking at some formula, but only by looking at actual scores from actual players.


I'll outline my idea again. You compare two files by simply taking pairs of scores, the best score from the same player on each file, and comparing them by how far they are from the AAA (that is, PA). After you discard pairs that don't tell you anything useful (people who get over 100 boos on either file, people who haven't completed both files, people who have the same score on both, etc.), the file that has a worse score in >50% of the pairs is more difficult. A technique like this will produce a useful measure of difficulty, even in the worst case scenario, because it automatically gives a higher rating to files most players in the community have difficulty with.
__________________
Best AAA: Policy In The Sky [Oni] (81)
Best SDG: PANTS (86)
Best FC: Future Invasion (93)
qqwref is offline   Reply With Quote