Kann man sich tp zusammenfarmen indem man leichte maps spielt, die unbliebt/ unbekannt sind und auf denen man so leichter einen höheren rang erreicht? (im vergleich zu pika girl, gangnam etc)
Nein.
Ask @Tom94
Previous
Does SV have any effect on the difficulty rating of a map?
Yes, obviously it changes the speed at which sliders move, thus indirectly matters for aim difficulty on maps with fast sliders.
Are speed points mostly given by fast, streamy maps?
Yes. It's about clicking speed, not movement speed.
Related users
"BPM doesn't really say much about the map. For example big black is 360 BPM while it plays like most 180 BPM maps in terms of speed." but it would still be nice, for example i want to find maps around difficulty lvl 50, but i dont want anything above 200 bpm or below 160 bpm.
Shouldn't speed difficulty be a well enough indicator for it then? I might add BPM sorting though if it keeps getting requested.
eta on acc/speed/aim banners?
Either this week or in more than a month. I'll be in Israel starting from Sunday.
Say if you had a very high aim and low speed/acc, does that mean that doing high speed/acc maps will give more ranks than high aim?
Yes.
Could you please add so that is it possible to see the BPM of a map when you look at it on tp page?
BPM doesn't really say much about the map. For example big black is 360 BPM while it plays like most 180 BPM maps in terms of speed.
I am sorry but your argument for high ar reading is invalid,your logic would apply for all mods then, I could learn every hr/hd ''by heart'', we are talking about skill measurement here yet high acc hd/hr gives insane amounts of tp just cos of acc, and the actual skill needed for ar11 is not awarded
Acc as well as aim can't be "learned by heart". True, AR reading is a different skill set, but I can't increase the weighting of it unless I create a completely new diffvalue which would be "reading". Might come at some point, and it wouldn't only feature AR but also pattern shape and overlaps.
did you taste milkshake's milkshake yet?
Why yes, she makes very good milkshakes.
http://ask.fm/Tom94/answer/47204357326 does it change anything if we do a 100 on a slider or circle? if yes, would it be possible that when we have 1 less combo than max combo it could suppose one 100 is on a slider or something like that? (1-2)
That'd only work in the one scenario that you are 1 below maxcombo without misses. As soon as that becomes higher the possibility of a single sliderbreak for ANY maxcombo difference is there. :/
I continued the idea of calculating accuracy based on standard deviation. Here are images of the formulas (for Mathematica) and some results: http://i.imgur.com/yMAqdIm.png http://i.imgur.com/Uswkl4J.png http://i.imgur.com/fXDARSJ.png http://i.imgur.com/XyNsmSO.png. What do you think?
What I am currently doing is using max(0, (300s - Sliders - Spinners) * 6 + 100s * 2 + 50s * 1) / (Hitcircles * 6) to calculate accuracy. (Note: I don't have information about whether there were Slider 100s / 50s or not, same with spinners etc.)
Yours also makes sense and I am not yet sure which one to use yet. (They're pretty damn close anyway.)
Using the expected std deviation sure is a nice idea, however all it does is transform one monotone scaling into another. Even though I like the way it does the high-combo- and OD-scaling for me, so I might switch to using it instead. Thanks for the idea!
(Also, dw, I know the formulas, had a lecture about probability and statistics a few semesters back.)
Yours also makes sense and I am not yet sure which one to use yet. (They're pretty damn close anyway.)
Using the expected std deviation sure is a nice idea, however all it does is transform one monotone scaling into another. Even though I like the way it does the high-combo- and OD-scaling for me, so I might switch to using it instead. Thanks for the idea!
(Also, dw, I know the formulas, had a lecture about probability and statistics a few semesters back.)
how can i know what ar, od, cs has a map? doesn´t appear anywhere i mean without looking in the tp web
Go into edit mode in the game and check the difficulty settings.
Does that mean that 100% acc is infinite times as good as any other acc? Wouldn't be better to calculate accuracy based in how stable someone has to be to get that accuracy (modeling errors as a normal distribution, without using statistical data of course). Example, in OD 10 (+-18ms error to get a
That's how it's done already, at least partly, since I have no per-hitcircle information about what the user got or what his timing was.of course 100% isn't infinite times as good as any other acc, because accuracy boils down to an approximation of the player's stability from below. That means the little error the player still has when he scores 100% is washed away by the leniency of the game.
All that being said, accuracy scales up exponentially as it approaches 100%. (And is combined with an exponential scaling of OD or the +-ms... doesn't really make a difference because both are linear dependent.)
All that being said, accuracy scales up exponentially as it approaches 100%. (And is combined with an exponential scaling of OD or the +-ms... doesn't really make a difference because both are linear dependent.)
as you said, HD increase acc just 2%...i think thats not enough...seen like here: http://osutp.net/scores?bid=107747 ..just 1% acc difference and the dt rank beats the dt,hd rank....dont think that this pictures the reality D:
I hope you are aware, that 99.5% is -three times- as good as 98.5%. That is due to the nature the percentage is calculated, you need to compare the distance to 100%, not the distance to 0.
99.5% only has 5 100s on that map while 98.5% has 15 100s. I think it is very fair on that map.
99.5% only has 5 100s on that map while 98.5% has 15 100s. I think it is very fair on that map.
HD bonus ought to be proportional to the AR of the map (the current system rewards AR0 HD as much as AR10.3 HD, which is clearly not right.) What do you think?
That's not true, the current system already rewards HD much more at lower AR. Don't just assume things. :)
In my opinion HD should increase acc more because it is harder to be as accurate without the approach circle
HD does increase acc. But only by 2%. Most of the acc doesn't come from reading but from listening to the hitsounds / song.
http://ask.fm/Tom94/answer/47010015950 it'd be unfair for those who can't play the map in future (some unranked map may never be ranked again for various reasons)
Keep in mind that tp doesn't use statistics like pp, so there (at least in theory) shouldn't be easy maps that give lots of tp. So in consequence it also wouldn't be unfair for people who can't rank on the map anymore... since the effort to tp ratio should be equal to any other map.
When a ranked map is unranked, but an update has already been performed and the tp added, will the next update remove the tp (due to the map being made unranked), or will it remain due to the score having still been achieved?
The way the system works right now it'll keep the scores and keep the map ranked internally. I don't really see a reason for not doing it that way, too.
Do you take the best rated scores or do you take the best aim scores to calculate up for the aim? Because my best aim score doesn't even come close to being a top rated score for me and I wonder if it doesn't count for that reason. Same for speed and aim.
The best aim scores are taken for calculating your aim value, same for speed and acc.
http://osu.ppy.sh/b/23984?m=0 why this beatmap doesn't count in tp?
Should be fixed with the next daily update. Appearently I was missing about 10 ranked beatmap sets from my listing.
OH MY GOD OSU TP WORST SITEN EVER WORKS OMG N000B FGT 420 BLAZIET
A statement from GladildOol. I feel honored.
http://osutp.net/scores?bid=138830 you think this big difference between HD and HR is right? (by this example Hr 99,48% = 196tp; HD 100% = 133tp) I mean expecialy on maps with ar8 or less, HD can be much harder then Hr even though +Hr would be Ar10. I think the relation could be better :)
You need to take a look at the aim / speed / acc values seperately. Aim barely increases while the big difference between those scores is the accuracy value. (HR dramaticly increases OD after all.)
Also keep in mind, that the 3 values -independently- affect your aim / speed / acc value. That means if you play the map with HD HR instead of HD you barely gain any aim rating, but you gain a lot acc rating.
Also keep in mind, that the 3 values -independently- affect your aim / speed / acc value. That means if you play the map with HD HR instead of HD you barely gain any aim rating, but you gain a lot acc rating.
Right now, very small amount of players can play ar11 actually, players like beatrice and 5321kinoko and many others simply can't play ar11(ar10.3), it deserves a LOT more tp.My only ar11 fc gave less than kpop shit, also nitro,metro,shadowsoul > TTTL,dragonhuman.Really disappointed from the system
ShadowSoul / Metro / Nitro all can jump better than TTTL. Not even speaking of their far superior accuracy compares to those lower players that you mentioned. This is not only about speed and aim, osu! is not called a "rhythm" game for no reason. Accuracy is already worth only a third of your tp ranking and as you surely have noticed usually has a lower value than speed or aim.
Now regarding AR: Reading high or lower AR sure is a special skill, but it doesn't have so much to do with the actual aim / speed / acc required. You can always learn a map by heart and play it with FL and you won't even notice the AR difference. That's why I think a smaller bonus for those ARs is fair enough. (To give you numbers: AR 10.33 gives a 15% bonus, AR11 scales up to 45%. Anything below AR8 also gives a bonus.)
Now regarding AR: Reading high or lower AR sure is a special skill, but it doesn't have so much to do with the actual aim / speed / acc required. You can always learn a map by heart and play it with FL and you won't even notice the AR difference. That's why I think a smaller bonus for those ARs is fair enough. (To give you numbers: AR 10.33 gives a 15% bonus, AR11 scales up to 45%. Anything below AR8 also gives a bonus.)
Liked by:
Tshemmp
I type this cuz im bored, not like my life depends on it but I am surprised noone said that(didn't see at least).The system became close to pp with the recent acc implementation, a 99% dt average(not average but even n00B) score surpasses my best 220-230bpm ulta hard for me to do dt 93%
It's not like pp is entirely wrong. I didn't want to make a system that only relies on aim and speed. Also keep in mind that your high speed and high aim scores ARE considered a lot, even if their total value is lower compared to your accuracy scores.
It would be nice if the banner showed in each player profile (with overall rank), your thoughts on it?
It would. But I might have a little solution for you: https://chrome.google.com/webstore/detail/osutp/cjdcemlbechjblaeknimeoljkafkecpi?hl=de&gl=DE
Oh, and for the FireFox users: http://puu.sh/3fQUQ.xpi
This AddOn will eventually be released on the osu!tp website... Credits to Nafar for making it.
Oh, and for the FireFox users: http://puu.sh/3fQUQ.xpi
This AddOn will eventually be released on the osu!tp website... Credits to Nafar for making it.
Next
