@Janio : Adding more random notes to make harder chordstreams and whatnot is not changing the gameplay. Autoconverts are bad in the way they are... "mapped" :that is, there is no logic in how the notes are falling in anyway. There is a difference between a stupidly hard map without any sense and an extreme but well mapped map.

Tom94 wrote:

it would be great if you could give me hints of what the hardest patterns are and how to rate difficulty. Any X-key feedback is welcome!

technically there's no such thing as harder "patterns" (due to shape etc), although pretty much everybody would agree that ring/middle trills are a problem to play (which is also why shields are hard) - at least in the beginning - but this is easily neglected by players playing better and doesn't weight enough to make a big impact (as in, trills are just about as hard with all other fingers too)

in the end it's all density (where 1 LN = 2 density) and jacks/finger usage repeat in short timeframes.

also stuff like 8 1/4 notes is harder than 4 chords of 2 notes, so 'amount of timing snap divisions peaks' might also be a "thing".

Tom94 wrote:

I'd like to mention a few things considering how I plan on tackling this.

The general direction:

1. First of all, everyone can decide on his/her own which maps he deems good or bad, or which ones he/she prefers playing. We shouldn't be forcing certain maps on people by making them worth more for no sensible reason.
2. I'll try to make an algorithm that's fair for both auto-converts and mode-specific maps. We will all see how it turns out, and if it's alright, then I see no reason to reduce the value of auto-converts. As mentioned above you are still free to play the maps you prefer playing. I can however understand where the auto-convert argument is coming from, and it'll definitely be addressed if there won't be a good solution that weights all maps the same.
3. Default X-key modes exist for each specific beatmap for a reason. My initial plan is to give slightly more points to the default X-key mode of a particular beatmap than any custom one. This is not set in stone yet and most likely will change in one way or another while testing.
4. Unfortunately X-key specific ranking is not going to happen. That'd be for one part a way too huge amount of things to manage and for the other would clutter up everything.

Beatmap difficulty:

This is were you all can help the most. My current concept is somewhere along the following: Similarly to how tp works with osu! standard maps I would like to instroduce strain values to measure how hard it is to play a certain map. This essentially means, that a higher note density will also be considered harder, while the algorithm is aware, that using a single hand or finger multiple times in a row is more demanding than simply alternating.
For mods the map will simply be transformed accordingly to what standard does in the moment. HR will increase OD and AR, DT the speed at which the map plays back, etc.
I have not yet gone in-depth and it would be great if you could give me hints of what the hardest patterns are and how to rate difficulty. Any X-key feedback is welcome!

You can look into the thread I linked earlier (my side note post). There is pattern repository which can give you an idea of what's hard and what's not.

Please do not make HR change AR. AR is something that is really depedant on players. Some play better with lower, some with higher. I like it how it works now (everyone can choose their scrolling speed). Also I really appreciate your attempt! awesome
If you need like any specific help, feel free to message me. Like I don't feel like actually coding myself atm but I wouldn't mind helping out here and there on the actual algorithm. I'm not a great mathematician nor programmer nor player, but I do have a some understanding of how they work.

these last arguments I've seen make me glad I'm not involved in this

"everybody is wrong except me" - everyone

autoconverts were not made for mania, that's a fact, autoconverts don't play like real mania maps- especially on higher levels, that's a fact too. as a matter of fact they play like shit and shouldn't ever be used as an example of mania done right but you're free to enjoy your shit and roll in it as much as you want

map compelxity is not given only by density, that's a fact too (e.g. so called 'gay' patterns)

if you'd say the devil is in the details a tiered system for charts has been the way the insane bms scale does it and it works. o2pf does this as well, and guess what. it works.

calling a shitty autogenerated chart better than charts that have been thought from scratch (and in detail) is honestly insulting to all the communities that came before osu!mania like it's the grand savior of vertical scrolling rhythm games

guess what: it isn't. thousands of shitty autogenerated charts that enjoyable are miles away from the real deal, and it's easy to see that a lot of people here have been spoiled (which I feared from day one)

but hey, I'm some sort of troll in everyone else's eyes so disregard me

Xcrypt wrote:

Please do not make HR change AR. AR is something that is really depedant on players. Some play better with lower, some with higher. I like it how it works now (everyone can choose their scrolling speed). Also I really appreciate your attempt! awesome
If you need like any specific help, feel free to message me. Like I don't feel like actually coding myself atm but I wouldn't mind helping out here and there on the actual algorithm. I'm not a great mathematician nor programmer nor player, but I do have a some understanding of how they work.

I don't know what I was thinking when I wrote my post. Of course, HR will not change AR in mania, my bad. Must have been with one of the other modes in my head.

Agka wrote:

these last arguments I've seen make me glad I'm not involved in this

"everybody is wrong except me" - everyone

autoconverts were not made for mania, that's a fact, autoconverts don't play like real mania maps- especially on higher levels, that's a fact too. as a matter of fact they play like shit and shouldn't ever be used as an example of mania done right but you're free to enjoy your shit and roll in it as much as you want

map compelxity is not given only by density, that's a fact too (e.g. so called 'gay' patterns)

if you'd say the devil is in the details a tiered system for charts has been the way the insane bms scale does it and it works. o2pf does this as well, and guess what. it works.

calling a shitty autogenerated chart better than charts that have been thought from scratch (and in detail) is honestly insulting to all the communities that came before osu!mania like it's the grand savior of vertical scrolling rhythm games

guess what: it isn't. thousands of shitty autogenerated charts that enjoyable are miles away from the real deal, and it's easy to see that a lot of people here have been spoiled (which I feared from day one)

but hey, I'm some sort of troll in everyone else's eyes so disregard me

Your opinion is not disregarded. However it's not the purpose of a ranking system to guide the players towards what they want to play. It is merely there for rating performance, and that depends solely on how hard a map is, not on how subjectively well-made it is.
Of course "hardness" already is a very subjective thing, but let's not overcomplicate it by adding more factors. As I said, if a general system won't work, then another solution will have to be found.

Tom94 wrote:

For mods the map will simply be transformed accordingly to what standard does in the moment. HR will increase OD and AR, DT the speed at which the map plays back, etc.

So mods are going to come in and affect your pp?

For your difficulity judging: An algorithm is nice and all, but I think there is also a alternate possibility of letting experienced players determine how hard that map is before it becomes ranked. That would probably be quite accurate (sort of like the level system in o2, except better because o2 has a lot of 'wrong' levels on maps.)
Also Agka I think you are right but I also think you are a troll haha. Maybe you can share how the insane scale BMS or o2pf system works?

Alright, after seeing that side of it, I'm gonna have to go ahead and join your side.

At first I thought autoconverts shouldn't be counted. They're pretty meh compared to actual mania-spec songs. I do play a good amount of converts though, but only because I like the song. I'm not really big on how they're converted. It irks me. Because of that, I wouldn't mind if they weren't ranked, but I didn't think about the other side. I forgot what it was like to be in that position. When I started off, I couldn't even touch mania-spec songs. They are challenging and the patterns can be off-putting. I wanted to have fun, not fail instantly.

I'm going to go ahead and say that autoconverts should be counted in rankings for now. Make them worth much less than mania-spec songs, but still include them. At least until we have a sizable amount of songs that less-skilled players can have fun with. If we kick out autoconverts, we are kicking a large amount of people from the rankings. That doesn't sound very fair. Just because they can't pass or don't enjoy the mania-spec songs because of their difficulty, doesn't mean they shouldn't be allowed to be ranked.

The problem is that autoconverts actively hurt the mode and its community. They give people a wrong idea and mislead new players in multiple ways.

1. The keymode of autoconverts goes up with difficulty, what makes people think that you can "work up" from 4K to 7K, while that's not the case. You could be playing 4K for years and still be unable to play 7K at all. You're supposed to learn 7K (or maybe even 8K) as soon as possible, after you can play it you can manage all lower keymodes with just a few plays. We have a lot of players who started with 4K and never moved up because they got comfortable with their 4K skill, while they'd have to start over to learn 7K.
2. They often have nothing to do with music. That makes people think it's what the mode is about, while real mania is about playing the music, not playing along with the music.
3. They don't feature patterns. You could get great at mania but as soon as you see something like brackets or complex LNs you're clueless in real mania. Autoconverts don't even feature something as simple as one LN starting at the same time another one ends.
4. They distract players from the fact we don't have a lot of maps. Unranking autoconverts would make people notice that and in addition to more real mania players we'd also get more mappers.

Agka wrote:

calling a shitty autogenerated chart better than charts that have been thought from scratch (and in detail) is honestly insulting to all the communities that came before osu!mania like it's the grand savior of vertical scrolling rhythm games

Just to clarify... I'm in no way calling autoconvert maps "better constructed" than mania specific maps. Yes, I realize that a lot of work goes into making actual mania specific maps, which I'm sure are great for people at that level. The problem that I ran into was whenever I tried to play even the "easy" 1 star mania specific maps was that they were too complex for my skill level. I'd compare them to "difficult normal" or "easy hard" autoconvert maps.

I don't know about most people, but I couldn't make the jump straight into mania specific maps due to their complexity, which left me feeling discouraged from either constantly failing the supposedly "easy" maps, or barely keeping up with them, which killed my enjoyment of the song. This is why to me autoconverts were always more enjoyable.

Should people be locked out of the ranking because they aren't that advanced yet? I would hope not... All I want is for people to be ranked in accordance to what they can play - where naturally most autoconverts would have less of a difficulty rating, which would translate into a fair amount of pp.

Anyway, I hope no one took what I said as an insult to the mapping community - you all do fabulous work I'm sure, but I'm not skilled enough yet to be able to enjoy it very much.

As a side note, this discussion has prompted me to give mania specific maps another try and I'm pleased to report that I'm now able to complete the "easy" maps. Though I don't find them very easy at all lol...

MandyJS wrote:

Should people be locked out of the ranking because they aren't that advanced yet?

uh, yes?

do you have any idea what's a competitive ranking

if you're having trouble playing [Easy] diffs then your rank should be the least of your concerns

Kanachann wrote:

Tom94 wrote:

For mods the map will simply be transformed accordingly to what standard does in the moment. HR will increase OD and AR, DT the speed at which the map plays back, etc.

So mods are going to come in and affect your pp?

Just the map, not pp i guess?...
And Tom also mentioned that HR will not change AR in mania previously.

https://osu.ppy.sh/forum/t/147552

This may be of some help regarding difficulty.

IMO there are 3 main areas of difficulty in osu!m:
1. Note Density (As stated previously by many others) - Both density of notes for burst sections and overall NPS (Notes per Second) should be one of the main areas to judge difficulty
2. Holds/Long Notes - Holds have their own set of difficulty and really test finger independence. While it may not be as dense key-count-wise as a burst of dense notes it adds its own complexity via hammerheads, inverted holds, etc (More in Drace's thread that was posted earlier https://osu.ppy.sh/forum/t/146615).
3. Speed Value(SV) Changes - Some files may not be as dense nor as complex hold-wise, but can trip people up with SV changes. This requires a lot of adaption with reading skills and can be really difficult at times.

Uhh... Does autoconvert counts with the pp for mania? If not i kinda disagree.. Many players rely on autoconverted maps as they could just score better than the usual mania map. If the autoconvert does not count in the pp then i assumr you really need to upgrade the community for mania. Modder/BATs etc. Because with such lack of people who are doing mania maps how can we rank if autoconvert doesnt count as the pp. Either way if you can only gain pp in a mania map only. I think it will be very hard fot some people that doesnt start to play mania yet. And thus the pp will work very slow if you cannot gain pp on autoconverted maps.

Addition: we really need to work up on judgement rate for mania

Aqo wrote:

MandyJS wrote:

Should people be locked out of the ranking because they aren't that advanced yet?

uh, yes?

do you have any idea what's a competitive ranking

if you're having trouble playing [Easy] diffs then your rank should be the least of your concerns

it may sound elitist

but yes, what are you doing in competitions if you can't even do the basic stuff? Do you really go to a robotics competition without knowing how to build your own robot and expect to compete with your basic electronics knowledge? Or do you try to compare what you know with doctors when you only have high-school level biology?

Xcrypt wrote:

Also Agka I think you are right but I also think you are a troll haha. Maybe you can share how the insane scale BMS or o2pf system works?

o2pf works a bit like this.

There's different skill levels- High class, Demon, Monster, Lost Humanity (off the top of my head)

To move from one skill level to the next you complete a set of challenges and move onto it. Some of the challenges include being able to one-on-one someone of that skill level and beat him/her, as well as the basic "pass this map under certain requirements" or so.

You could easily adopt something similar while accounting to the different dimensions of skill that mania play offers.

Overall, accuracy, LN, etc..

Agka wrote:

it may sound elitist

but yes, what are you doing in competitions if you can't even do the basic stuff? Do you really go to a robotics competition without knowing how to build your own robot and expect to compete with your basic electronics knowledge? Or do you try to compare what you know with doctors when you only have high-school level biology?

Forgive me if I'm wrong, but I thought ranked competitions were made to rank ALL players according to their skill. If I'm the worst mania player in existence, then give me the very lowest rank. That way I'll know where I stand and can celebrate small increases in rank as I become a better player.

To say lesser skilled people are not even worthy of being ranked isn't in the spirit of competition, from my point of view anyway.

Just a few comments from someone relatively new to mania (my ranking used to be around 880).

I agree with all the previous comments saying that auto-converts should be weighed a LOT less than mania specific maps. No matter how hard the song is in osu mode (lets say FOUR DIMENSIONS Freedom Dive, or 0108 Scarlet Rose, or even Madness Precipitation for 狂気ちんでん), auto-converts never reach surpass a certain level of difficulty. Unlike most MX or SHD maps (or even HD maps), auto-converts never include (from my experience, not necessarily accurate, sorry if I'm using personal terminology, I'll try my best to be clear, I might be editing this list):

1. Holding long notes before the previous long note is finished (nearly every HD+ map includes holding long notes halfway into another long note, or letting go of one long note before the other one is finished, if you really need an example, fripSide - future gazer http://osu.ppy.sh/s/115715)
2. "Burst" maps, even though not necessarily a good determinant of skill, but still a good reference (i.e: KOTOKO - unfinished http://osu.ppy.sh/s/102519)
3. Melodic sweeps (I'm not saying the occasional 5 note sweep, I mean shit like MEPHISTO MX) and sweeping holds (bj. Halo - Ende SHD http://osu.ppy.sh/s/79132, or MEPHISTO SHD)
4. Actual beatmaps, not rhythmmaps (Silent Siren - Stella http://osu.ppy.sh/s/81166)

I'm perfectly aware that 4k can definitely get as hard as 7k, but I also believe that 7k should be the standard for mania (or else why would the system be limited to hitting a maximum of 6 notes simultaneously). Anything out of 7k should be left just for recreation purposes, and should be weighed less in comparison to 7k. Accuracy shouldn't be weighed as heavily as in other modes, as it would just encourage people to play auto-convert and easy maps. You could consider counting the MAX to 300 ratio, as I think that would be a rather accurate way of determining one's skill. You're making a brilliant move in tallying all scores (not only the top 50), as all mode-specific maps are a hell lot harder to go into top 50. All mods should be left the way they are right now.

I'll try to add to this whenever I have time, quite busy right now.

MandyJS wrote:

To say lesser skilled people are not even worthy of being ranked isn't in the spirit of competition, from my point of view anyway.

I'm fine with this, but then why would you weigh autoconverts the same skilled people play?

that's the point

Agka wrote:

I'm fine with this, but then why would you weigh autoconverts the same skilled people play?

that's the point

I'm sorry you've misunderstood me. I've always been stating that since autoconverts aren't as challenging as the mania specific maps that they should be weighted accordingly for pp.

Here's my previously posted thoughts on the subject so you don't have to do searching...

MandyJS wrote:

I'm sorry you've misunderstood me. I've always been stating that since autoconverts aren't as challenging as the mania specific maps that they should be weighted accordingly for pp.

yeah, as I said as long as autoconverts are weighted very, very low, almost negligible (as I harshly said before)

Please, autoconverts should NOT be included in the ranking formula.

I'm sure this was said, but autoconverts really kill the rankings. It's a little bs seeing myself being beaten by ranked 20k people and then I beat a rank 20 something.

EDIT: Or rank autoconverts difficulty accordingly to the mania star difficulty. For example, people are complaining that a 5 star autoconvert is as hard as a one-two star o!m specific map. When changing into mania mode, they should adjust the star difficulty accordingly, this way they'll have less weighting.

I think that autoconverts should be ranked too, but as stated before, they should be lowly weighted to the ranking system. Since you should play mania-spec maps to get a better rank I think that also would be a good motivation for some players that actually want to be good at the game and not just have a high rank

Letting ACs be ranked at all (no matter how much they weight, 0 is only proper value) should make hitsounding and correct timing of objects the ONLY ranking criteria for osu!mania, I see no logical reason why a bot can make something superbad that is accountable and a human can't. So either remove them from new rankings or change o!mania demands for chart being rankable.

first of all, i'm sorry my english to sucks (trying to makes everyone understand of this post)

maybe is it just me or i don't wanna seeing pp system ?

why i rejecting pp system ? the reason is simple : farmer player will try gaining pp as he / she can, and the result, rank owns anything right ? (from outside view)

and the idea of only gaining pp by playing mania-specific-beatmap is still not good, if you cant play hardest diff, just try the easiest diff, get SS, and get so many pp, done.

and i'm not gonna say anything about auto-convert beatmaps, since some auto-convert beatmaps for me is fun (maybe just for listening song ?), and when i arrive to this game we're only get a very few mania-specific-beatmaps (and because of that, pro + ET players tend to play a lot of unranked, pending beatmaps)

+ the idea of mods gaining more score is kinda terrible too... i'm not good almost at all mods and the acc in this game really damn hard (especially if you try FC / Perfect in 1 beatmap, but the pattern makes it kinda impossible to do, and now we're getting some mods = ++ score ? hahaha hell no dude...)

and sorry i know this maybe a little bit OOT, but what i'm trying to suggest is only this :
- trying to find new system ranking (some players find a ranking makes the game competitive right ? but old pp system kinda makes me lost there...)
- and the most of all, try lighten a bit of osu!mania mapping ranking rules, so in the future this mode have a lot of mania-specific-beatmap, lots of osumania players, and the most of it, mappers are enjoy to create them, and targeting to makes his / her beatmap to ranked...

that's all i got to complain and suggesting...

but autoconverts are so wrong it's like playing scatman on standard vs these new maps

Aqo wrote:

-Blossom- wrote:

First of all, there are a lot of converted maps pretty hard to pass.

uh, no?
I completely suck at mania and at this point can pass all the autoconverts in existence. (Some are hard to play though, extreme fantasy etc)
If you're having trouble passing autoconverts then... -_-

There's already 80 ranked mania mapsets, with usually 3-4 diffs per one. That's already plenty, and with the increased amount of mania BATs the ranking speed of new mania charts has drastically gone up. Autoconverts are no longer needed.

Completely judging a map set on your own skills doesn't determine the actual difficulty for the general mania players. Someone's version of 'sucking' at a game can be considered as somewhat advanced in another's eyes.
I agree that mania maps are harder, yes... but just rank them properly in accordance to the real difficulty. Why is it so hard just to increase the stars they were given? There is no right in taking out autoconverts completely. Some may prefer the 'fun' maps that have been made for the osu!standard mode.
As for the original keys set for osu mania (such as sdf etc) people change the keys to how they prefer to play. Some may play better with the original keys set, while others prefer a custom key set. Decreasing a score based on what key sets people use allows no common sense in this matter.
Also, calling players who play the osu!standard converted maps 'not real mania players' is utter and complete bs. Play 0108 converted maps and tell me if that's 'too easy'.
Ganbatte on the new system tom. Hope everything works out well for the ranking system in development.

in my opinion

why not big 500 got a few pp such 1-10 pp(1-2 for easy,3-5 normal,5-7 hard and 7-10 for insane)

but 1st place got a bonus pp 20 and big 50 got bonus 10 pp and big 100 got 3 pp


it should be easy for make an opinion like this :3

but i'm afraid if it will make people confuse XD

Pearl Phoenix wrote:

As for the original keys set for osu mania (such as sdf etc) people change the keys to how they prefer to play. Some may play better with the original keys set, while others prefer a custom key set. Decreasing a score based on what key sets people use allows no common sense in this matter.

Unless I'm mistaken, people weren't talking about the actual keys uses to play the game, but rather the key mods. Like if the map is originally 7 key, but someone plays it using the 4 key mod, the score would be decreased.

MandyJS wrote:

Pearl Phoenix wrote:

As for the original keys set for osu mania (such as sdf etc) people change the keys to how they prefer to play. Some may play better with the original keys set, while others prefer a custom key set. Decreasing a score based on what key sets people use allows no common sense in this matter.

Unless I'm mistaken, people weren't talking about the actual keys uses to play the game, but rather the key mods. Like if the map is originally 7 key, but someone plays it using the 4 key mod, the score would be decreased.

I was replying to Tom's idea.

Forgot to mention that anyone who thinks the converted maps are all 'easy' should go SS all the hard's and insanes.

Pearl Phoenix wrote:

Play 0108 converted maps and tell me if that's 'too easy'.

They are not easy actually, they are quite hard, but for the wrong reasons

Also, please we should stop worrying about autoconverts for now. If there's a system that applies weights to the difficulity of the maps, autoconverts become worthless for everyone except new players. Help Tom with what he needs help with instead, that is determining the difficulty of the maps, and it's not that easy to do.

If this is to be done right, the auto-converts shouldn't be ranked at all.

First of all, those autoconverts wouldn't pass the ranking process if they were submitted like mania maps. Many maps gots notes mapped to stuff that mania maps normally wouldn't. The long note's release are not correctly timed. And no matter the map, you'll never have something layering multiple tracks which is pretty much the essence of multi key games like this. These maps were converted from a 1k source and the patterns processed by a semi-controlled randomization.

With autoconverts, speed is the ONLY difficulty factor. You will never find difficult patterns. "Difficult patterns" are patterns that would still be difficult when slowed down because of how the notes are placed. I would agree to a certain extent with the people that mentioned that the people that play autoconverts exclusively aren't true mania players. The whole difficulty ladder in all similar rhythm games is learning and recognizing different patterns after extended play, even the standard mode. But the maps converted from there are stripped from the clever patterns they once had and become a useless junk with single-dimension difficulty.

Imagine converting a mania map to standard, and having circles timed with the notes (or chords) with randomized placement, yeah. Someone who never played standard would think "what's wrong with that?". But I'm pretty sure every standard player who read that just cringed. That's how it is for mania ACs.

There is no reason at all these should be ranked. They can be there for a larger song library, but there is no sense in having stuff that doesn't accurately determine a player's skill play part in the ranking system. We got mania maps for that and that's all we need.

To everyone who doesn't understand what I'm saying look at this. http://osu.ppy.sh/forum/t/146615 and scroll down to the pattern repository. Now try and find autos that feature anything but beginner patterns. All those "omg so hard" autos are just fast beginner and maybe novice level patterns. Keep in mind that this is a just a fraction of "common" patterns. Harder maps all feature unique stuff to make them memorable.

(I apologize for the the self-promotion, but I honestly don't know of a better repository.)

----------------------------------------------------------------------------------------------------------------------

Enough saying what already known, I'm gonna introduce a couple other points here.

People are saying that maybe there isn't enough mania maps to run the ranking system. Well truth is there's more than enough. There is currently more ranked mania than any DJmax games, any IIDX machine, (I'm assuming that goes for DDR too), than any Rock Band or Guitar Hero games (excluding DLC) and mania's playerbase is smaller than any of those games. Having those 30'000+ (diffs) ranked autos was waaaaay too much anyways. People with no skill could accumulate PP by going no life and playing every single one of them.

And we could also think about the mappers, what motivation can they have to make maps when the majority of players are playing autos? Their map will just get lost and unappreciated. And what about modding? The playerbase playing autos have a bunch of songs to play and the idea of helping getting more maps ranked doesn't even cross their minds. I'm pretty sure unranking the autos will not only motivate mappers to try and rank more maps, but also spark interest within the mania playerbase to help as well. A well needed kick start.

I personally believe unranking the autos is the best possible move, lets forget those atrocities ever happened and start fresh.

-- About PPv2
It's nice and all that ppv2 gives huge priority to mania maps, but honestly it's still useless if easy/normal are still giving more pp than hard/insane simply because there's more ranked plays.

What if the pp system weighed the songs the total unique players ranked on all difficulties, and then weighing that again per diff on the mapset?

Regarding the big auto-convert discussion:
I'm fairly positive, that a good difficulty algorithm would rate auto-converts extremely low, because they (mostly) are so easy compared to mania charts, so you would naturally have them worth less, not artificially by excluding them. Keep in mind, that the ranking system is meant to work for everyone, even the people who just start with mania who can't just dive into the mania-specific maps, or want a big map diversity at their lower level of skill.

The current system does not look at how many ranked plays there are or anything. It is not done calculating yet, so please don't judge it yet. And even when it's done a way more precise version will follow in a few weeks when the difficulty algorithm is implemented. You will be able to give feedback on beatmap weightings earlier than that; I'll be posting some lists and your feedback will decide what to take.

I can't help but wonder, will pp still be delivered accordingly to how many other players who has actually played a certain map. Will there be any reward in passing the high difficulty osumania specific maps that has a low success rate?

Any answer is appreciated. Thanks.

Edit: I really appreciate the time and effort you're putting into this Tom.

Just to clarify, by unranking I don't mean removing their ranking boards, I mean making them not count towards PP.

If you're new to the point that the easiest mania map is too hard for you, what good would a rank be anyways? Do you want to shove the fact that they aren't good in their face by displaying that 2mil+rank?

And It's completely impossible to make an accurate algorithm for difficulties, it's best to limit the pool of possibilities to what we know is good material instead of depending on an algorithm to tell it for us. AKA only mania-specific maps.

As far as I know, ppv2 for mania will function like osu!tp - where your best scores determine your rank, making it impossible to farm at a higher level. If anything, this will encourage people to play the more difficult mania maps instead of ac.

Also, thanks for your work Tom

handyg11 wrote:

why not big 500 got a few pp such 1-10 pp(1-2 for easy,3-5 normal,5-7 hard and 7-10 for insane)

but 1st place got a bonus pp 20 and big 50 got bonus 10 pp and big 100 got 3 pp

That way you get the same points with 3 top 500 and with a top 1. I think higher scores should be rewarded much better.

That said, I think that only mania-specific maps should be ranked.

Cozzzy wrote:

As far as I know, ppv2 for mania will function like osu!tp - where your best scores determine your rank, making it impossible to farm at a higher level. If anything, this will encourage people to play the more difficult mania maps instead of ac.

Also, thanks for your work Tom

This is how plan on doing it. Only your score and the map are relevant for your score - other players and their scores are not. Doing statistical analysis on 2 different maps usually is not working well, because not the same players play the same maps.

For judging difficulty, I'd say we should do this right and introduce a new BAT-like group of expert players to give maps a starcount of a custom difficulty scale. I personally know a fair amount of active ETs that would gladly volunteer.

Though of course this would never work if the autos still count for PP.......................