User avatar
osu!team
4,482 posts
Offline
I am pretty sure that multiple rankings and the debate, which attribute is most important aren't really an issue, but rather an enhancement to the game, as it gives people incentive to specialize in different things and/or compete in different skill assets globally.

If I would be able to design freely and come up with spontaneous ideas, I would replace or rework the current Player Level system, making the Player Level raise for each user accumulating ranks/points in the various skill assets. Therefore, those with higher levels are more "vesatile", as the points for levels you get from one skill asset eventually soft-caps out, while starting to train a different skill asset will start giving you more points to next level easier. I dunno, something like that. Instead of ranking people then, I would just give people levels. I don't think it needs a global overall ranking, because the skill in this game can't possibly be boiled down like that so simply. Players could choose which rank to display and maybe the display could even alter based on your specialization (sort of like osuskills.tk awards you different "titles" based on your specialization combination, I really liked that idea)

There are a lot of ways to make this interesting and make it account for the multitude of skill assets. I proposed something like this internally a few times already, but it was considered too "complex", which still saddens me, to be honest.

All in all, the pp-system is based around a biased view of what defines "being good" and also weightens it around the biased idea of what is "worth more" and "worth less". This entire balancing shenanigans can also be circumvented with splitting it simply into different rankings.

It will also cause mappers to specialize into mapping things shipping towards the specific rankings, increasing diversity of mapping just by that. Mapping generic patterns to yield the most reward out of the least effort will still continue to exist, but it will certainly create more different target audiences for beatmaps, as not everyone is competing on the same leaderboard that only aims at the three same skill assets always. And the sole creation of new target audiences creates demands, which eventually will lead into supply.
Last edited by Loctav on , edited 2 times in total.
Image
User avatar
Rhythm Incarnate
2,908 posts
Offline
Well, over the past years most of the work behind the curtain has gone towards "Simplifying things". Look where that's gotten the game - most active players/modders are disheartened, bored, and demotivated because something so simple is quickly figured out and inherently boring.

It would be really good to press to intentionally loosen up on simplifications and allow for more complexity and just let the community sort it out on their own. Like Loctav said, there will always be a large chunk of the game that leans towards simplicity - but a community is not fueled by the lurkers or casual players, it is fueled by content creators and active players. Too much has gone towards making the game more accessible to the masses, to the point where the specializers are actively hampered.

Talking about it is nice and all, but there's no possible way to actually DO anything without convincing people with coding capabilities (and admin access) to work things out - ideas are a dime a dozen, implementation is gold.
If you can see this it means I'm posting outside of the mapping forums and you should take anything I said with heaps of salt, or as purely intended for humor.
User avatar
Beatmap Nominator
628 posts
Offline
Oh boy, my favorite mapping related topic. This type of discussion briefly appeared in the upheaval thread, some of what I said has been commented on already but I want to expand on it. I already see a lot of discussion about the ranking process, etc. so i'll try to add some newer concepts to think about.

General motivation

Quote:
I think the only people in the end who decide the amount of innovation and creativity that go into the ranking system are the mappers themselves. A map that uses perfectly meta aesthetics, comfortable flow, very safe style, etc, is obviously much easier to rank than something that pushes rc boundaries. So if mappers feel more comfortable mapping in that way from experience then they'll do so, and push those for rank. The thing that will most stop them from pushing experimental types of maps is the boundaries that the ranking system provides- so it does have to do partly with the qualification system, but there can't ever be a perfect system (even though we can come as close as we can). There has to be some sort of quality assurance for ranked maps in the first place.


Rewards

Leading off of what i said above. If mappers are intrinsically motivated to adhere to mapping styles they're familiar with, I think that we need place a bigger focus on ways we motivate mappers to be different and take risks in their mapping. The spotlight works, sure, but I think at the moment people are looking at the spotlight to see what cool maps might be ranked, that's nice and all. But if we want map spotlights to act as an incentive to mapping creativity, we need it to leave more of an impact and it needs to be hyped up. I've seen mappers not even know that their map received spotlight, if we are going to use this as encouragement to create notable maps then that needs to change. Still, what does spotlight mean to the mapper? I would suggest some kind of icon on the mapset webpage as a reward, maybe something to add to the mappers profile also. (Probably not a badge or a title though because skilled mappers who can earn spotlight multiple times would clutter their profile with that.) Just realized that this was mentioned but expanding i guess.

With that being said, we should also expand on other rewards besides just spotlight. Spotlight can in theory act as a bigger, more prestigious reward, but something smaller that mappers can earn for creating something neat (but not exactly groundbreaking) can help encourage mappers with that little bit of recognition. Like an 'honorable mention" sort of thing. I don't know what that reward might be (probably something that recognizes that map specifically, not just supporter) but an add on to the spotlight for honorable mentions can work. I've seen a lot of skilled mappers with potential say they're quitting, one reason why I think a mapper might leave the mapping scene is that they haven't received enough attention from their maps, so it comes down to how passionate they actually are about mapping, they don't get enough recognition telling them that their maps are cool. To be fair though the ranking system is largely demotivating and can be pretty unforgiving for maps that attempt to push boundaries, which is already kinda adressed with the qualification system in this thread.

The loved section

fix loved lol

User rating can help?

On a separate note, one thing that I think should hold more weight on maps is user rating. It seems as if the most played beatmaps get more attention than those with great user feedback get a lot more attention. I think this encourages pp mapping in a way because players seeing a farm map = lots of retries = more plays on map = front page of website & reddit for big pp plays (and yeah a new pp system as previously mentioned can add to this). With that being said, the amount of exposure a map gets is affected by, and sometimes reliant on, how much pp a map gives. We all know this in some shape or form. I would honestly just replace the most played section on the home page with something that factors in user rating - the only thing here though is that it's easier to give a map a good rating than it is to repetitively play it, and there's bias toward mappers we already know and like, so I'm just suggesting more of a balance in between the two. Thinking about this, CCBC hot like 250k plays in the last 24 hours but ended up with a rating of 6.5 lmao.

That being said, this type of system might just give exposure to creative maps rather than create more variety overall. But I still think it can make that "most played" panel more valuable and give mappers another thing to aim for.

BN flexibility

I saw a little bit of talk about this and think its a great step forward toward diversity. The only problem I see with this is - how? You can say that bns are allowed to change their perspective and take a little more risk but will they? The only problem is that this isn't something that you can enforce, but mostly just encourage. idk anything too specific about the future of the bng, subdivisions, etc. but if we were to penalize a subdivision for nominating a mapset that got a pop or dq, that would discourage taking risks bubbling something that's non-meta conforming. Idk what much else to say about this, but I'd like to see bng structure and map diversity more aligned, if that makes sense.


I think that's it, sorry if its wordy. :o
~SP~
User avatar
osu! Alumni
3,217 posts
Offline
What Probox said fully resumes the core problem that limits mappers while mapping. We can't and we won't be able to do anything with PP ranking system. Even if the PP system changed the mappers would anyways adapt again to the new system so that their maps will be mostly played due to the effects it has on the player's ranking. If we want to discuss about the nowdays mapping styles and ranked beatmaps we don't have to consider the PP system tbh and we should only focus our selves to find a different solution that doesn't connect the mapping with PP system. When the score system was still above the PP ranking we had bunch of different mapping styles ranked (symmetric maps ex. Andrea, maps with linear patterns ex. djpop, unconsistent spacing but flowing maps ex. siriru, and etc.) while now we only have jumping PP maps and technical ones (that can be considered anyways a style to implement with the ones I mentioned above). Only way I think it may work is to IMPLEMENT the PP system in the mapping and not IMPLEMENT the maps in the PP system, it's a different concept: the ranking should adapt to the map and not the opposite. Well, I will now propose some of the ideas I got with reading your ideas as well. All my ideas are referred to the concept of beatmap category I am gonna to explain.

Beatmaps Category
I would add multiple categories to the qualified/ranked beatmaps in which every beatmaps with specific features and elements that reflect the specific category will be placed at. Categories like: Jumping maps, Techinical maps, Symmetric maps, Flowing maps, Innovative maps, Constant rhythmically maps, etc. where maps can be placed when qualified/ranked and this would push the mappers use different styles on different songs. Of course this would mean that the "PP system has to be adapted to the mapping and not the opposite": hardest maps of that specific category will give more points and PPs. Yea I am still talking about PP but this time it has been adapted to the style of the specific map and not the opposite. Mappers would map more different styles instead of the same annoying PP ones. Well, how can BNs and QATs decide in which category the future ranked map should be placed? This depends by the mapper, modders and BNs/QATs them selves. Once the mapset is finished and the mapper wants to push it on pending beatmaps he should specify in which category he wants the map goes so when the BN/QATs and modders will mod the beatmaps can even vote if they agree with the mappers or not. Of course a beatmap can be mapped with different styles so when I try to search a map to play I can select multiple categories.

Specific BNs
There should be a revisiting on selecting specific BNs for specific Categories. This means that BNs can be able to nominate/qualify different styled maps because they are able to mod and judge those specific categories. This would ofc mean that the team has to be enlarged so that we have a good amount of BNs for each categories (QATs should be able to mod/qualify all kind of categories).

Testplayers Group
Group with selected skilled players can be created to verify if that kind of hard map with that specific category can be played properly and if the map itself respects the style of the category the BN/QAT is gonna to select while nominating/ranking the beatmaps.

Well.. We can't negate that the standard mapping is the hardest one because compared to osu!mania, osu!taiko and osu!ctb every mapper can map as he wants on a full editor grid. It's like impossible to adapt the ranking system to the mapping styles (let's be honest) but the ideas I mentioned are focused especially on that and not on how to revisit RC to give different styles the chance to be ranked and to be played with same intensity of PP maps. I am aware that these ideas are kinda pushed because it would mean some big changes on PP system, BN/QAT system and the system of beatmaps in different categories. But well, despite this I wanted to mention them and nothing. Good topic tho.
User avatar
osu! Alumni
3,464 posts
Offline
I posted this in 2015 and it was promptly rejected, but seeing some of the discussion here I guess it's probably relevant again so I'm linking it again: t/371530

The gist of it is basically to add another state for maps, which would be a "frozen" state for when discussion takes place. The map won't advance in the qualification progress, but it also won't lose any progress until it is properly DQ'd. Only if the map is disqualified due to a significant change does the qualification timer reset, as then the change is big enough that you should have more people look at it again. If the change is minor, or it's a non-issue, the map gets fixed, and unfrozen to continue its journey through qualified.



I'd also like to add that spotlights and whatnot are nice, but they really need to be integrated into the game client and exposed to players. Even just having, say, "most played maps of the week" or "highest user ratings of the week" categories with 5-10 maps showing up in the game client could help promote at least some diversity.
Wafuu~/Boats
ImageImage


Image
~Mod one of these and get a cookie!~
Mod queue.
Feel free to poke me in-game if you have no idea what I'm talking about in my mod.
User avatar
Rhythm Incarnate
546 posts
Offline
I personally don't think mappers need incentive to map creatively, and I certainly don't think there are ONLY pp maps being made today.. The issue is not lack of mapper creativity, the issue is lack of support for creative/different maps by Staff/Beatmap Nominators. Most BN's are too afraid to rank anything that is something other than anime jump pp maps, which causes there to in turn be nothing but anime jump pp maps...

Don't believe me? Look at this literal quote from a BN on his userpage:
Naotoshi wrote:
most of my bubbles are on fairly bad maps, or rather fairly uninteresting ones. i saw it as necessary, this game needs new content, and without bns nominating rankable-but-not-amazing stuff there just arent enough good mappers out there to keep the game alive.


I think many BN share this same sentiment. The ranking criteria and the pressure of the system not to "rank anything you can't judge" has caused such a stagnation in quality, unique maps being ranked.

If you want to fix it, loosen the ranking criteria and let BN run free ranking whatever content they deem fitting. BN should never feel AFRAID to push a map because of potential backlash from staff/other BN just because it's different than a 1-2 anime jump map. It's so frustrating as both a player and mapper to hear people say that there are no unique maps being made today when it's clearly not true, one look at mappers like dsco or handsome show just how wrong that sentiment is.

Question about quality, would you rather have a map like Miss You, where 50% of people think its absolutely genius and 50% of people loathe it, or a random 5 star anime map that everyone thinks is meh/average ranked?

The current system, from the 24 hour icon bubble popping rule, to the strict regulations on BN, to the lack of quality assurance that comes from having a Quality Assurance Team is what's causing this influx of pp mapping, not mappers "not having the incentive to map creatively". You need to stop thinking of mappers as children that need to be goaded into being creative, and instead put your power and make a difference in the maps being ranked.

It's not necessary to change the pp system to fix mapping, just qualify better maps... Better maps than pp maps exist and are waiting without a willing nominator for months/don't even go for rank because the ranking process would prove literally impossible for them. You as the staff have the power to set the precedent for what can and can't get ranked, so just do it instead of wondering aloud "hmmmm how can we make mappers map creatively again??"

The answer isn't even stopping the ranking of pp maps: if you rank more different/unique maps, more people will in turn want to make different/unique maps because they'll have the feeling that they'll actually be able to rank something that's not the same anime jump map or generic wub mapping style map
Image
User avatar
Elite Mapper
3,070 posts
Offline
agree with kaifin except the "don't rank things you can't play" part because that's a rule that's in place for a reason and is usually forgone anyways by people who are experienced enough. it just keeps people at 80k ranking a 8* on their 2nd month in the bng or something. extreme example but the rule is fine (also not what this thread is about)
Image Image
User avatar
Rhythm Incarnate
3,155 posts
Offline
Hollow Wings wrote:
2. Perception of difficult maps in cyber vision.
pp system failed in recognizing hard ones, but did well to get rid of easier ones from them, in case of the rank rules we built long times ago: we forbidden proper things in lower mapsets. so we can just focus on harder ones.
How we can call a map is difficult to play? We used to trust the pp system and think now we don't. i always wonder why we call a map's difficulty "Extra" when we have all dthd scores on the ranking board, shouldn't it be "Insane" level if so much people can do that? I'm not talking about the use of classification as diff names done, but from the automatic processing way machines do to give ranks. And as we known computers can hardly see if some map's reading is tricky or tough even from technical way, it's much better to judge it directly from the score board: less people performed good and then that map is surely difficult. Some people may say that it's more like a ppv1 version system, but actually and slightly not. i cant image a much better way to solve this problem for now, and it can be done right now imo. (even i don't now the specific detail way in programming, i know it can be figured out by statistics.)

- This new system will take more power in judging performance of players if the map is a "difficult" one, and do the opposite way if it's easy, let current pp system do the rest of work, it actually still did it well.
Just going to say HW hit the nail on its head on this one. The skill chart Loctav suggest is going to work on everything but reading. Essentially you can do aim, speed, precision, reaction, which are essentially the same things that influences players to play hard maps now, minus precision (small CS). I would love to hope that this will someday be calculated using formula, but upon realization that the future will likely involve new gamemodes that players would want to be ranked, and that those gamemodes may have even more complex mechanics, this will become as unrealistic as you can image. I figure the best way to approach this is to interpret maps as a black box and do statistics on them like HW mentions.
User avatar
Star Shooter
196 posts
Offline
(most important part of this post after the ~~~~. no tl;dr, read all of it, if any. most of the time i spent on this post was spent shortening it. everything below is a miraculous work of artistic brevity compared to the incomprehensible stream-of-consciousness babble it was previously. if you're really impatient just skip down to the bolded parts)

Slightly less important part first, quick and simple: stop pursuing flawlessness.
Creative maps are harder to rank because ranking criteria doesn't value merit, it values flawlessness. They are different things.
As a result, many ranked songs are boring but "flawless" maps.
Proliferation of flawless meritless mapping in ranking section influences community to continue this trend. It's an unhealthy cycle.

Do you want more diverse mapping?
Start ranking more diverse maps.
That means to value flawlessness less, and value merit more.
The mapping community, BNs, and QAT shouldn't be so scared of pushing creative maps even if they have some flaws.

If the ranked section becomes more diversified, the community will follow suit.
Yes, the community generally likes to shit on creative maps, just because they're different from what they know.
But if BNs and QAT keep ranking them regardless, then tastes will change for the better.

So stop being such perfectionists, and have an open mind. That's all. It's too easy to be a perfectionist when you're judging someone else's map as a modder or QAT or BN or whatever, so just stop being a perfectionist and stop being so protective of the ranked section, it stifles the state of mapping. My standards are higher than anyone's, but I'd much much rather see some shitty maps in ranked section than the homogenized maps that fill it today.


~~~~



ppv2 incentivizes homogenized mapping. ppv2 is THE biggest problem, I don't need to tell you why (but i will anyway lol). ppv2 takes the most boring mapping and places it at the top of a subconcious and completely arbitrary heirarchy. ppv2 is pretty much the sole reason mapping is what it is right now. Want to get your map played often? Put in big jumps in a simple-to-read map. Want to map a memey song from sesame street, but also get it lots of attention? Put in big jumps in a simple-to-read map. Just an innocent mapper who likes to map in a style that happens to have jumps? Now everyone will heap attention and plays on you, which tells you and the rest of the community that that certain style is the only way to go.

ppv2 takes a single hyper-simplified style of mapping and makes it the focal point, the bread-and-butter pattern of the entire game. It's much bigger than just a pattern you see in most maps. It's a paradigm that the whole game is measured by, which is stupid. Even jumps themselves, a perfectly respectable pattern, are now vilified precisely because of the pp system. ppv2 as a system has taken the spirit of competition away from the players themselves and has injected them into the maps. The spirit of competition is good for a game, but is unspeakably cancerous to creative endeavours such as mapping. Instead of competing with each other for ranks on the map leaderboard, they are scrutinizing the maps themselves. If they lose or gain pp, it's no longer because they are better or worse than other players, it's because of the map and its patterns and its resulting difficulty-to-pp ratio. If you want to fix things, shift the spirit of competition away from mappers and back towards players. That kind of competitive milieu doesn't belong to mapping, not if you want mapping to diversify and flourish instead of stagnate.

I know I go on and on about ppv2. People know it's a problem, I just wanted to drive home how much of a problem it is. It's very easy to heap blame on mappers or QAT or BN or any other people, but it's not the people, it's the system. ppv2 as a system is the root cause; everything else is a symptom. PP mapping, PP mappers, community recoil from unfamiliarity -- all are not the problem. ppv2 is. And that's not to say that ppv2 is worthless and tom94 has destroyed mapping; it's been through some rough spots but ppv2/star system is valuable as a very rough approximation of difficulty.

Here's my personal suggestion; calculate skill with leaderboards again. You'll never be able to accurately put difficulty into an equation, it's too subjective and complex, so stop trying. Calculate skill by what rank you get on the leaderboards of a particular map. That will get people looking for different, more interesting maps, it'll create a more diversified "pp canon" which will in turn result in more interesting mapping purely through exposure to different styles. I wasn't around for ppv1, but from what I hear this is similar to that. Of course, the problem with ppv1 (from what I hear) is that because there was no quantified measure of difficulty, a hard diff from osuplayer111 would give more pp than a nice play on freedom dive, for instance. So here is what I am proposing:

0-2 star maps = tier 1 (pp value:1)
2-3 star maps = tier 2 (pp value:3)
3-5 star maps = tier 3 (pp value:7)
5+ star maps = tier 4 (pp value: 20)

PP of map = (100.00 minus rank percentile on map) x (map tier pp value)
Values and star rating can be adjusted and ppv2 values can be fit in somewhere, this is just so you get the general idea.


Reasoning for tiers: I grouped everything above 5 into one because I don't want someone bumping a 5 star map to be a 6 star map just to give it pp. In my experience, people play 5/5.5 star maps to 7 star maps at about the same frequency. I don't play 4 star maps anymore (except ones I REALLY like) but I'll still play 5, 5.5 star maps if they're interesting, and I do all the time, even though most of them are easy for me. The most apparent issue is Apparition being in the same tier as Hitorigoto, but then again, Notch Hell is 5 stars too. And getting top 10 on Apparition and top 10 on Hitorigoto both require significant skill. In this way, the system starts to pit players against each other instead of against the map.

If a map requires little skill to FC, then people will flock to that map to compete for pp, and the value of an FC on that map will decrease. Using mods will obviously remain advantageous (I personally would advocate removing FL from pp, because pure memorization to me is an entirely different skill from every other mod, but that's a debate for another time), but nomod won't be irrelevant either as long as the map is hard enough. In this way, via competition (this is actually a pretty capitalistic system that's based on competition and incentives, rather than forced and rigid regulation), values of scores will be equalized based on their difficulty decided not primarily by an inaccurate equation, but by weighing the performance of a player against other players.

Player scrutiny moves away from the map, which is good; that "spirit of competition" is eliminated from the mapping process, because no matter what mappers throw out there it is against each other that they have to compete. If they lose PP, it's not because they didn't fc because "the map is dumb and plays bad," it's because someone else did better than they did. They don't have to FC to get PP, they just have to do better than everyone else.

Maps below 5 stars can still be rigidly tiered because if you're mapping less than 5 stars, you're not mapping for pp anyway.

This should also mitigate the problems created by ppv2 by de-emphasizing jumps. It will take some time to have a long-lasting effect on the public consciousness, which by now is shaped extremely heavily by the ppv2 paradigm, but it will happen.

One problem I can see arising from this is the enormous amount ranked maps available. Because it's so hard to find unknown or old maps, plays on those maps will be enormously inflated. A solution to this would of course be to develop a map sorting system that can sort from lowest # of scores to highest # of scores. The date ranked should be easily visible as well, just so players know what era of mapping they're getting into. I think in general it should be easier to explore the ranked section, there's a lot of hidden gems out there that I have no idea how to get to. There might also arise a fun pp culture that involves just exploring the ranked section and looking for undiscovered maps to play. There are too many maps rotting in obscurity in both the ranked and graveyarded sections.



In my opinion, anything short of completely revamping the pp system can only partially ameliorate the negative effects ppv2 has had. Piling more extraneous features and categories like loved won't do much to fix the current situation, it'll all just be ineffectual flailing. But if a ppv3 is too drastic, then just stop being a perfectionist about the ranking system and things will be at least somewhat better.

edit: better spotlighting is also a good and less iconoclastic idea, handsome has some nice suggestions.
User avatar
Beatmap Nominator
1,406 posts
Offline
i think kaifin completely missed the point of my userpage quote. and that's compounded by the fact that i also nominated 2 of those dsco maps lol..

in any case: if you don't think pp causes this issue, go look on top played maps. they are ALL pp maps except for the occasional pop song that appears for a week then falls down.
as long as i'm getting mod replies like http://puu.sh/xyHBA/111a402cd5.png then it's pretty clear what most mappers map for. i mean, people want to be popular, and the easiest way to do that atm is to make a map that completely abuses pp algorithm and gets you tons of plays. yeah you get faded out and irrelevant after a week or a month or whatever but if you make enough of them (doormat, taeyang, lami. slayed, etc etc) you end up being well known regardless.
as a result, i agree that the main problem is the pp system not incentivising interesting maps and as a result the vast majority of players simply ignore anything that isn't 1/2 spam with a diffspike for that sweet peppy points.
aside from that, it's much easier to make a decent anime map (i mean cmon look at all those harumachis and sweets weekends and whatnot lol) than something actually hard, so naturally it's harder to get it ranked. i don't think that's an issue of bns being scared or whatever (we dont get penalized for bubble pops anyway) its just cuz the maps are held to a different standard than generic anime map #29583.
User avatar
Combo Commander
405 posts
Offline
PP does cause the problem but not if you don't let it :/ Sure most players will only want and appreciate pp maps but if you really want to push interesting stuff, there's no need to look at the players too much. There's tons of mappers who prioritise good mapping over wanting to be famous for shitty generic farm maps.
Hi
User avatar
Elite Mapper: Aspirant
602 posts
Offline
I must say I'm a big fan of having pp tied to map leaderboards as mentioned earlier in this thread. I tried coming up with a working system but there are several problems with pp being tied to leaderboards as listed below. Ultimately I'm optimistic about the possibilities of this system though and feedback on this would be much appreciated as this could be made into something for real. Now for the problems I mentioned earlier that would have to be solved.

Biggest problems to solve with this proposed system:


1. Maps that aren't played giving inflated pp.Basically super old/unpopular maps with little to none plays on them would give more pp than they should simply because of nobody having played them. Solution to this is simple though; a formula that takes into addition total playcount on the map and compares the total playcount to amount of full combo plays with or without specific mods. In case of no full combo plays on the map, the highest acc A or B plays etc. all the way to amount of nomod passes or if those don't exist, best passes with ez/ht/nf mods. Problem with pp being tied to leaderboard + map playcount would be the fact that one could essentially boost his own pp value by adding to a maps playcount on purpose, therefore raising the value of his score. This would have to be taken into addition somehow, perhaps by restricting map playcounts to only plays which pass the map or by having a max cap of playcount that a player can add to a maps pp (while still counting the plays on his userpage) or a combination of these two.

2. PP tied into leaderboards would have to be balanced around players' own playcount / amount of time to play which is not a good thing as the more you have time to play = the more time you have to farm leaderboards (Theoretically this is how it works nowdays but with this new system you would be rewarded for having more plays on unpopular maps as not everyone would have time to chase the scores on them). This is the solution I came up with: let X amount of time pass for all the maps to be properly contested and then the system would balance itself when all unpopular maps would be farmed to acceptable levels. Also limit pp calculations to persons 25-50 best plays (same number for everyone obviously). Basically the downside to this would be that in the beginning pp values would inevitably be inflated but the longer this system would be in place, the more it would balance itself out. Since the leaderboard system tracks all online plays made on a map, the most extreme scenario (this is impossible to occur but very important to see how this system would age) for this system to face would be a situation where all maps that are ranked are all full comboed with all mods that boost pp value. In this case the only thing differentiating the ranked maps would be their total amount of playcount as some maps have 70k ranked scores on them while others only have 3k. Therefore in the most extreme scenario this system would have to prioritize the 3k fc maps over the 70k because of their rarity but if this were to happen then the 3k fc maps would get more plays until they would be equal to the 70k playcount map since the 3k would give more pp. Overall the system balances itself out.

3.PP would have to be flexible in a sense that it can go up or it can go down. Up in the scenario of you making an impressive score, down in the scenario when your impressive score wouldn't be so impressive anymore (other people replicate it or make an even more impressive score on the same map). This is not exactly a big problem but adding this flexibility would certainly give mixed responses as losing pp is something that could never happen before (except in rare cases) and might be seen as a negative thing.


4. How to seperate an 8* rare score from a 5* rare score. This system would be balanced the closer one would be to the highest rank but balancing this in the 5k-500k region would mean that a map has to have a base pp value because otherwise ranking up would be impossible as your best plays are very impressive to you, but very common overall. So as mentioned the solution would have to be adding a base pp value to a map. This base value is the biggest issue here because calculating a base pp value perfectly taking everything into addition is not possible. Wherever this base value would be poorly balanced, this poor balancing would become the point of pp exploiting. Basically worst thing to do would be to tie this to a visible thing like current star rating system as testing out the weaknesses of it is easy and one can quickly tell how to exploit it. My suggestion would be to tie this base value to a new map difficulty rating system (essentially an upgraded star rating system which takes into addition slider velocities, low ar and such more accurately than currently) and this new system would never be revealed to the public to prevent blatant exploitation from developing. The old star rating system would stay unchanged and would still be the go-to option for players to evaluate how hard the map they are about to play is (It works well enough and we have already become accustomed to it although admittedly changing star rating system isn't that unheard of as it has happened before). Ultimately after time passes people would still probably figure out where this new system would be balanced around and try to exploit it but since the playcount + fc% is there to balance the system, noticing what the base pp is formed on would be really hard and should take a long while.

Those were the major problems I came up with, I'm sure there are more. Now to move on, why should we adapt this new system into place then if its so much work? I seperated the benefits into 3 categories below: Mapping/Mappers, Playing/Players & Overall development for the game.

What would this new pp system mean for mapping?


1. In theory this system should be "farming proof" (unless I overlooked something which I might have) as if someone were to make a pp map for farming, everyone would farm it and therefore it would lose pp value as a higher % of players scored a really good score on the map. Niche pp maps that are super hard to fc would obviously reward the players who are able to fc them and their unique plays would get more meaning which means that the harder the map, the more contested the best scores would be, up to a point where the maps get too hard for people to be physically able to play for todays standards(9*+ deathstream maps etc.).

2. Freedom in the sense that there is no point in chasing the optimal pp maps anymore because of them being balanced on playcount + fc%. The initial thought would be to try to make the hardest map possible which is still fcable but if the map is fcable then it would lose pp value when multiple people would be able to fc it hence rendering this idea useless and those who would be able to fc the map would ultimately deserve the pp it would offer.

3. Making fun maps to play would be encouraged as the lifespan of a map would be something like this: Fun to play but not a good pp source -> People who play for fun play it and enjoy it and because of them setting relatively rare scores on the map it would prove to be good pp (not why these people play the map) -> More people play the map because its fun as told by their friends OR because of it being rare it offers more pp than some other maps of similar difficulty do -> Even more plays -> Devaluated pp scores at the "cost" of many more people enjoying the map. Ultimately the end result is that a map made for fun was played by all types of players and gained more recognition and playcount than it would previously had = Mapping whatever is more viable with the new system = Mapping diversity.

What would this encourage the players to do?


1. This kind of system would encourage the players to intially seek for rare fcs on forgotten maps but as time passes this would become much less optimal. Having forgotten/low playcount maps be played with an initiative is a good thing though (in the case where the map was previously overlooked because of its poor pp values and not because of its poor quality) since this would give exposure to many maps that haven't been played much.

2. Seek very hard accomplishments in terms of fcs or high acc plays on super hard maps and specialize in certain aspects of the game to try to get scores nobody else could get.

What would this mean overall for the game?


1. More balanced pp system while still being ultimately flawed because of the necessity for base pp to exist. However this is a better alternative to current pp and not as drastic option as rebuilding the pp into seperate categories like I previously suggested.

2. Most players would discover new maps to play from the already existing maps as the system encourages playing different maps instead of farming the popular maps like everyone else.




Ok I think this was all I had to say. If someone is legit reading through this entirely and ended up here then thanks for taking your time to listen to me trying to make sense. Please give me feedback and challenge my arguments, I tried to do so as best as I could myself but ultimately writing this was tiring and I most likely said some things that make little sense because of this.

tl;dr With effort this system (pp tied to leaderboards) could be made into a better pp system than the current one which would have overall a positive impact on the game

Perkele
Image
User avatar
Rhythm Incarnate
626 posts
Offline
before we go to all those fancy stuff, i think it's better for us to take a look on how the editor can be improved

not going to write a 3-page essay, but what i mean is that there needs to be an easier control on how objects are able to be made. it's hard to make an aesthetically beautiful slider when you start taking account on curves between bezier sliders.
more extreme SV ranges needs to be "unlocked". currently, we can only choose between 0.5x and 2x (without doing weird things) which isn't really enough for mapping something a lot more technical
handling timing points needs to be easier. putting a timing point for every slider takes time to do, and it should be easier if we can have an sv change that's bound to the slider
rotating stuff aren't really handy atm. i've proposed something that would make mapping symmetrical things easier to do t/626782

that's just some of them. i couldn't come up with any more ideas so i'll just put things that has concerned me since i started std mapping. also note that we have 4 modes so these problems are just small amount of them

and these can be easily ignored by more professional mappers. making editors more friendly is a way to make weirder maps easier, especially for newcomers

i just realized that my post is 1000000000x worse than the others help
Image
User avatar
Rhythm Incarnate
3,847 posts
Offline
I am going to pass on making a real wall of text on this one as others already gave a lot of reasons.
The one I consider the most grave for this debate is the effect of ppv2 on the community and mapping.

All I really have to say is that we knew the impact it had one year ago. We knew it two years ago. People posted it in the great thread that probably none of the dev has been reading for a long time (provoking claim with no backup but there hasn't been feedback from them anyway).
People posted it in other places.
It became apparent when tons of feature requests about pp popped up and got shittons of stars.
Players and mappers from pre-ppv2 eras complained about what ppv2 made of this community and what it leads new players to.
There have been several attempts to create alternative ranking systems that fix the weaknesses of ppv2.

People also made a lot of other suggestions how the editor, the modding system, kudosu stars, qualification and everything could be improved to provide the best environment for mappers to be creative. I'm positive that almost every suggestion coming out of this thread has been voiced before.

Point is: Most of it was ignored. Now, I'm not going to make a shitty "osz2" joke because the lazer-github is on fire since the start of the year but fact is that barely any major development on osu! has been witnessed in the past ~2.5 years (probably longer considering the age of "osz2" but I can only talk about the time I'm here) and that a lot of thoughts people are having and partially also voicing aren't getting the attention they might deserve.

How does all of the above help the issue at hand? It doesn't really.
What I want to say though is that I trust the people driving osu!lazer to get rid of this problem of stagnancy forever. Because people will get the possibility to get changes going with an open-source client.
It is nice that Desp as a popular community member can actually get this discussion going (despite this being nothing but a compilation of (duplicate) feature requests in the wrong subforum) and I'm interested to see what will come out of it as a short-time result.
Personally I can wait for lazer and until then I will do my best to support maps that I like.
User avatar
Rhythm Incarnate
926 posts
Offline
It should be noted that all of the systems proposed above will break pp systems in other modes without additional measures.
Relative to mode specific sets, converts are rarely played. Having a system based on playcount or number of FCs will drastically change how converts are weighted, as their playcount is significantly lower.

Lets just go through what converts mean for each mode:
  • In Taiko, converts are mainly played for pp and devalue specifics. Converts disregard a lot of the RC with unreadable SVs and boring note usage. Specifics are usually more fun and challenging.
  • In Catch, converts very popular and loved by the community due to the diversity of unique patterns and playability. However, nearly all converts give little/no pp and I’ve seen many amazing/unique plays go unnoticed due to this, while image material FC #30 gets posted to reddit.
  • In Mania, converts are badly charted, underweighted, and aren't fun to play. There’s little reason to keep them around aside from farming leaderboard spots or ranked score

If the new pp systems are added, there need to be new kinds of leaderboards for these modes to accommodate, similar to the speed/aim ones suggested before:
  • Taiko needs to have separate leaderboards for converts and specifics. There are still some impressive convert plays, but they should be in their own category. This encourages more people to play specifics, which is usually more enjoyable.
  • Catch is the same, add a convert and specific leaderboards, but to encourage more people to play converts. There are many fun converts that aren't played as there is little reason to. Many top players have quit because the leaderboards are almost specific only, and it's boring considering we only have ~9 pages of ranked maps.
  • Mania needs to have key specific leaderboards. Converts maybe have their own leaderboard, but it's almost unnecessary. Doing this would encourage players to learn other keys, as there is literally no short term benefit to right now. Key specific leaderboards would also lead to far more 5/6/8/9k maps being pushed for rank.
  • Standard mode would also benefit from the speed/aim/stamina thing mentioned above, I haven't mentioned it much as it's already been discussed

If any of the pp system ideas proposed beforehand are implemented to other modes without these leaderboards, their pp systems will break.
If the ideas above are implemented, it gives tens of thousands of players who may have quit a reason to play again, it would almost be as exciting as introducing another gamemode.

If/when the pp system is reworked, new leaderboards should be the first step. They are required for pp in other modes to work (if based on map's play count), it's faster to implement with clearer short-term benefits, and it affects players from all modes to a greater extent.
Reply 31 posts jump | prev  1, 2, 3  next

Users browsing this forum: None


Jump to: