Loctav wrote:

StarrStyx wrote:

Also, regarding the mp3 extension rule, I don't really see a reason to deny extensions if the mp3 is properly extended without any badly trimmed sections; plus I'd rather see an extension of a song rather than a crude 1-2 min cut. Imo those are even worse than extending an mp3 just by a few seconds. (Also the song compilations stuff that others have pointed out already so ya)

In my personal opinion, either way is bad, cutting and extending alike. Extending stuff by a few seconds just to adhere to personal laziness to avoid mapping a fullspread and making the set accessible to the entire playerbase (instead of just to a small minority that can even play most Approvals, which are mostly Extreme level) is contradicting some core philosophy we have been trying to defend for years (make stuff accessible and enjoyable for the majority, not just to only the top players). Like it or not, but those who play Extreme level maps are in the sheer minority. Encouraging to circumvent the necessity to produce content for the majority of our playerbase is unwanted, because with that new people will eventually not find content they like and they can play.

While the argument usually pops up that there are "already loads of mid level content to play", don't forget that newcomers to this game usually look for music they already know. And as time goes on, new music gets produced and therefore new osu beatmaps on these tracks. If these tracks are all available but only for the top tier player segment, it discourages newcomers to actually stay in osu! and enjoy it with us together. (because if you are into the hottest newest Trash Metal album and beatmaps exists of that in osu!, I doubt you can be bothered to play 500 Anime opening maps first before you can even remotely play what you actually came for)

Cutting songs is horrible, too, especially if done poorly (like literally just cutting it) and is also some sort of epitome of laziness, because mapping a full spread on 4.49 minutes long songs is definitely tedious.

I understand both sides of the argument and I can relate, like every mapper, with the laziness that comes along with mapping fullspreads, but the limits get stretched more and more. First someone extends a 4.58 minutes song to 5 minutes, then they start with 4:50, then 4:40 and as longer this goes, we have 2:00 songs just being looped five times just to avoid the fullspread. This shouldn't be a thing, in my opinion.

Can you guys stop shaming mappers for being "lazy" just because they try to avoid mapping full spreads for 4:59 songs? Their laziness has absolutely no correlation with the map's quality, or the mp3's quality, or anything that matters quality wise. The only thing potentially negative about extending mp3's is poor mp3 editing, and apparent laziness. So what if people are lazy? You are operating under a slippery slope fallacy that because 4:58 songs are sometimes extended, songs that are 4:00, 3:30, 1:00 etc... are all subject to that extension practice. Just because X happens, it doesn't mean Y is bound to happen.

Laziness should not be your primary argument, it has no tangible bearing on mapping quality which is the concern of the Ranking Criteria.



Regarding the spread issue, the entire reason we even put this on the draft was because "8 is the maximum number Loctav thinks is acceptable/defendable to developers". We never got feedback directly from dev's regarding how accurate Loctav's claims were, and seeing Ephemeral being open to both sides of the argument makes me think dev's aren't adamant on restricting difficulties to 5-6 etc... Imo it's entirely unnecessary to restrict the number of difficulties on a set. There is no quality improvement in setting a cap from my perspective as someone who has both nominated many large sets, and made many myself. Rather, as people ahve already mentioned, only negatives. You can argue for "cohesion between difficulties" all you want, but good luck describing that. It seems entirely like an argument someone developing the game, with no knowledge of the mapping meta, would consider a reasonable argument. Do consider that often times, guest difficulties are there not for cohesion, but for variety and contrast. Also, how exactly will capping difficulties improve cohesion, if this wasn't already an abstract enough concept. Additionally, do you think cohesion is actually that important of a feature in mapsets? What defines a cohesive mapset? I don't believe this reflects the current mapping meta in the first place, and is too far removed from philosophies the current mapping community considers high quality. It's something you have to be active in the mapping and modding community to understand, honestly.



Anyways, just voicing my ideas. It's sad how lonely I was in the ranking criteria discussion. I'm supposed to be playing the devils advocate and trying to poke holes in the draft, and ways to abuse the current wording, so we can iron them out before the public proposal. I shouldn't be the only one who is arguing against capping difficulties... (though i guess council members are inactive too...) This issue is clearly something the vast majority of the community has disagreed on...

I think the limit of 8 diff in a set will be beneficial in the sense that mapper will stop that think that do a mapset is equal to make 5 extras because pp.
But its true that doing this a rule will be bad idea, like in Kenterz's example.

I propose make this a guideline, where if any mapper wanna broke the guide, must to justify this very well


(Now, in a personal side, i can see how most players usually play only one or two extras in a set... so limiting the amount of extras in a set will be what the data is showing that player really wish?)
:thinking:

I also think a guideline is a better idea for the amount of difficulties.
The criteria for breaking that guideline should be either spread or unique approaches. I can see the appeal of no restrictions for the amount of difficulties but all to often you get multiple extra diffs that only differ in minor details stylewise where it is really questionable what justifies the existence of that difficulty when it is really just bloating the set, bloating the work for modding and qualification and achieving nothing gameplaywise that sets it apart from the other difficulties.
I really like the reasons for having smaller sets as it also comes with a potential increase in quality for these sets. Not to mention that the musical diversity might profit in the long run because instead of mapping 5 similar extras for the same song, we might see an extra for 3 other songs instead.

A problem would be how one could get mappers to not make GDs that match each other as seen here.
Because it would be lame to tell them later to remove one of the diffs due to duplication.

haven't read op+thread, just got linked to mons's post and from composer pov I disagree with non-purposeful extending. Unrelated to the diff-spread issue obv.

Monstrata wrote:

Can you guys stop shaming mappers for being "lazy" just because they try to avoid mapping full spreads for 4:59 songs? Their laziness has absolutely no correlation with the map's quality, or the mp3's quality, or anything that matters quality wise. The only thing potentially negative about extending mp3's is poor mp3 editing, and apparent laziness. So what if people are lazy? You are operating under a slippery slope fallacy that because 4:58 songs are sometimes extended, songs that are 4:00, 3:30, 1:00 etc... are all subject to that extension practice. Just because X happens, it doesn't mean Y is bound to happen.

Laziness should not be your primary argument, it has no tangible bearing on mapping quality which is the concern of the Ranking Criteria.

You're right in defending the fact mappers aren't being lazy for doing this but mp3 editing goes against the creator's work; a single measure being repeated to make it 5:xx+ instead of 4:59 could make the track be full 4/4 + 5/4 on one occasion which makes no sense etc if it wasn't structurally/musically intended. What would the guidelines/rules be required to say in that case; "extending is allowed if it's subjectively ok sounding and insert-more-details-that'll-eventually-be-more-restrictive-in-a-year-or-two instead of doing w/e" ? Longer extensions could be swell but then we'd have to 'set limits' on subjective povs in that
direction as well so it'll lead to issues as well. Write it out nicely if this does get allowed so such things don't come up.

Otherwise osu! has stuck to its 5:00 limit for quite some time so why not stick with the current Loved or unranked plan if a mapper doesn't want to do multiple difficulties; the Loved category also exists due to this issue, not only due to lack of mods/etc.

Other possibility: mappers can work on doing their own remixes: would even small variations to an existing track be allowed? In the music industry it isn't but this game disregards their copyright/etc laws already so it could be handled differently and mappers+staff can come to a compromise in this direction.

For the extension issue, I would again argue it is not about "why should we ban extended mp3?" but "why should that 1 second matter so much?"
As I explained before, if the spread rule could vary according to the song length, everything will be clear.

People are not afraid to make 1 or 2 more difficulties, but worried about making 5 more diffculties just because of that 1 second. would be nice if we could think about smt like this: t/432739/start=0

Actually, it is not reasonable to rank a single X diff, even it is an approval diff, in my opinion. i really wish the managing team understand the App should not be a reason for mappers to edit mp3, but serve to make the spread of ranked sets better.
Thus, mappers do not really need to worry much about extending mp3, therefore problem solved.

alice soft wrote:

haven't read op+thread, just got linked to mons's post and from composer pov I disagree with non-purposeful extending. Unrelated to the diff-spread issue obv.

Monstrata wrote:

Can you guys stop shaming mappers for being "lazy" just because they try to avoid mapping full spreads for 4:59 songs? Their laziness has absolutely no correlation with the map's quality, or the mp3's quality, or anything that matters quality wise. The only thing potentially negative about extending mp3's is poor mp3 editing, and apparent laziness. So what if people are lazy? You are operating under a slippery slope fallacy that because 4:58 songs are sometimes extended, songs that are 4:00, 3:30, 1:00 etc... are all subject to that extension practice. Just because X happens, it doesn't mean Y is bound to happen.

Laziness should not be your primary argument, it has no tangible bearing on mapping quality which is the concern of the Ranking Criteria.

You're right in defending the fact mappers aren't being lazy for doing this but mp3 editing goes against the creator's work; a single measure being repeated to make it 5:xx+ instead of 4:59 could make the track be full 4/4 + 5/4 on one occasion which makes no sense etc if it wasn't structurally/musically intended. What would the guidelines/rules be required to say in that case; "extending is allowed if it's subjectively ok sounding and insert-more-details-that'll-eventually-be-more-restrictive-in-a-year-or-two instead of doing w/e" ? Longer extensions could be swell but then we'd have to 'set limits' on subjective povs in that
direction as well so it'll lead to issues as well. Write it out nicely if this does get allowed so such things don't come up.

Otherwise osu! has stuck to its 5:00 limit for quite some time so why not stick with the current Loved or unranked plan if a mapper doesn't want to do multiple difficulties; the Loved category also exists due to this issue, not only due to lack of mods/etc.

Other possibility: mappers can work on doing their own remixes: would even small variations to an existing track be allowed? In the music industry it isn't but this game disregards their copyright/etc laws already so it could be handled differently and mappers+staff can come to a compromise in this direction.

If such things occur, they should be brought up in the modding process anyways. The rules shouldn't replace BN's abilities to make sound judgements. I've never come across a poor mp3 edit getting past qualification anyways though .

I guess if you want to do some 7 extra mapset
you gotta at least have different styles on each difficulty and not have them be exactly the same,otherwise there wouldnt be a need for the other difficulty to be in the spread

Loctav wrote:

Like it or not, but those who play Extreme level maps are in the sheer minority. Encouraging to circumvent the necessity to produce content for the majority of our playerbase is unwanted, because with that new people will eventually not find content they like and they can play.

Do you want to give numbers to back up your data?

Let's take a top 5 played map like No Title, since this is probably a decent choice that represents what people want to play given its playcount https://osu.ppy.sh/b/766475

At the time of posting, here are the amount of passes and plays in diffs at various ratings:

Difficulty
Passes Plays

E
413892 650246

N
299507 614919
374551 1264139
============
674058 1879058

H
480281 2936976

I
671031 3449540
261651 2375593
195770 2166451
211253 1877997
============
1339705 9869581

X
45155 628154
129675 1354980
116640 1179720
138373 1566427
75497 1072797
133822 1602187
============
639162 7404265

Passes:
I > H > N > X > E

Plays:
I > X > H > N > E

I'll leave interpretation of the data up to others, but it certainly seems that most people rather play I and X rather than N or E.

bad edits to mp3s shouldn't be reason to stop mp3 extending / cutting. Bad mp3 edits getting ranked shouldn't even be a thing right now; BNs / modders aren't stupid. They can tell if an mp3 edit is garbage and would probably tell the mapper to do a better one if possible.

Really feels to me like a lot of the reasoning provided to defend these changes show a real disconnect between the higher ups / devs and the mapping community to me. I don't want to make it sound like I'm insulting anyone, it's just what I feel is the case.

blissfulyoshi wrote:

Loctav wrote:

Like it or not, but those who play Extreme level maps are in the sheer minority. Encouraging to circumvent the necessity to produce content for the majority of our playerbase is unwanted, because with that new people will eventually not find content they like and they can play.

Do you want to give numbers to back up your data?

Let's take a top 5 played map like No Title, since this is probably a decent choice that represents what people want to play given its playcount https://osu.ppy.sh/b/766475

At the time of posting, here are the amount of passes and plays in diffs at various ratings:

Difficulty
Passes Plays

E
413892 650246

N
299507 614919
374551 1264139
============
674058 1879058

H
480281 2936976

I
671031 3449540
261651 2375593
195770 2166451
211253 1877997
============
1339705 9869581

X
45155 628154
129675 1354980
116640 1179720
138373 1566427
75497 1072797
133822 1602187
============
639162 7404265

Passes:
I > H > N > X > E

Plays:
I > X > H > N > E

I'll leave interpretation of the data up to others, but it certainly seems that most people rather play I and X rather than N or E.

Exactly, most of the player base plays I/X or superior. Even high digit players play I or X instead of H or below.
That's also why adding a restriction to 8 difficulties per set it's just going to kill the Easy difficulty. No one is going to add an Easy instead of an Extra or another Insane. In fact, if you look at the data, you'll find that usually in every single set, the lower difficulties are often played 3-4 times less than the top difficulty.

Those numbers are not adequate for an analysis. You must take numbers only in 1 month or so.
If you take all plays, any conclusion will be wrong.

ErunamoJAZZ wrote:

Those numbers are not adequate for an analysis. You must take numbers only in 1 month or so.
If you take all plays, any conclusion will be wrong.

Well you can repeat the same exercise with Hitorigoto (https://osu.ppy.sh/s/596704) and get the same result, except there are no X diffs, so you only see I plays.

But finding maps that were ~1 month old had really small sizes for numerical purposes, but here https://osu.ppy.sh/b/1251083&m=0

I think there are more players who play ENH, but the ones that play IX retry many more times, driving up the playcount.

Still not a reason to limit the number of difficulties in a mapset, which everyone should have realized by now is a terrible idea.

xxdeathx wrote:

I think there are more players who play ENH, but the ones that play IX retry many more times, driving up the playcount.

Still not a reason to limit the number of difficulties in a mapset, which everyone should have realized by now is a terrible idea.

Thinking and proof are 2 different things. Without proof, your opinions mean nothing.

xxdeathx wrote:

I think there are more players who play ENH, but the ones that play IX retry many more times, driving up the playcount.

Still not a reason to limit the number of difficulties in a mapset, which everyone should have realized by now is a terrible idea.

Actually the number of (true) people who play ENH difficulties is pretty low. Most people who play and retry lower difficulties are people who HDDTHR them to be in the leader boards. The only change happens on old popular maps, but that's because it's one of the first maps new players download so it only represents that popular maps are the gate to new people.

pishifat wrote:

A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due. Drain times for collaborative difficulties must be listed in the creator’s words for via storyboarding.

Weird wording there, "for via storyboarding".

ErunamoJAZZ wrote:

I think the limit of 8 diff in a set will be beneficial in the sense that mapper will stop that think that do a mapset is equal to make 5 extras because pp.
But its true that doing this a rule will be bad idea, like in Kenterz's example.

I propose make this a guideline, where if any mapper wanna broke the guide, must to justify this very well


(Now, in a personal side, i can see how most players usually play only one or two extras in a set... so limiting the amount of extras in a set will be what the data is showing that player really wish?)
:thinking:

Endaris wrote:

I also think a guideline is a better idea for the amount of difficulties.
The criteria for breaking that guideline should be either spread or unique approaches. I can see the appeal of no restrictions for the amount of difficulties but all to often you get multiple extra diffs that only differ in minor details stylewise where it is really questionable what justifies the existence of that difficulty when it is really just bloating the set, bloating the work for modding and qualification and achieving nothing gameplaywise that sets it apart from the other difficulties.
I really like the reasons for having smaller sets as it also comes with a potential increase in quality for these sets. Not to mention that the musical diversity might profit in the long run because instead of mapping 5 similar extras for the same song, we might see an extra for 3 other songs instead.

A problem would be how one could get mappers to not make GDs that match each other as seen here.
Because it would be lame to tell them later to remove one of the diffs due to duplication.

I like this idea

blissfulyoshi wrote:

xxdeathx wrote:

I think there are more players who play ENH, but the ones that play IX retry many more times, driving up the playcount.

Still not a reason to limit the number of difficulties in a mapset, which everyone should have realized by now is a terrible idea.

Thinking and proof are 2 different things. Without proof, your opinions mean nothing.

how about just look at the ranks of people and the range of those ranks can perform at, its not rocket science to see most people play ENH, but many more relevant people who contribute to this game play I+

pishifat wrote:

Spread

Rules

All rules are exactly that: RULES. They are NOT guidelines and may NOT be broken under ANY circumstance.

1. If a hybrid mapset includes osu!standard difficulties...
   
   1. One osu!catch difficulty may be included. It must be at least an Insane difficulty.

Does this counter this rule now? t/292401

The current standing explains that catch the beat diffs on hybrid sets could not contain ONLY an X-level difficulty (Overdose). The new rule implies that only one difficulty can be added and it can be either I (Rain) or X (Overdose) level? Unless I'm missing something here, isn't this a bit counterproductive towards the growth of our mode, considering a good amount of hybrid sets have H-level (Platters) on them?

Also why are we limiting the amount of difficulties on a hybrid mapset? It's like we're being discouraged from providing content to our users. It's not like we've ever had a problem with too many difficulties in a set before in our mode, so there's almost no reason to enforce something on us like this. Our mode is dying, please don't kill it more. It's already bad enough that I have 1/2 of the ranked osu!catch maps in the last 2 months lol

blissfulyoshi wrote:

xxdeathx wrote:

I think there are more players who play ENH, but the ones that play IX retry many more times, driving up the playcount.

Still not a reason to limit the number of difficulties in a mapset, which everyone should have realized by now is a terrible idea.

Thinking and proof are 2 different things. Without proof, your opinions mean nothing.

borborygmos wrote:

how about just look at the ranks of people and the range of those ranks can perform at, its not rocket science to see most people play ENH, but many more relevant people who contribute to this game play I+

Pretty much this when I said what I said. I shouldn't have to go off topic to argue the semantics of my slightly incorrect word choice when the proof is in plain view.

borborygmos wrote:

how about just look at the ranks of people and the range of those ranks can perform at, its not rocket science to see most people play ENH, but many more relevant people who contribute to this game play I+

xxdeathx wrote:

Pretty much this when I said what I said. I shouldn't have to go off topic to argue the semantics of my slightly incorrect word choice when the proof is in plain view.

What plain view? What logic? Prove it with data. I'm not saying any of you are wrong, but if people interpret the same source of data differently, then whatever you say means nothing to those seeing that data, unless you back it up with something more solid than your own conclusion.

Note: I never said anything about H diffs.

Monstrata wrote:

If such things occur, they should be brought up in the modding process anyways. The rules shouldn't replace BN's abilities to make sound judgements. I've never come across a poor mp3 edit getting past qualification anyways though .

Can't say I have either so point taken.

Ascendance wrote:

pishifat wrote:

Spread

Rules

All rules are exactly that: RULES. They are NOT guidelines and may NOT be broken under ANY circumstance.

1. If a hybrid mapset includes osu!standard difficulties...
   
   1. One osu!catch difficulty may be included. It must be at least an Insane difficulty.

Does this counter this rule now? t/292401

The current standing explains that catch the beat diffs on hybrid sets could not contain ONLY an X-level difficulty (Overdose). The new rule implies that only one difficulty can be added and it can be either I (Rain) or X (Overdose) level? Unless I'm missing something here, isn't this a bit counterproductive towards the growth of our mode, considering a good amount of hybrid sets have H-level (Platters) on them?

Also why are we limiting the amount of difficulties on a hybrid mapset? It's like we're being discouraged from providing content to our users. It's not like we've ever had a problem with too many difficulties in a set before in our mode, so there's almost no reason to enforce something on us like this. Our mode is dying, please don't kill it more. It's already bad enough that I have 1/2 of the ranked osu!catch maps in the last 2 months lol

Nope, it's not meant that way, just a case of bad wording again LOL
It's supposed to say pmuch the same as that old rule, so I guess smth like "..having only one osu!catch difficulty suffices when it's an Insane or lower" would work better

In the interests of fair and accurate discussion on the matter, I pulled some stats down for beatmap plays across all difficulty ranges of ALL approved, ranked, qualified and Loved beatmaps:



Make of that what you will.

Ephemeral wrote:

In the interests of fair and accurate discussion on the matter, I pulled some stats down for beatmap plays across all difficulty ranges of ALL approved, ranked, qualified and Loved beatmaps:



Make of that what you will.

Just checking, this is across all time?

Across all ranked plays, yes. You'll note that it adds up to almost the exact number of purported "total ranked plays" at the top of the page as well.

But then the consideration is that there are still some sets that may not have an Extra or an Insane at all so it's still not really a perfect representation of what the player base plays these days. (Since you mentioned all maps Im just gonna assume it includes the older maps before 2013, and back then we didn't have much Extra difficulties (or sometimes not even Insanes) compared to now.)

edit: ok so it is indeed all maps

Good point.

2014-present:


2015-present:


2016-present:

------

2008-2015:


2015-2016:


2016-2017:


2017-present:

It would be interesting to see how those numbers corresponded to Playcount. I would like to believe that the more dedicated players, people who put time in this game, will make up the majority of the Insane/Extra plays, and people just visiting the site / playing when they're bored for like 2 hours a month or something would make up the majority of Normal/Hard plays. All conjecture though of course.

I would prefer catering to more dedicated players in the community over people who just play once in a while.

A reasonable assertion would be that NHI-HIX spread would adequately encompass about 60-70% of the playerbase based on these statistics alone.

Easy difficulty outlier-status is easily explained by its incredibly high pass ratio compared to other difficulties - most easies are probably too easy for 95% of all players, even newer ones, and are frequently only ever played once or in very few numbers before players graduate on to Normal as the defacto "entry-level" difficulty.

blissfulyoshi wrote:

borborygmos wrote:

how about just look at the ranks of people and the range of those ranks can perform at, its not rocket science to see most people play ENH, but many more relevant people who contribute to this game play I+

xxdeathx wrote:

Pretty much this when I said what I said. I shouldn't have to go off topic to argue the semantics of my slightly incorrect word choice when the proof is in plain view.

What plain view? What logic? Prove it with data. I'm not saying any of you are wrong, but if people interpret the same source of data differently, then whatever you say means nothing to those seeing that data, unless you back it up with something more solid than your own conclusion.

Note: I never said anything about H diffs.

The fact that there are 10 million users and no more than the top few hundred thousand that are capable of playing insanes is not enough to support my original hypothesis?

there are more players who play ENH

That's the data and I see no need to analyze it to the point where you're convinced when most others are able to take it for granted

I think it's less an issue of easiness, and more an issue of sheer quantity. Every mapset that is not approval has an Easy difficulty, but the learning curve in progressing from playing Normals > Hards is a lot steeper than Easy > Normal (And Hard > Insane even steeper still). My point being, because there are steeper learning curves the harder you get, the more a player is likely to play more maps with similar difficulties. You might have to play ~1000 Hard maps before you've improved enough in skill level to properly play Insanes. But you probably would only need a fraction of those plays before progressing from an Easy to a Normal. Therefore, you wouldn't feel the need to "try" as many different Easy-difficulty maps.

It's why imo Easies are so untouched. Tbh I think the same holds true for Normal, but the number is offset by the sheer quantity of people who just play osu as a past-time like maybe for an hour every month or something.


But anyways. interesting statistics, thanks for sharing! I wasn't expecting Insanes to have that many plays. I always thought it would be the Hard or Normal difficulty judging from all the bundled beatmap playcounts. Reaffirms my opinion that a spread cap is unnecessary.

Important to note that the 2014-2015 discrepancy appears to be due to a change in the way beatmaps are recorded serverside, and is instead an aggregate of 2008-2014.

since spread cap of 8 isn't really needed here, would changing it so that it's a diff specific cap work better?
eg. capping at 4-5 insanes.
prevents things from getting spammy without limiting the max difficulty, eg if an a hyper or advanced was needed for spread

I think a rule limiting difficulties that are not significantly different would do well to address the bloat issue while still keeping things fair for creators. The difficulty (hah) then comes in determining what is "significantly" different.

As data displayed, the majority of plays can be found in the segment of Hard, Insane and Normal - and Expert difficulty being in the minority (together with Easy, which is clear because not all sets must have an Easy). The amount of Expert difficulties produced in recent history are in no relation to the amount of plays they actually receive. The excerise of increasing the set's amount of difficulties to add *even more Experts* (although they are not even targetted at who actually plays the set) just adds content bloat - a lot of the same, without a distinctive difference.

I debated with Ephemeral for a while and I can get behind two alternative proposals being made by fellow community members, one being to turn the set size limitation into a guideline, saying that "sets shall not exceed the amount of 8 difficulties per game mode, unless the exceeding ones are a significantly different approach of interpretation"

It is a middle ground between "we need a limit" and "whoever actually exersizes varied mapping and adds an actual multitude of map designs as alternative to their mapset, is allowed to do so, if the alternatives are distinctive enough from each other"

But I still cannot see any valid and reasonable points about the necessity of putting limitations of mapset size.
For building up difficulty of ranking and modding process? If a mapper meets difficulty on pushing their map forward, they will automatically reduce their mapset size. And if they do not want to do that, they take the consequences by themselves. Same for modders and BNs. And I dont think it is reasonable for the criteria to handle this scheme.

For the majority-minority issue? Actually there is no determined sign and valid reason to conclude if with this new guideline, more beginner friendly maps will be created. It might just increase the proportion of simple difficulties overall but not the total number of them. This cannot solve the problem you put forward at all and even might reduce the range of choices for more pro players. Also it is not sure if people who wanted to rank a huge mapset initially will make two sets instead, because of the limitation.

responding to things that i have a different view on and why if i deleted something from your quote you can guess that it's because my viewpoint is sufficiently explained in other posts on the thread

UndeadCapulet wrote:

update wrote:

Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose.

If a mapper is going to go out of their way to have the mp3 edited to reach marathon length, this rule won't get them to make a full spread. This rule is choosing between "have a 5 minute map for a song that isn't 5 minutes" and "have no map". As long as the actual play experience of the map is still pleasant, it is better to have the map than not have the map. There is never anything bad about having more maps.
There are plenty of songs that really don't fit for a full mapspread, but just barely don't reach minimum drain time. My 4:45 300BPM death metal anthem is not going to make for a good experience as a Normal difficulty. Requiring a mapper to design a full spread for certain songs that clearly don't need it isn't beneficial to rank quality.
To be honest this rule is almost unenforceable anyway since the line between user remix and mp3 extending is super blurry. Right now I can't add sounds to the beginning or end, but I CAN add sounds consistently throughout the mp3, call it an Edit ver. and rank it that way. People will always find loopholes to unnecessary rules.
It's also unclear whether separate songs are included in the "adding sounds to the end of the mp3" or not. Song compilations can be considered unrankable with this current wording.

If a mapper is going out of their way to custom-extend a mp3 to dodge a spread they do just that: dodge a spread and thus content they would normally have to provide to the game's different target audiences which is detrimental to a mapset's quality overall and not just plainly lazy.

The whatever diff they make out of it's quality would remain untouched if they made a full spread plus it would offer other target audiences something to play. You claim that some songs don't need a spread whereas I think at this point in the thread it's kind-of obvious that adding a spread will not be detrimental to a map's quality and if you let it influence your maps quality you should probably rethink why you are ranking stuff in the first place.

We went for explicitly listing what constitutes as abuse for the 5 minute draintime ruling on marathon maps to avoid the mentioned scenarios in mapping as well as mp3 editing for this purpose to get away with it, song compilations are a thing that would probably need talking about with this wording, i agree.

update wrote:

A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due. Drain times for collaborative difficulties must be listed in the creator’s words for via storyboarding.

change "for" to "or" fair
The mapset owner is in charge of not just mapping, but frequently also asking for every guest diff, finding mods, hitsounding, storyboarding, timing, balancing spread, and ensuring every other included difficulty is rankable. The set owner basically always deserves credit regardless of how much work they actually did.
If the guest difficulty mappers are okay giving mapset credit to a mapper who did "less work" than them, there should be no reason to not let them. If they weren't okay with it they wouldn't be in the set.
it has been agreed upon that the one hosting a set has extra tasks which he voluntarily takes over by hosting the set, but to be able to do so there needs to be some sort of failsave to assure that the mapset's host has contributed to the mapping side of a set that he hosts unless we want to see people ranking gd-only sets again.
This minimum effort that someone has to put into a set in order to claim to be the host has been decided to be the amount of draintime they mapped to leave out wiggle room for ambiguity. obviously mapped objects would not work as a measure because then it'd dictate that more dense = better so the only component that went into this rationale was the draintime someone mapped.

I think as a rationale this makes a whole lot of sense because it aims to avoid people going for ranking with content that they cannot even claim to have done the majority on.

update wrote:

Usernames indicating possession of a guest difficulty should be consistent between multiple mapsets. Varying nicknames for one user makes interpreting who created a difficulty confusing.

If the mapper doesn't care whether they're credited or not then that's entirely on them. The real username will be in tags and description anyway for users to find them so it doesn't matter what nickname they use in the diff name. This is just a "No Fun Allowed" rule that doesn't affect map quality in any way.

To be honest this can be said about every diffname rule/guideline. Is it really necessary to police them so strictly? With the new star rating the map difficulty can be somewhat reasonably determined without any difficulty name, so I don't see any reason to carry over old rules.
At least an explanation as to why you're using a different alias every goddamn set without changing your username for no other reason than lolz would be nice. This guideline is intended to bring forward more clarity about who a guest diff indicating possession is actually crediting because with recent developments this became quite unclear in some cases

Minimum draintime rule should talk about abusing the 5 min because the extend your song to get 30 seconds thing is actually encouraged

for anyone wondering a few of the compelling reasons for the first guideline were given by https://osu.ppy.sh/b/765299&m=0
second guideline shoudl not only recommend but require clear and appropriate relation to a song

requiring one or two difficulties if you do a marathon length set to encompass the majority of the playerbase if you are doing an expert level difficulty is something i could agree with.

As for the 8 difficulty spread limitation: I think this thing is already being really lenient (but probably only because i know all our alternative ideas for this lol), but if we want it to work as a guideline then we should remove the sentence that states that the highest level difficulties can be out of spread and would allow reasonable high-end spread to be one of the reasons to break the guideline

Additionally if we see sets that are top heavy with a lot of redundant content bloated into them a problem then allow a maximum of <N> diffs around the same difficulty to allow <N> distinct interpretations though we would then need to define what constitutes as such because if the majority of the rhythms and concepts used in a map are identical it'll become really hard to draw the line, this would need definition of further extra tier difficulty levels along with standardizing naming schemes for them which goes along xexxar's idea which he posted but never really bothered to answer my concerns from 6 months ago on

Also @Monstrata: contrary to popular belief I do actually sleep (a lot more on weekends, by the way) so just concluding / asserting that we are completely inactive when it comes to dealing with this thing is ridiculous

Ephemeral wrote:

Good point.
2014-present:

2015-present:

2016-present:

------
2008-2015:

2015-2016:

2016-2017:

2017-present:

Thanks Eph!, I love it.
I knew about this behavior looking in amount of plays in my loved set, but I did not spected that hards were the most tbh.
That explains why hitorigoto have a lot of plays...

Data > Opinions! <3

Okorin wrote:

If a mapper is going out of their way to custom-extend a mp3 to dodge a spread they do just that: dodge a spread and thus content they would normally have to provide to the game's different target audiences which is detrimental to a mapset's quality overall and not just plainly lazy. The whatever diff they make out of it's quality would remain untouched if they made a full spread plus it would offer other target audiences something to play.

Again, mappers that extend mp3 to 5min are not going to be making a full spread because of this rule. It's just not going to happen. They'll just pick a different song to map. Extending mp3 can't be detrimental to a mapset when that mapset wouldn't exist at all without the extension. Between 1 map and 0 maps, clearly the 1 map is the better option.
Keep in mind that not everyone has enough friends to gather mods/gd's/bn's for a 4:58 ENHIXU set, only a small handful of even well-known mappers can make that happen. Limiting the kinds of songs that can be reasonably pushed forward is definitely detrimental to the quality of the ranked section.
From my personal experience as a new player, experiences I've had with irl friends getting into the game, and just talking to less skilled players, in general maps above 4 minutes are not appealing to new players. They're just too long and exhausting, constant focus on not failing for 4 minutes straight is a skill only Insane/Expert players, and maybe some Hard players have developed to begin with. Making a rule forcing mappers to appeal to a target audience that would never find the map appealing is not beneficial to rank quality.

Okorin wrote:

You claim that some songs don't need a spread whereas I think at this point in the thread it's kind-of obvious that adding a spread will not be detrimental to a map's quality and if you let it influence your maps quality you should probably rethink why you are ranking stuff in the first place.

Feel like I may have implied something by mistake here. I didn't mean to say the rule would be detrimental to map quality, just that it wouldn't improve it. It does nothing at best. It only punishes "lazy mappers", reduces the diversity of songs in the ranked section, and doesn't provide any major benefits outside of this.

If you really think it necessary to restrict mp3 editing to this degree, then please instead implement something closer to what Shad0w1and suggested here. This would lead to more maps for more players to play that would actually be played. Seems like a reasonable compromise.

Okorin wrote:

it has been agreed upon that the one hosting a set has extra tasks which he voluntarily takes over by hosting the set, but to be able to do so there needs to be some sort of failsave to assure that the mapset's host has contributed to the mapping side of a set that he hosts unless we want to see people ranking gd-only sets again.

I don't see how gd-only sets is a bad thing. All the gd-er's are aware and consenting to being in a set that the host didn't participate in. They are the only people who should be concerned about getting credit for their work, and they're still in the set. If they don't care, why should we? Is it just an issue of staff not wanting to give contest/BestOf rewards to someone that didn't map anything? Because the rewards are still given in the case of someone who only mapped an Easy for their ENHIIXXU4K5K6K7KTaikoCTB set. There's little difference between rewarding this and rewarding a gd-only set.

Okorin wrote:

At least an explanation as to why you're using a different alias every goddamn set without changing your username for no other reason than lolz would be nice. This guideline is intended to bring forward more clarity about who a guest diff indicating possession is actually crediting because with recent developments this became quite unclear in some cases

Then why isn't the gd'er fully required to put their username in the difficulty? I think it's way more unclear to have "Insane by Kibbleru" than it is to have "Quibboo's Tragic Love Insane feat. Kibb by Kibbleru".
But as I said already, any diffname restriction is unnecessary in my opinion, because star rating shows map difficulty to a reasonable degree for the average player to have a reasonable expectation of difficulty. We've just recently had multiple cases of the mapping community rejecting mappers' silly Yuri Imouto with a Sword diffname, showing the RC limitations aren't helpful (these were all rankable) or needed (community can oversee itself to ensure nothing too ridiculous is ranked). This isn't exactly the most serious game out there (our lord creator peppy makes this clear enough), osu! is just a fun hobby, so what's wrong with a teeny bit of fun? Diffname affects map quality literally 0%, restricting in any way is excessive imo

---

In regards to 8diff as a guideline, I don't see how that's enforceable. There's no clear line for what makes 2 diffs of the same level distinct enough experiences to say "this set is unrankable under this guideline, while this other set is perfectly rankable." It would be case-by-case, heavily swayed by the current atmosphere of whatever was recently ranked, inconsistent, and lead to unpleasant community interactions.

In regards to 8diff as a rule:
It's been said the rule is intended to break up top-heavy sets, but the current implementation would be breaking up sets with an even spread that are just really big, like Hitorigoto, the most-played map since it was ranked. Why are we removing sets clearly approved of by players?

And I still don't see how top-heavy sets is a bad thing. It's still more maps being ranked. More maps is never a bad thing. Loctav used the term "content bloat", but breaking up the Expert diffs into other sets would create even more bloat, since now there are also multiple low diffs, across multiple sets, which requires multiple downloads, and makes it harder to find a different song when browsing through newly ranked. If anything we should be encouraging mappers to combine their sets for player convenience.

Since 2015 Insanes have been the most-played difficulty judging by Ephemeral's numbers. There are plenty of sets that are Insane-heavy as well as Expert heavy, that this rule would cut out. This is yet another way the rule would be going against the playerbase's desires. Really disagree with a rule that goes against what both players and mappers like.