forum

[Proposal] Relaxation/revamp of the osu!mania Spread Requirements

posted
Total Posts
37
Topic Starter
Quenlla
Context

During the last month, various threads and proposals have arised to discuss and propose changes to the spread requirement sections of Ranking Criteria. The excessive usage of graveyard as a permanent map storage was deemed partially a result of the stagnation of the mode's ranking section in an initial discussion that intended to increase the limit of pending maps; this thread saw a common opinion be conceived, about both hitsound requirements and spread requirements being major roadblocks to achieve a higer diversity and quality in the ranked section. Full mapset hitsounding is not necessary anymore after this week's changes, so now it's the time to finally get to work on spread requirements.

Abraker contributed to the matter with two proposals vouching first for a relaxation of the requirements, and later a complete removal of those. These discussions, as expected, weren't totally able to reach a common consensus but gave a clear hint that a majoritary part of the community saw changes on spread requirements as necessary. This is understandable, as it constitutes a much broader topic than hitsounds with longterm consequences on the mode's ecosystem way larger as well, as spreads seem to be both a pillar and a limitation to the current Mania ecosystem.

This difficulty for reaching for consensus with a lineal forum discussion was what drove us to conduct a proper data collection process in order to listen to the community and ensure that every opinion is heard and accounted for.

Collected Data

To gather the playerbase and mapper's opinions on the topic, we designed a survey in which respondants were asked both general profiling questions and spread-related ones. This has been repeated over several times, but of course the survey was designed in order to be as neutral and objective as possible, and was reviewed by multiple NATs and BNs to ensure it.

The results and numerical findings of the survey have been publicly shared in the following website:

> Survey Results <

From here on, we would like to conduct some unbiased result analysis:
  1. In all cases, we have seen that Keeping the current rules is an inmensely minoritary option, even among mappers used to ranking sets and therefore used to the current spread criteria. There is a consensus that a change needs to happen.
  2. We have observed an always theorised disconnetion between the mapping and playing faces of the community: most non-mapping players give a higher priority to a complete removal of the spread requirements (>63%), while those who do map lean towards a relaxation of the requirements, and a good portion of them tend to disagree with a removal. It's important to take into account though that the data here is certainly biased and may overrepresent removal preferrencies between the playing face of the community: more than >80% of respondants where rank #10k or higher, and therefore naturally less dependant on full spreads for their playing necessities.
  3. Lower skilled players have a higher tendency to believe Keeping the current spread rules would be more benefitial for them than Removing them (albeit keeping still being the minoritary option). In all cases and skill ranges, a relaxation is still the overall preferred option, both as first and second choices.
  4. Mappers with ranked maps have a stronger tendency to prefer a Relaxation than mappers who don't have any ranked maps. This could be interpreted in different ways: new faces to the ranked section could arise with a more aggresive revamp of the requirements, and/or mappers experienced in making spreads understand the cruciality of mantaining a sufficient influx of content for players in all skill ranges. Both hypothesis are consolidated looking at the results from the "Mapper-specific questions" in the survey results.

Conclusions and Proposal

As previously stated, the gathered data shows an overwhelming consensus from the community that a change in the Spread Requirements section of RC needs to happen as soon as possible to help solving the current stagnation of the mode as whole, and specially to improve the quality and variety of the ranked section. How to achieve this change is of course a matter of disagreements, and therefore we have tried to modulate the proposal in order to accomodate the preferences of all positions involved.

Proposed RC wrote:

If the drain time of each osu!mania difficulty is...
  1. ...lower than 2:30, the lowest difficulty of each included keymode cannot be harder than a Normal,
    OR each keymode must provide a spread starting at least 3 difficulty levels below the highest difficulty.
  2. ...between 2:30 and 3:15, the lowest difficulty of each included keymode cannot be harder than a Hard,
    OR each keymode must provide a spread starting at least 2 difficulty levels below the highest difficulty.
  3. ...between 3:15 and 4:00, the lowest difficulty of each included keymode cannot be harder than an Insane,
    OR each keymode must provide a spread starting at least 1 difficulty level below the highest difficulty.

Difficulties below the highest difficulty can combine break times with drain time to meet the above thresholds. This does not apply to difficulties with less than 30 seconds of drain time. Judgement regarding the suitability of gameplay elements used for any lowest difficulty Hard and Insane difficulties is up to the Beatmap Nominators and Nomination Assessment Team members for the corresponding game modes.

For difficulties above the lowest required difficulty level, the spread cannot skip any difficulty levels and there cannot be any drastically large difficulty gaps between any two difficulties. On difficulties Insane and harder, a proper spread is defined by relative difficulty gaps similar to those formally defined between lower difficulties.

Current General RC wrote:

If the drain time of each difficulty is...
  1. ...lower than 3:30, the lowest difficulty of each included game mode cannot be harder than a Normal.
  2. ...between 3:30 and 4:15, the lowest difficulty of each included game mode cannot be harder than a Hard.
  3. ...between 4:15 and 5:00, the lowest difficulty of each included game mode cannot be harder than an Insane.


Difficulties below the highest difficulty can combine break times with drain time to meet the above thresholds. This does not apply to difficulties with less than 30 seconds of drain time. Judgement regarding the suitability of gameplay elements used for any lowest difficulty Hard and Insane difficulties is up to the Beatmap Nominators and Nomination Assessment Team members for the corresponding game modes.

Reworded General RC wrote:

If the drain time of each difficulty is...
  1. ...lower than 3:30, the lowest difficulty of each included game mode cannot be harder than a Normal.
  2. ...between 3:30 and 4:15, the lowest difficulty of each included game mode cannot be harder than a Hard.
  3. ...between 4:15 and 5:00, the lowest difficulty of each included game mode cannot be harder than an Insane.


Difficulties below the highest difficulty can combine break times with drain time to meet the above thresholds. This does not apply to difficulties with less than 30 seconds of drain time. Judgement regarding the suitability of gameplay elements used for any lowest difficulty Hard and Insane difficulties is up to the Beatmap Nominators and Nomination Assessment Team members for the corresponding game modes.

osu!mania beatmaps are binded by a different set of spread rules defined in the osu!mania-specific Ranking Criteria. This is to allow for easier approachability to osu!mania mappers of different upbringings, and to account for the mode's rhythm construction.


This proposal's objective is to act both as a relaxation and also as a re-definition of how spreads are built and defined: instead of only being able to define spreads from the lowest required difficulty, mappers would now be free to define them from their topdiff. This model would come with various benefits:

  1. More difficulty-focused or complex sets due to song choice may not need to create oversimplified or throwaway Normal difficulties, which was a common request between experienced mappers in the survey.
  2. Mitigates the problem of "rigid thresholds" from the current model or a simple relaxation, specially for harder sets, by means of more homogeneous spread creation requirements:
    1. Example: For a set with a very hard U top difficulty (i.e. Blocko's We Want To Run), a difference between 5:00 and 4:59 draintime meant that 5 additional difficulties needed to be mapped to meet spread criteria. With the proposed model, this would be reduced to only 1 additional difficulty, as the spread can be constructed from the top difficulty. We hope this gives the "drain times" model a fairer look compared to the current implementation.
  3. Protects the influx of basic difficulties on the type of sets that currently provide this type of content (mainly simpler ENHI anime TV Size sets and ENHIX rhythmgame-core/VSRG sets), which expectably make up from the overwhelming majority of ranked sets in the mode.

All in all, we believe that this will is a final, sensible proposal to change the current Spread Requirements in the osu!mania RC, with views on improving the viability of more varied and higher quality ranked sets, yet always preserving the ecosystem for newcomers and enjoyers of larger/easier spreads. This (hopefully) definitive idea has been designed and revised by members of BN and NAT members of the mode, to ensure we've reached a consensus further supported by the data we've collected by the players' opinions.

Cheers! ♥

P.S: The PR in Github to implement the changes will be created in the next couple days after we figure out where to include these in the mania-specific RC page
Ery
Agree
Monheim
yes pls :)
RandomeLoL
This went through tons of read-throughs. And the drain times are appropriate with what was found on the data collected. Moreover, another spread choice as to not enforce certain "Lows" has been discussed and added to make spreads even more flexible. Overall, while it doesn't abolish them completely, it at least offers more flexibility from the mapper's side, as this can potentially mean not having to map difficulties that certain songs won't be very fitting of while the mapper can express itself by focusing on those difficulties that would both add value to the spread and that would be enjoyable all around!
-NoName-
Everything said here looks good.

At the most, the specific drain times could be moved around based on feedback, but at this point relaxation would certainly be a bit better than a complete removal, which is too much too soon.

Since there is a ton of keyboard VSRGs out there and people very frequently start osu!mania as a non-beginner to KB VSRGs in general, it should not be treated the same as the general RC.
Paturages
Yes please.

The "3/2/1 difficulties below highest difficulty" seems pretty reasonable for songs that guarantee enough complexity for a very hard map. However just to make sure, what about songs where complex mapping wouldn't necessarily fit? Fitting 3 difficulties below a "Hard" would seem a bit weird to me, although such cases should be pretty rare...
abraker

Paturages wrote:

Yes please.

The "3/2/1 difficulties below highest difficulty" seems pretty reasonable for songs that guarantee enough complexity for a very hard map. However just to make sure, what about songs where complex mapping wouldn't necessarily fit? Fitting 3 difficulties below a "Hard" would seem a bit weird to me, although such cases should be pretty rare...
My understanding you either start with Normal as lowest diff and fill up as desired, or start with highest diff then fill additional 2 down as required. So Normal and Hard should be ok
Topic Starter
Quenlla

Paturages wrote:

Yes please.

The "3/2/1 difficulties below highest difficulty" seems pretty reasonable for songs that guarantee enough complexity for a very hard map. However just to make sure, what about songs where complex mapping wouldn't necessarily fit? Fitting 3 difficulties below a "Hard" would seem a bit weird to me, although such cases should be pretty rare...
In case the wording isn't clear enough, in no cases anything below a Normal will be required for 0:00-2:30 draintimes, below a Hard for 2:30-3:15, etc. We can reword those lines if they're hard to grasp.

Abraker wrote:

My understanding you either start with Normal as lowest diff and fill up as desired, or start with highest diff then fill additional 2 down as required
Exactly this! (though in that particular case it'd be an additional 3, not additional 2)
RandomeLoL
Yep, as Abraker said, it's very important to take notice on the OR clause on every Drain Time range. This is the kind of flexibility that would help the end user with spreads, as one clause would normally help harder spreads while the other would still make it feasible for lower ones.
epic man 2
this is pretty cool honestly, im glad that the sort of discussion has managed to come to an ending point finally

gl to any newcomers going for ranked, as they say!
Ventilo le vrai
I was more into remove the spread requirements but I really like the solution that has been found. I'm all with it
Unpredictable
definitely support 100%
[LS]Ham

Paturages wrote:

Yes please.

The "3/2/1 difficulties below highest difficulty" seems pretty reasonable for songs that guarantee enough complexity for a very hard map. However just to make sure, what about songs where complex mapping wouldn't necessarily fit? Fitting 3 difficulties below a "Hard" would seem a bit weird to me, although such cases should be pretty rare...
3 difficulties below a hard would never be required because the RC states that for a mapset lower than 2:30 it would need either a Normal OR 3 diffs below highest diff. So you could have just NM and HD and that would be rankable if I'm understanding correctly.

Edit : Someone already said that lol, sorry.

Would just like to say this is a fantastic solution to discussions that have been popping up recently.
Roasted Chicken
Agree 100%

Don't mind me I'm just shooting kudosu stars

★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★
AutotelicBrown
Happy to see the inclusion of the "each keymode must provide a spread starting at least x difficulty levels below the highest difficulty" clauses as I felt that form of relaxation was kinda overlooked in the discussion threads and I personally think it's more significant than the time thresholds.

Anyway, very well written and I do agree with the proposal.
Maxus

Proposed RC wrote:

OR each keymode must provide a spread starting at least 3 difficulty levels below the highest difficulty.
So in my understanding, for example for the map like last resort which have Jakads' Lasting Legacy as highest difficulty, and you filled the 3 difficulty levels below that (Streamy's Heavenly, Ancul's Maximum, and Lolicide's Gravity) , that means in that particular spread, Lolicide's Gravity will be the lowest difficulty isn't it?

While i'm pretty much okay with that rule in 4K scenes, i'm unsure how this will translates into 7K scenes, because in general 7K mapping field have bigger difficulty spread than 4K (and most likely it will be even bigger in the future), so you probably can differentiate the ruling based on the keymodes itself (for example, using the 3 difficulty levels ruling for 4-6K , while 5 difficulty levels ruling for 7-10K).

It's just particularly a small concern from me mainly for 7K field because how their spread works are sort of different from 4k field, so probably it can be taken into consideration.

But otherwise i do agree with this proposal and do support this! ^^
Topic Starter
Quenlla

Maxus wrote:

It's just particularly a small concern from me mainly for 7K field because how their spread works are sort of different from 4k field, so probably it can be taken into consideration.

But otherwise i do agree with this proposal and do support this! ^^
Hi Maxus,

As you point out, 7k spreads could be the only "unsafe" part of the proposal, and they have brought up our attention in the BNG and NAT during the last couple days: we understand your concern.

Our insight into this, after discussing with 7k-focused BNs, was that 7k's problem is the lack of spreads as a whole; right now, no large spreads with U+ difficulties are getting ranked in 7k, as the keymode has always seemed to have a higher focus into marathons. Of course, we believe this is an issue and wish for a bigger influx of large spreads in the mode.

With this in mind, we are certain that the best idea is commiting ourselves as BNs into incentivizing mappers to push larger 7k spreads instead of trying to avoid them due to the larger workload they tend to carry. 7k lacks from a lack of content more than a lack of accesibility as of now, so we believe this change would still be a move in the right direction. Some changes into our group's activity requirements, such as increased activity points for nominating multi-keymode spreads, should also help with this, with more to come. Hopefully, we will also achieve the inclusion of more non-4k-focused BNs into the group (Kawawa and Asherz007 have already rejoined in the past week!).

We will do our best to ensure the increased liberty from this change and the one regarding hitsounding relates to an improvement of the 7k mapping scene, instead of being exploited exclusively for XU-only sets.
Maxus
That's great to hear, Thank you for still provide the solution for the problem from different angle! I appreciate there's more incentive to be done for 7K mapping field.
Mipha-
Strongly in support of this.
guden
100% support this, already discussed this in the BN server but yeah I think this is the best of both worlds and geared towards what the community would like!
Guldakh fanboy
Definitely yes.
AncuL
so good. o2jam maps please come out
Garalulu
Sure
Antipole
Yes.
vernonlim
I fully support this.
Madoka2574
I have some nitpicky concerns about the current wording.

For example, if the drain time is below 2:30, mappers are allowed to create a spread with the lowest diff starting from Normal OR map a spread with 4 diffs.

There are 2 points I want to mention:

1. How to define the reasonableness of spread? I can see that the RC can help people creating a spread like 5.0 - 6.0 - 7.0 - 8.0 under 2:30, but what if there's a spread like 7.0 - 7.1 - 7.2 - 8.0? There's no lowest diff limit so mapsets might tend to become higher SR, yet we have no rules or guidelines for Extra or above.

2. As for spreads at the lower SR, people can still map 3.0 - 3.2 - 3.4 -3.6 for those under 2:30 and call them ALL in Extra level, because I have 4 diffs already, so they can simply apply "OR each keymode must provide a spread starting at least 3 difficulty levels below the highest difficulty." and no diff of the spread needs to follow the RC in lower levels which can lead to chaos. I have seen the explanation that says Normal is definitely required but yeah I think it needs to be writen clearly in RC as people are not likely to browse the forum long time after new RC got applied.

I believe this is not what the proposed RC was meant for, and overall I'd agree on it, but I felt it can be more rigorous, because leaving all the judgement to BNs may cause some subjective issues.

Thanks for your efforts on improving the ranking field!
Murumoo
Agree
RandomeLoL
1. How to define the reasonableness of spread? I can see that the RC can help people creating a spread like 5.0 - 6.0 - 7.0 - 8.0 under 2:30, but what if there's a spread like 7.0 - 7.1 - 7.2 - 8.0? There's no lowest diff limit so mapsets might tend to become higher SR, yet we have no rules or guidelines for Extra or above.
This is quite a nice topic to bring up, and the reached conclusion from the BNG would be that it will be up to the BNs to intersubjectively analyze the spread and see the "sense" it might make. Even with that being a possibility, BNs should be rational when making harder spreads, which is mostly seen in higher Keymodes.

One thing that was proposed is, that for cases such as these, at least an Insane was added to still have a degree of "openess" so the mapset wouldn't just be for 60 people or so. But as that was very unfeasible to write in the RC as a general rule/guideline, the final assessment as said will fall upon the BNs themselves.

Your second concern would theoretically not be possible. Keep in mind that despite the RC changing, the actual criteria that determines whether something might be too big or too small of a gap is still under BNs discretion, the same as previously. Having such a reduced gap would definitely stand out, plus doing these kinds of things in mania tends to be harder due to the layering concerns and conflicts that normally pop out.

In short as to answer your concern, said rigurosity would hinder more than help. As these are things expected for BNs to check for, engraving it in the RC will only narrow the maps and, therefore, make a big difference on spreads if they're forced yet again to follow strict guidelines, even those that this reform precisely tried helping out.

And it is true! The thing with Subjectivity is a double edged sword. A double edge sword that has been applied to a plethora of other cases currently, in the past, and even on spreads with the current criteria. Hopefully that approach tries to encapsulate the "why" said rigurosity is lackluster on this iteration of the RC.

I cannot speak for EVERYONE that is, so take this take as just what has been taken to a conversation prior to the proposal's release as these same concerns are not new to the discussion, which is perfect to see for others to comment on!
lemonguy
yes
Topic Starter
Quenlla

madoka wrote:

I believe this is not what the proposed RC was meant for, and overall I'd agree on it, but I felt it can be more rigorous, because leaving all the judgement to BNs may cause some subjective issues.
I understand your concern! First of all, I think its important to be clear that judging what difficulty level a chart belongs to is already among BN's work, so we can definitely allow for this liberty because we will be checking a proper spread with the required difficulties is completed. As you say, Extra+ is a bit more tricky, but there's no way to define gaps there in the RC because the type of patterning doesn't change. Proper spreads through that difficulty range already exist without any specific RC binding them, and it's up to us to ensure they are properly spaced from now on as well. So yeah, a bit more work for the BNs to ensure the spreads are constructed correctly, in exchange to a bit more freedom for the players when creating them.
McEndu
Agreed.
Kyousuke-
yes.

totally agree
lenpai
glad to see my shower thought (top diff based spreading) get refined and put into proper context by everyone involved in this proposal

thanks for your efforts lets make this work!!!

@madoka as of late, sets that reach into very high difficulties have been reasonable and spread out well. Last Resort (as an example if it only had the top 4 diffs) was an outlier and will not fly should this proposal pass as it has a lot of overlapping diffs.
Linlime
100% Agreed!
Scotty
i still have some doubts but overall i think this is a very good middle ground solution so i'm in support of this too
Horrifying
Perfect!
abraker
Please sign in to reply.

New reply