Xexxar asked in #modding the other day about ranking a 22min song with 3 acts in the style of The Unforgiving:

act1..........: diff1
.....act2.....: diff2
..........act3: diff3
act1.act2.act3: diff4 (marathon)

the idea's pretty cool and nobody seemed to have an argument that this wouldn't be fun to include in Ranked. after further discussion in #modding it was concluded that such mapsets would be acceptable as long as they are mapping a "contiguous album" (i.e. having a clear order or reason to be mapped together, no "Song Compilations" or "my favorite anime OPs") or a multi-part song like Xexxar's.

this isn't supported by current RC though, that's why I'm posting here.

problems and reasoning:

RC wrote:

Directly re-using your own Ranked beatmaps in other Ranked beatmaps is discouraged. This is to avoid unnecessary bloating of Ranked content.

this guideline suggests that the duplication of parts of the marathon into their own diffs would not be allowed. however I think this is a good case for a rare exception to made, as we've seen with The Unforgiving that there is value in doing this; leaderboards have a completely different kind of competition going on per-diff vs. the whole marathon, and providing the individual diffs gives ppl ranked maps to play if they weren't interested in the marathon.

RC wrote:

If you do not beatmap the last 20% of your beatmapset's audio file, it must be cut. The intro time is not included. This does not apply if more than 20% of the outro is occupied by a storyboard/video.

depending on how you read this rule, either:
- this rule isn't broken, because you map the whole audio file across the set (you read it as applying to one beatmapset)
- this rule would make diff1 and diff2 impossible because their outros are super long and you have to cut it (you read it as applying to one beatmap)

either way it's clear this would be a source of confusion or argument for a map like Xexxar's, so I think it's to be considered in a possible exception.

RC wrote:

If the drain time of each difficulty is... [that whole block]

in Xexxar's case he actually dodges this because each act is longer than 5min anyway. but if some diffs were short enough and high enough difficulty to conflict with this rule, it'd be a problem. my argument for why this is a valid exception to "ignore" the spread rules is that the mapset could be ranked with only the marathon already, and the only difference is that you would include extra difficulties for parts of the map, so you aren't subtracting any potential content from Ranked. in fact, as said in response to the first RC quote, this type of mapset gives more ways to play the songs included in the marathon, definitely not less.

suggested changes:

ok so it's rly hard to reword each of those above things to include exceptions for this, and it's so rare that I don't want to clog up the reading there just for this. instead, something like this would be its own "exception":

proposed RC wrote:

Single-difficulty marathon beatmapsets of contiguous albums or multi-act songs may include extra difficulties that split the marathon into single songs or acts. These extra difficulties are exempt from other rules and guidelines regarding duplicate content, audio file coverage, and drain time spread requirements. For example, Armin's The Unforgiving includes difficulties for each song in the album, as well as "Marathon" for the whole album.

I know that's not typical for RC, but I can't think of a better way to write this in.

How about you just split the "marathon" and make an own mapset of each song?

This reads like a cheesy way to skip the "full spread" requirements.

it's not skipping any requirements because the marathon is rankable as-is. and having the option of both playing each song alone or the whole marathon has been proven to be useful & interesting with The Unforgiving, that's why I'd say it's better than either only the marathon or separate sets for each song/act.

that being said, you're right some bad actor could try to do a kinda lame version of this setup where it does just feel like a cheesy thing to pack together--- that was the reason for having this exception only be made for albums or multi-act songs.

Yeh, so just rank the marathon as-is and make new sets for each of its songs? For me, this just looks like a way to avoid the spread creation for the individual songs.

mapper could just have alternate dl in desc with each song mapped separately in a different diff, having the same map exactly ranked twice on the same set seems pointless and just inflates pp or whatever

Loctav: that would go against the duplicate content rule (first one I cited in OP), assuming one of the diffs is the same as what's included in the marathon. if none of the diffs are the same as marathon, you can already do that and that's not really what this thread is about...

Nao: did you read OP or #modding about this? it's not pointless and The Unforgiving is a perfect example where the leaderboards are absolutely wanted and have made for cool & interesting types of competition, despite being duplicated from the marathon difficulty. that was the main inspiration for saying this should be allowed via some exception to the duplicate content rule

When it comes to duplicate content, I don't see how having the parts as diffs in the same set as the marathon differs from giving them their own sets. After all, we'd have the same amount of inflation in pp/content in both cases. You said in the OP that "this is a good case for a rare exception to made", and all of your reasoning (leaderboard competition, non-marathon content) for that applies to separate sets just as much as it does to the proposal.

As such, the exception should extend to giving parts of the marathon their own sets as well. They'd have to abide by normal set requirements, of course, so the option might not see much use, but they should still be allowed for consistency.

As for the proposal itself, I don't have a strong opinion in either direction. I did come up with a way to abuse it, though: if you had an album of songs that are short enough that they'd need a spread in separate sets and you wanted to map most of them, you could make a marathon just for the sake of the separate diffs and avoid making several spreads. It's a somewhat niche case, but there's an album I'd totally try to do that with if the proposal went through, so it isn't unrealistic by any means.

lewski wrote:

if you had an album of songs that are short enough that they'd need a spread in separate sets and you wanted to map most of them, you could make a marathon just for the sake of the separate diffs and avoid making several spreads

Correct if wrong, but I think you can do that now. If you want to do it now, you would just have one marathon diff instead of optionally additional diffs each containing sections of the album.

abraker wrote:

Correct if wrong, but I think you can do that now. If you want to do it now, you would just have one marathon diff instead of optionally additional diffs each containing sections of the album.

the separate diffs for each song are the entire point of the idea, that's how you could rank all of them as their own songs but without making spreads

maybe I need more clarification in the wording: the point is not to make new diffs and avoid spread rules just for the sake of it, this idea is supposed to let each difficulty have exactly the same content as a part of the marathon, so it's just two different ways to play the same maps in the same set.

and the reason I'd say splitting it across multiple sets is confusing is that you wouldn't be required to download them together, and the connection between them would be nonobvious, so a player would get a feeling like "wait, haven't I played this before?" and it's just not very intuitive overall. then, since those potentially <5min require spreads in their own set, you'd get a set where some diffs are new content and one diff is copy-paste from another ranked map... seems confusing to me

I did come up with a way to abuse it, though: if you had an album of songs that are short enough that they'd need a spread in separate sets and you wanted to map most of them, you could make a marathon just for the sake of the separate diffs and avoid making several spreads. It's a somewhat niche case, but there's an album I'd totally try to do that with if the proposal went through, so it isn't unrealistic by any means.

this isn't abuse, it's exactly a case I'm writing this thread for. the marathon is already rankable and all you would be doing is making each part of the marathon available as a diff as well.

lewski wrote:

abraker wrote:

Correct if wrong, but I think you can do that now. If you want to do it now, you would just have one marathon diff instead of optionally additional diffs each containing sections of the album.

the separate diffs for each song are the entire point of the idea, that's how you could rank all of them as their own songs but without making spreads

Except clayton is proposing restrictions on that

clayton wrote:

after further discussion in #modding it was concluded that such mapsets would be acceptable as long as they are mapping a "contiguous album" (i.e. having a clear order or reason to be mapped together, no "Song Compilations" or "my favorite anime OPs") or a multi-part song like Xexxar's.

Additionally, it's not as simple as bypassing RC to avoid spreads.

proposed RC wrote:

Single-difficulty marathon beatmapsets of contiguous albums or multi-act songs may include extra difficulties

You can't bypass it if you don't intend to include a marathon diff in the first place.

Therefore you are certainly not loosing potential spreads because marathons dont require spreads in the first place and because you can't compile random songs to avoid spreads either.

small amendments:

proposed RC wrote:

Single-difficulty marathon beatmapsets of contiguous albums or multi-act songs may include extra difficulties that split the marathon difficulty into single songs or acts. These extra difficulties are exempt from other rules and guidelines regarding duplicate content, audio file coverage, and drain time spread requirements, and are allowed to use the songs' or acts' names as difficulty names. For example, Armin's The Unforgiving includes difficulties for each song in the album, as well as "Marathon" for the whole album.

"difficulty" there should make it clear that the map is copied, and the second part is because I forgot that rule would be broken too

The RC doesn't use explicit beatmaps as examples so if an example is to be included it should be something like the one posted at the top like

act1..........: diff1
.....act2.....: diff2
..........act3: diff3
act1.act2.act3: diff4 (marathon)

but with better wordings.

any reason why not? I thought a real-world example would be helpful for understanding since the rule is a bit wordy.

Firstly, I agree with not adding a specific map as an example because it is generally discouraged to let a certain map be a precedent for future maps since many ranked maps have flaws that should not be taken as an example.

Also, I honestly disagree with letting such mapsets skip the spread requirements for the individual difficulties because it is indeed an easy way to avoid making a spread for those songs. If the drain time is more than 5:00 like in the case of Xexxar's map, it's fine because those diffs don't need a spread anyway. It would also work if the individual parts are longer than 4:15 and they're all Insane diffs for example. However, mapping a bunch of songs from an EP or something that are like 3 minutes long each with only Expert diffs and then putting them together in a compilation would allow you to skip Normal, Hard and Insane diffs for multiple songs which doesn't seem fair for players who want to play those songs on a difficulty that is accessible to them. This could be taken even further by making a TV size cut of each song and doing a compilation with those. Of course you can say that the marathon would be rankable by itself and this is just splitting it up into multiple parts, but that is exactly what makes the difference: If the individual songs are not long enough to skip certain difficulties, there is no reason to allow them to be ranked without a spread. The marathon on the other hand is much longer and therefore not required to have lower difficulties.

I agree with not adding a specific map as an example because it is generally discouraged to let a certain map be a precedent for future maps since many ranked maps have flaws that should not be taken as an example.

the way I've worded it here I think it'd be very hard to interpret as showing The Unforgiving has no flaws or w/e, it's just saying how the diffs are laid out. regardless I'll see if there's a good way to word this without linking to specific map then, I just feel like it's gonna get wordy again and defeat the purpose of adding an example

Of course you can say that the marathon would be rankable by itself and this is just splitting it up into multiple parts, but that is exactly what makes the difference: If the individual songs are not long enough to skip certain difficulties, there is no reason to allow them to be ranked without a spread.

I don't see the logic in this, the marathon is rankable by itself, yet adding more content for the songs to that set should make it unrankable due to "skipping" requirements?

This could be taken even further by making a TV size cut of each song and doing a compilation with those.

fwiw this is prevented by the restriction to contiguous albums and multi-act songs. but the proposal I'm making has nothing to do with individual songs'/acts' lengths

clayton wrote:

I don't see the logic in this, the marathon is rankable by itself, yet adding more content for the songs to that set should make it unrankable due to "skipping" requirements?

Yes, because the content that is being added has a different length compared to the marathon diff. The reason why spread requirements depend on the length of the map is that newer players who play Easy/Normal/Hard diffs tend to prefer shorter maps because they require less stamina and are just easier and less tiring to play. Therefore, the compilation of all the songs/parts is most likely too long for most newer players to enjoy because they're still learning how to play the game. However, this is not the case for the individual songs. Under the assumption that they are usually not much longer than 3-4 minutes, it would mean that they are short enough to be reasonably playable by less experienced players (of course the enjoyment of playing a map is a highly subjective matter but these are the main reasons why the spread requirements were changed so I'm just using those as an argument). Having only Expert diffs on relatively short songs would simply cut out a large part of the playerbase and go against the purpose of spreads to begin with. It's like making a 4:15 Insane diff and then adding a Normal that only maps the first 50% of the song. The Normal would only be additional content too but that doesn't mean it should be allowed to skip certain requirements (in this case the last 20% of the audio file needing to be mapped).

This could be taken even further by making a TV size cut of each song and doing a compilation with those.

clayton wrote:

fwiw this is prevented by the restriction to contiguous albums and multi-act songs. but the proposal I'm making has nothing to do with individual songs'/acts' lengths

Based on the current wording, it is not prohibited to do that. This example would be an even bigger contradiction with the current spread rules.

if you cut up songs in a contiguous album and piece them back together, it's probably not going to be contiguous or representative of the album. same goes for multi-act songs. that's why I'd think it's not possible. in any case this concern is just a small wording change away from being 100% clear so I don't wanna get too held up on it

---

imagine you're a player who can only play ENH. you see this album marathon set with IIIIIXXXXXXX (top diff being Marathon) and think something like "damn I wish there were lower diffs", because it's a little disappointing that you saw a song mapped but can't play it well.

would you feel less disappointed if instead that set were just X? now, in addition to not being able to play a song well because of its map's difficulty, you can't even play the song without slogging through a bunch of other IX stuff that you also can't play well. the ENH player gains nothing from the transition from IIIIIXXXXXXX to X, which is exactly what would be happening if my proposed exception is denied and a mapset following this setup has to be changed into only a marathon.

Having only Expert diffs on relatively short songs would simply cut out a large part of the playerbase and go against the purpose of spreads to begin with.

then, with the point I'm trying to get at via this example player, the problem isn't in all of these IX diffs, but the fact that the X marathon was allowed in the first place. I can already make something that "goes against the purpose of spreads" (in your opinion) by creating a bunch of 2 minute X maps, i.e. maps that should warrant a lower spread to allow more of the playerbase to play them, and then packing them all together in a song compilation because I'm too lazy for spreads. the playerbase that could not have played my 2 minute X maps is certainly not benefiting from me making this song compilation, nor are they missing out on less.

but I'm not here to talk about the purpose of spreads or if album/compilation marathons are going against that purpose, my argument is only this (using the example again cuz easier): the IIIIIXXXXXXX set should be rankable if and only if the X set is rankable. and the case here is that X set is already rankable so I'd say the IIIIIXXXXXXX should be too.

personally, it seems cool, but i don't really know how it'll go.

I still feel like you can't really compare a long map that consists of several parts/songs with the individual difficulties for each part/song. The gameplay experience is completely different and the length is also a deciding factor so I don't really consider this additional content because it's just the same thing being split into multiple parts. Aside from the duplicate content issue mentioned by other people before, by doing that you are changing the map from a marathon/compilation to a regular map of a single song. I think this change warrants a different approach regarding the spread. Having a single difficulty for the marathon makes sense because it's very long, but I simply don't see a reason why only one difficulty should be necessary for a fraction of the marathon when a normal mapset of that song would require an entire spread. The argument about additional content not changing the rankability of the map would make sense if the added content were similar to the marathon difficulty, but it's vastly different as explained earlier. The analogy with Songs Compilations already allowing to rank multiple songs with only Expert diffs is also not really logical because those maps only have 1 difficulty for the entire compilation, not one for each song. Again, you can't compare one long difficulty with several short difficulties because they're not the same thing.

Sorry if this is a bit off-tangent, but my concerns lie more with the pp-inflation this change will bring with it than spread rules

Allowing players to essentially gain pp for the same patterns twice opens up for even larger abuse of the pp system than we're currently faced with (assuming content from diff 1, 2 & 3 are put together into diff 4) and introduces content bloat into the Ranked Section

Though I can see the use-case for leaderboards on each song, I don't think this is the right direction to go because of this

A potential fix could be to restrict pp-gain to the marathon diff, but that'd open up another can of worms...

I don't think that is a problem. pp should measure skill, and if you have sufficient skill to do the map, then you deserve the pp. If it gives too much or too little pp, that's an issue that lies in the pp formula and not ranked.

Hypothetically speaking, having 20 copies of the same map ranked is going to earn you as much pp as as 20 different, varying, maps of similiar difficulty. There being 20 copies of the same map in ranked would be less of a pp issue and more of a quality issue.

Therefore concerns about pp-inflation with this proposal are not valid concerns

@abraker never thought about duplicate maps like that in respect to pp but I think you're right. and not to mention that a lot of farm maps' difficulty comes from a select few patterns that you can easily find as the most difficult element of plenty other farm maps already, so... a literal duplicate map actually wouldn't be very different. (for readers, remember that individual diff vs. whole marathon would require a whole different skillset though, despite being copies of the same map)

@Pennek not a tangent, thanks for bringing that up. aside from what abraker wrote, the kind-of-conclusion in #modding when this was first brought up was that limiting it to only albums and mulit-act songs would play a big part in preventing abuse like that

bringing it back today, i really don't know if this'll actually be a good addition. i think we should try to test this out, and see what a group of mappers, players, and bns/nats think.

now, another problem is how inconvenient it would be for mappers. mappers would need to apply the mods to the marathon and then to the diff. in nevo's case with his oral cigarettes comp, imagine if he had to copy individual diffs for each song. if you attempt to go to the discussion page for the map, and onto the timeline, you'll get a inconsiderable amount of lag, so it'll be near impossible to actually mod it

I agree with your suggestion. That rule can get more motivations for mappers & players. Some mappers like map marathons, but usually people too lazy for play long maps. Low playcount and non popularity can demotivate them to make marathons. Your suggestion can help for both sides: mappers & players.

you say this could be pp abuse but isn't the marathon harder than all of the diffs combined because of the consistency required for length? i think in really rare cases it could work, like 2 really easy songs and a hard one at the end, but really its kind of an insult when you consider how easy it is to just move to farming 2 similar maps for other songs lol, the effort in abusing this pointless. i don't think it's feasible that pp could be abused but just make the "marathon" diff unranked but with a leaderboard if it's really an issue...

Li Syaoran wrote:

bringing it back today, i really don't know if this'll actually be a good addition. i think we should try to test this out, and see what a group of mappers, players, and bns/nats think.

now, another problem is how inconvenient it would be for mappers. mappers would need to apply the mods to the marathon and then to the diff. in nevo's case with his oral cigarettes comp, imagine if he had to copy individual diffs for each song. if you attempt to go to the discussion page for the map, and onto the timeline, you'll get a inconsiderable amount of lag, so it'll be near impossible to actually mod it

this is why splitting the marathon into multiple difficulties isn't required.

Uniform wrote:

Li Syaoran wrote:

bringing it back today, i really don't know if this'll actually be a good addition. i think we should try to test this out, and see what a group of mappers, players, and bns/nats think.

now, another problem is how inconvenient it would be for mappers. mappers would need to apply the mods to the marathon and then to the diff. in nevo's case with his oral cigarettes comp, imagine if he had to copy individual diffs for each song. if you attempt to go to the discussion page for the map, and onto the timeline, you'll get a inconsiderable amount of lag, so it'll be near impossible to actually mod it

this is why splitting the marathon into multiple difficulties isn't required.

the answer is just waiting until mods are finished to create the marathon diff...

knightc0re and i were talking about this in a stream. one viable compromise for this could be handled in the same way as old TAG maps. Just like how TAG diffs were loved and subsequent diffs were ranked, why not just love the act/chapter diffs and rank the full marathon? that removes the concern of pp inflation. in order ot get pp you have to play the full map. very surprised this hasn't been proposed yet

maps aren't formatted in a munti-act manner very often so even if this manually has to be done it shouldn't have to be too often

also hurry up and pass this so that i can maybe rank i am shimmer

pp isn't an issue (community/forums/posts/7524471, https://osu.ppy.sh/community/forums/posts/7742593) and I don't think settling for a hacked up loved+ranked solution just cuz a few people are scared is the way to go

the amt of writing here arguing for what should be a clear net benefit for players and mappers (OP, community/forums/posts/7492799, https://osu.ppy.sh/community/forums/posts/7496209, community/forums/posts/7492085) is kinda ridiculous already and I've got nothing more to add. I'll ask some NAT to read it if it doesn't get any activity soon.

community/forums/posts/7499257 still needs to be addressed.

There's just a contradiction with the current spread rules which makes this change not coherent. The same logic could be applied to a full spread of a song by adding difficulties for the first half and for the second half of the song, it would technically also be additional content because the map would be rankable without them, but in the end they're only duplicates. I feel like it's not really necessary to have leaderboards on the individual parts anyway, people can still play them regardless.

Serizawa Haruki wrote:

There's just a contradiction with the current spread rules which makes this change not coherent.

that's the point, it's a special case/exception and does not fit into current spread rules, or at least I couldn't figure out a decent way to word it in there. for clarity when reading RC I'd at least put this very near the big block of spread jazz

Serizawa Haruki wrote:

The same logic could be applied to a full spread of a song by adding difficulties for the first half and for the second half of the song, it would technically also be additional content because the map would be rankable without them, but in the end they're only duplicates.

it could be applied there but that's not what I'm proposing. nobody asked for any random song/diff to be cut into parts and ranked separately, and I can't imagine people would enjoy playing it that much either...

Serizawa Haruki wrote:

I feel like it's not really necessary to have leaderboards on the individual parts anyway, people can still play them regardless.

I don't understand why lack of necessity is an argument against something that would be fun.

if a very specific written exception for the act/chapter splitting of a marathon/compilation beatmap into its specific parts is made, i'd be okay with allowing something like this

what i am worried about though is people taking this ruling if it was written incorrectly to create more "songs compilations" with 1 minute snippets of completely unrelated songs and stringing together giant sets of like 20 diffs + marathon with them, because people will do stuff like this, ploved has proven already that there is an appetite for sets like this.

if we do allow something like this, it has to be for contiguous works, like Varien's Valkyrie trilogy of tracks or The Unforgiving's 3 act stuff. it absolutely has a productive and engaging use case in those circumstances

clayton wrote:

it could be applied there but that's not what I'm proposing. nobody asked for any random song/diff to be cut into parts and ranked separately, and I can't imagine people would enjoy playing it that much either...

It doesn't matter if nobody asked for it, you can't make RC proposals solely based on a single example without considering other possibilities. Cutting a song into multiple parts can make just as much sense as cutting an album or multi-act, there are songs which are clearly divided into several parts that sound very different from each other, and they don't have to be extremely long either to be split up. But even without taking this into account, full spreads can also be made for 5+ minute songs, so you would essentially end up having several full spreads for different songs within the same mapset. Given how adamantly "content bloat" has been tried to prevent, this is rather contradictory.

There's even a specific guideline about it:

Directly re-using your own Ranked beatmaps in other Ranked beatmaps is discouraged. This is to avoid unnecessary bloating of Ranked content.

Ephemeral wrote:

if a very specific written exception for the act/chapter splitting of a marathon/compilation beatmap into its specific parts is made, i'd be okay with allowing something like this

agreed, the proposed bold part of the rule's got "contiguous albums or multi-act songs" in there

Serizawa Haruki wrote:

clayton wrote:

it could be applied there but that's not what I'm proposing. nobody asked for any random song/diff to be cut into parts and ranked separately, and I can't imagine people would enjoy playing it that much either...

It doesn't matter if nobody asked for it, you can't make RC proposals solely based on a single example without considering other possibilities. Cutting a song into multiple parts can make just as much sense as cutting an album or multi-act, there are songs which are clearly divided into several parts that sound very different from each other, and they don't have to be extremely long either to be split up. But even without taking this into account, full spreads can also be made for 5+ minute songs, so you would essentially end up having several full spreads for different songs within the same mapset. Given how adamantly "content bloat" has been tried to prevent, this is rather contradictory.

1. I (and the others from original #modding discussion) did consider "other possibilities", that's why the wording of the proposed rule is so specific
2. it is specifically for marathons because nobody asked or wants to be splitting up shorter songs like this, even if they are "multi-act"
3. it is specifically for single-difficulty sets because of this concern with existing spreads

I know that it goes against the content bloat rules, that's why it's a specific amendment/exception thing that spells out exactly which rules can be ignored in this case

I believe you're missing the point, which is that there's no difference between splitting a song and splitting an album from the point of view of how spreads work. Why should one be allowed but not the other? Similarly, why should only songs above 5 minutes apply? It's not true that only longer songs can/should be split up, there are actually other rhythm games where this is the case and it's actually not a bad idea because shorter maps are suitable for such games. I also don't see why it should only apply to single difficulty sets. The only reason is because the proposal is made to fit one specific example instead of trying to adapt to different examples. The RC applies to everyone, not only those who ask for certain changes.

It also shouldn't be allowed to ignore content bloat rules because it's no different from doing the same thing across multiple mapsets rather than within one. For example, if someone mapped an entire album and then someone else mapped one song from that album and asked them to make a GD, they wouldn't be allowed to take that part from their own map and rank it again in someone else's map, so why should they be allowed to do it within their own set. Either the rule applies to all maps or none.

Serizawa Haruki wrote:

I believe you're missing the point, which is that there's no difference between splitting a song and splitting an album from the point of view of how spreads work.

so I'm proposing to change up the rules slightly with this amendment---there will be a difference, if this gets written in RC. I'm a little confused why the current RC is being referenced to say that my proposed RC is not allowed (?)

Serizawa Haruki wrote:

Why should one be allowed but not the other?

the former is not wanted but the latter is (correct me if I'm wrong), plus a more broad allowance for this is easily open to abuse (community/forums/posts/7816440)

edit: by "former" i meant normal songs, and "latter" i meant albums or multi-act song

Serizawa Haruki wrote:

Similarly, why should only songs above 5 minutes apply? It's not true that only longer songs can/should be split up, there are actually other rhythm games where this is the case and it's actually not a bad idea because shorter maps are suitable for such games. I also don't see why it should only apply to single difficulty sets.

community/forums/posts/7817379

Serizawa Haruki wrote:

It also shouldn't be allowed to ignore content bloat rules because it's no different from doing the same thing across multiple mapsets rather than within one. For example, if someone mapped an entire album and then someone else mapped one song from that album and asked them to make a GD, they wouldn't be allowed to take that part from their own map and rank it again in someone else's map, so why should they be allowed to do it within their own set. Either the rule applies to all maps or none.

keeping the duplicate maps contained within a single set means that they are organized and clearly presented as duplicates in a place where they don't produce the harm that "content bloat" does. these duplicate maps don't mislead players, circumvent standards for other would-be-typical sets, or flood Ranked with identical maps(they keep only 2 copies, and with good purpose). the example you give is what will stay disallowed

so I'm proposing to change up the rules slightly with this amendment---there will be a difference, if this gets written in RC. I'm a little confused why the current RC is being referenced to say that my proposed RC is not allowed (?)

I'm not talking about the RC, it's about the logical standpoint.

the former is not wanted but the latter is (correct me if I'm wrong), plus a more broad allowance for this is easily open to abuse (community/forums/posts/7816440)

edit: by "former" i meant normal songs, and "latter" i meant albums or multi-act song

You can't really speak for the entire community regarding what is wanted and what isn't. There are many examples of songs where splitting them up makes just as much sense as splitting up albums or multi-acts and where this is wanted (e. g. beatmapsets/662260). It's arbitrary to consider mapsets with more than one difficulty abuse but not the example mentioned in the proposal because if splitting one difficulty is fine, it's also fine to do it on a whole spread. The difficulties are completely different from each other and therefore not more duplicated content compared to a single difficulty mapset, plus this could also just be circumvented by creating a new mapset for each difficulty, but the result is the same. The same goes for song compilations, if each song is cut in a proper way and the songs are related to each other (both of which are already required by the RC), I don't see why this would be worse. This type of "abuse" could also happen with the kind of examples you're probably having in mind, that's why I keep saying the proposal shouldn't be focused on specific examples as it only leads to bias.

keeping the duplicate maps contained within a single set means that they are organized and clearly presented as duplicates in a place where they don't produce the harm that "content bloat" does. these duplicate maps don't mislead players, circumvent standards for other would-be-typical sets, or flood Ranked with identical maps(they keep only 2 copies, and with good purpose). the example you give is what will stay disallowed

This makes no sense because all the discussions about content bloat took place on single mapsets without considering other mapsets. I actually do think that having different songs mapped within the same mapset is misleading because at first glance it seems like there are several different difficulties which cover the whole audio file and not only a part of it. Also how does this not flood the ranked section with identical maps? If maps which are only "similar" are deemed problematic, this should be even more so. There's no reason as to why the same scenario on two different mapsets should stay disallowed but not on one, there is no actual difference despite what you claim.

this proposal is focused on allowing a very specific example type of map because that's all I and the other participants of this discussion cared about, and the other mappers chiming in here so far have also not expressed desire to make it less specific. could it be made more broad? maybe, that's just not my proposal. I'm not trying to make this one general or all-encompassing because I've seen no support for that and I don't personally care, nor do I want to put in the extra thought to how exactly someone might abuse a broader version of this and come up with a more complex rule accordingly. it is "bias"ed toward the people who cared to speak to this proposal at all and our rough estimation of how other ppl would react based on a past example.

Serizawa Haruki wrote:

I actually do think that having different songs mapped within the same mapset is misleading because at first glance it seems like there are several different difficulties which cover the whole audio file and not only a part of it.

that's true, I was assuming people would name the difficulties after the individual songs or acts, at which point I would argue it's clear what they are (looking at The Unforgiving as an example again, since there seems to have been no issue with it after its approval). didn't write that into the rule though so next time I'm at a computer I'll update the proposed wording to factor that in

the rest of your post has already been addressed in this thread

Couldn't you just make one marathon diff, and then remap the individual section diffs so its technically not duplicate content?


The only thing stopping you then would be the ambiguous 20% audio thing.

Having a continuous flow between songs should allow to solve this problem over this way.

beatmapsets/1220237#taiko/2538450 is 52 minutes long and people don't want to play the easier parts first just to play one certain part, which can take up to 48 minutes.

I'd be okay with this suggestion if the acts as you call them have a reasonable length, the 20 % audio thing is written in the spirit that all difficulties in the set start and end at the same time so I'd not even apply that guideline to those sort of projects so long as the longest difficulty fulfills the guideline, the mp3 wouldn't need to be cut in my opinion.

This is an example of what this proposal is talking about and is in fact the inspiration for the original post.

beatmapsets/1275778#osu/2650704

As long as all songs/acts in the album are related, i see no problem with this and hope it will be implemented. Mapping whole albums takes a lot of effort (unless it's mapped by different mappers), i don't think someone would abuse that.

I actually don't think we need any change for this to happen. The "re-using" guideline can be broken under exceptional circumstances (as is the case with all guidelines) and this seems like a good reason to do so.

UberFazz wrote:

I actually don't think we need any change for this to happen. The "re-using" guideline can be broken under exceptional circumstances (as is the case with all guidelines) and this seems like a good reason to do so.

if this is the official position going forward, then this needs to be /explicitly/ stated in RC. we have seen how people's interpretation of the RC can vary, and this would benefit from not being subject to unfair vetoes and DQs just cuz it's not stated anywhere.

i think the proposal is good barring the example at the end (for reasons already stated above), so i would personally just consider:

Single-difficulty marathon beatmapsets of contiguous albums or multi-act songs may include extra difficulties that split the marathon difficulty into single songs or acts. These extra difficulties are exempt from other rules and guidelines regarding duplicate content, audio file coverage, and drain time spread requirements, and are allowed to use the songs' or acts' names as difficulty names.

the example was just to help explain inline, cuz at least to me it seems like everyone would understand this exception better as "you're allowed to make sets like The Unforgiving"(stretched a little)

doesn't matter to me if it's included or not, but like usual the wording and stuff can be figured out by ppl on the wiki side instead of here. i just want to get approval for this exception here :^)

I think this is an interesting proposal, but it would need to be extremely clear in the rule what is allowed and what isn't. The OP already does that by specifying contiguous albums and multi-act songs, but we can go further. I doubt this rule will be used very often, so it's worth being incredibly... ostentatious(?) about its implementation.

To indefinitely prevent abuse of this rule, I would add an exact length requirement, since many people might try to map EPs or LPs with very few songs in this manner, such as this 8 minute map of a mini-album. Obviously, this is not something that you would want split into 5 separate difficulties, all under 2 minutes each. I think that this length requirement logically carries over to multi-act songs for the same reason, and this would be fine since the proposal is obviously meant to be applied to these extreme cases of marathons (over 30 minutes) and not merely 10-15 minutes long maps. From what I know about music and the world, a full-length album is generally over 30 minutes, so that's where I would place the cutoff.

I don't think an example map is necessary if the rule is as clear-cut as I'm suggesting. Once somebody makes a modern example, people will look at that, but it's unwise to include maps in the ranking criteria without knowing how the game will look in the future.

Also, something may need to be added for storyboards (if some madman ever decides to do that), cause in my knowledge, maps don't end until the storyboard ends, so if someone maps the first 5 minutes of a 30-minute file with a storyboard, the player will be sitting there for 25 more minutes waiting for the results screen. The mapper would have to cut down the audio file and link it to the separate difficulties to avoid this. This might be able to be edited in the .osu file to end the map earlier than the storyboard ends, but I don't know enough about the subject to confirm or deny that, so I'll leave it for discussion.

With that, I propose a leaner version of the rule:

Single-difficulty marathon beatmaps of albums or multi-act songs over 30 minutes in drain time may include individual difficulties for each song or act. These extra difficulties are exempt from other rules and guidelines regarding duplicate content, audio file coverage, and drain time spread requirements.

A length requirement would make sense. What would you think of 20 or 15 minutes instead of 30? Nowadays, the runtime of an "album" isn't really in the definition due to the advancements of recording formats, and major charts & award corporations seem to have more lenient requirements for what counts as an album (see: UK Albums Chart and Grammy Awards). Multi-act songs are also usually not the length of whole albums, so I don't see a reason to be as harsh as to lean on the lengths of albums O_O

Nifty wrote:

Also, something may need to be added for storyboards... (cont.)

I thought that you can skip storyboards that keep going after the map ends? Unless you mean that storyboards should be cut so that players don't have to press skip?

30m is far too harsh of a cutoff, tbh i dont think any minimum length works as there are plenty of albums out there that are ~20m-30m and have 15-20 songs depending on genres i dont see a problem with a map like axer’s having individual difficulties?