Are huge sets a problem?

Hi there! I would like to use this blog to throw a few thoughts about difficulty spreads in general at the community.

Please keep in mind that this is my personal opinion on things and I intend to stir discussion about it, so some things may be provocatively worded or tinted by what I deem to be more important (as in, yes, that is intended). As many may have noticed, the amount of difficulties within the same set varies from very simple ENH to difficulty spreads to things like DJ Ozawa - Tokyo (Innovaderz Remix) or xi - Happy End of the World which top at over 20 difficulties in total. This article will take big sets with multiple difficulties per level as an example to discuss the pros and cons of even having them in the first place.

But first things first: many of you will probably remember the community outrage generated by a difficulty spread limitation rule proposed by ztrot a few months back (in case this went over your head, read up on what it was about here). The core ideas of it are as follows:

• Align sets in a linear and progressive ENHIX spread, where an additional Extra could be added under the placeholder term Ultra and be excluded from linear spread progression
• Difficulties were proposed to be limited to one per difficulty level per gamemode, so one Insane, one Extra, one Hard etc.

People with not-bubbled sets were asked to restructure their sets to obey these rules initially, which lead to massive community backlash due to the changes this would require for most contemporary sets, which e.g. commonly use for example 2 Hards (the more or less popular difficulty name Advanced comes to mind) or multiple difficulties of everything like Tokyo has. Structure-wise I will throw a few statements as hypotheses and arguments in here and then try to discuss these. Starting with:

When a difficulty is not necessarily required for a spread, all difficulties on around the same level will receive less plays overall.

What I mean with this is the following: Unless you are a really known mapper or your difficulty is a fundamentally different approach to the song, these Extras around the same difficulty level will be played based on popularity of the mapper and a few of them will get relatively low exposure in comparison to others.

Having multiple Extras/Insanes which are about equally difficult gives people no real incentive to play / choose either depending on their skill level, but rather on which of these mappers is the most known / popular / creative.

Examples:

A few of these Extra or Insane level difficulties may have been better off if they were put into a spread where they are actually a necessary element for progression throughout the set.

A different incentive to play a diff among higher difficulty levels (such as Insane/Extra or higher) may be that the difficulties are rhythmically or conceptually completely different approaches to the song and thus create an incentive for each of the difficulties to be actually played.

Examples:

Both of these examples feature difficulties which are vastly different takes on the song, while still being similarly difficult gameplay-wise. One of the sets is frequently at the top of most played so all difficulties are played a lot and approximately evenly, while the other set doesn't seem to get much attention. The differences between both of them are pretty apparent as the amount of plays is more balanced between the difficulties on No title while the Extras on TSUBAKI are also balanced, but each of them does not really get that much attention compared to the hardest Extra and the easiest one.

The point of choosing one difficulty to play based on a spread still stands, because the difficulties being vastly different is not really visible from the map selection ingame, even if you write that on the thread / map description on the forums. However, once you figure out that each of the Extra or Insane difficulties are vastly different from each other, you may go back to choosing to play the difficulty which you like the most (this assumes your first choice does not depend on what SR skill level you are at). This is highly situational, but different approaches which are equally difficult might end up at different SR, mostly because one of the difficulties is using more jumps and the other one maps more streams, or other similar reasons which distort the SR-algorithm a bit.

So... is having multiple similarly difficult difficulties to choose from a good or a bad thing for players? It probably is not too bad for players, since they can just ignore the additional diffs which they do not like and carry on with playing one of the difficulties which they like instead.

So... is this a good or a bad thing from a map-design standpoint? It probably is posing a problem, since you end up giving people no real incentive to play any of the difficulties based on their individual skill level and progression through the spread. Basically design-wise these additional difficulties seem obsolete and could be moved to a set that actually would need them.

So... are there other problems with having multiple difficulties of the same level? The next problem that can arise with these difficulties is not entirely related to the "same difficulty level" argument already discussed above, but rather to the total amount of difficulties.

Sets with a lot of difficulties heavily distort the worth of a nomination.

While getting a set bubbled is by no means a right, but rather a privilege, the worth of a bubble is varying vastly. An Easy/Normal spread on a 30 second song will require two people to nominate it, while on the other end of the spectrum Happy End of the World and Tokyo also require two nominations in order to get ranked. So the worth of one nomination varies anywhere between 30 difficulties of 6 minutes length to 2 difficulties of 30 seconds length. And a single difficulty of 5 minutes length needs three people to nominate it for it to be ranked.

The basic idea of a "marathon" set is, that it usually turns out to be too much to check for two people. Much more can "go wrong" on a set with this drain time as opposed to shorter difficulties where errors are easier to spot because there are less things you can screw up in a shorter song.

Except that essentially a marathon set with a full spread such as Happy End of the World can get ranked with only two people nominating it, which goes against the entire point of needing 3 people to check longer sets in the first place. A way to balance this further would be disallowing difficulties that are not necessary for a difficulty spread, where the attempt caused more or less community outrage. Taking "marathon sets need more people to nominate it" as an example it would make sense to determine how many people need to nominate a set based on total drain time of the set. With the general lack of 4 minutes or longer difficulty spreads in comparison to relatively short songs, determining such a limit can be really difficult.

It would impose further obstacles on people who are mapping 4 minute songs with an ENHIX spread in comparison to people who map just 2 minutes with the same spread. The general unhappiness about the enforced spread argument, combined with further obstacles will probably make 4 minute spreads more or less disappear. I am sceptical if this is the correct way to go about this, as the side effect seems to limit people who want to map longer songs. The main problems mappers seem to have with mapping longer songs for spreads are more or less the following:

• They take longer to make
• Less people are motivated to mod these
• Less people are motivated to nominate these

While no rules can fix any of those complaints (the argument of omitting parts of spread was being discussed on reddit, but that really is a weird option as it assumes your set does not have to stand on its own and beginners can just go and play other sets - which makes barely any sense from a map-design standpoint). Adding another restriction seems to only further discourage people from mapping more longer songs.

So... increase the amount of nominations based on amount of difficulties? This is the same thing as the last argument, except that it ignores the drain time of a difficulty. While it would limit sets that have lots of similar difficulties, it would also impact mapsets that have a proper spread from Easy to - let's say - something around 7.5 stars and sets with generally small difficulty gaps but linear progression.

So... limit the amount of difficulties, similar to what ztrot suggested? While that would get rid of "similar difficulty level"-spam-sets completely, it would force an inappropriately large spread on difficult songs in the way ztrot suggested it. In my opinion the main flaw in the initial draft of it is, that it assumes that Expert = Expert.

Everything above 5.25 Star difficulty is labeled as Expert as its own difficulty range, but 5.25 star difficulties and - let's say - 6.3 star difficulties are for the majority of cases largely different. So are 6.3 star and 7.4 star difficulties and 7.4 star and let's say 8.4 star difficulties. For this reason, a strict limit on the number of difficulties seems to impact the upper level of difficulty for really intense songs. For songs that require highly difficult maps, this assumption that Expert = Expert screws up the entire proposal. Also, a strict limitation like this seems to cause major community backlash. Currently there is an existing discussion about this in the Feature Request subforum (which has been originally denied 9 months ago), so if you are interested in this topic, join there.

A possible soft limitation version of this would be that linear spread is enforced, and additional difficulties that fall out of linear spread would require more nominations. However, such a limitation could be potentially too vague and has potential to create a ton of edge cases to consider in order to make this suggestion work anywhere near properly. The main thing that could confuse mappers about this would be the question "how many BNs do I need in order to get my creation ranked?!".

What do you think?

• Does having multiple difficulties around the same difficulty level cause any of the problems I see with them for you? Do you even think these are problems?
• Do you think the worth of a nomination needs further balancing? And which of the described ways would be favorable? Do you have a completely different approach from mine on this?
• Do we need more difficulty icons for the steadily growing difficulty of maps? What would be reasonable ranges if so?

I think this sums the things I intended to discuss with people up, while giving my opinion on all of the aforementioned. I discussed this with a few modders, mappers and players in private messages already, and each of these discussions went in slightly different directions so I look forward to what others think about this in a more public setting.

—Okorin