forum

[Proposal - osu!catch] Removing the second part of the Overall Difficulty guideline

posted
Total Posts
10
Topic Starter
Sanyi
Hi everyone,

for clarification - I am talking about the following guideline:

If a difficulty uses a lower Approach Rate than one or more difficulties in the difficulty level below it, then the Overall Difficulty should be equal to that of the highest Overall Difficulty value in the previous difficulty level.

I modded a map yesterday, where this part of the guideline applied, which made me think once more that this guideline is useless and potentially confusing. Here is the link to the discussion: beatmapsets/1577302/discussion/3240525/general#/2639688.


Why I think the guideline is useless

The only thing that OD affects in osu!catch is score, meaning OD has no effect on the map itself. It affects the score at certain increments, depending on the CS and HP value. More about score calculation can be found here: wiki/en/Score/ScoreV1.

In many cases, applying said guideline doesn't affect the score, making it useless. But even in cases where the score gets affected, it doesn't affect gameplay and therefore is irrelevant for map quality. There may be a few players who care about score farming, but the potential score loss from removing this guideline is negligible, considering how rarely that guideline ever applies and actually affects the score.


Why I think the guideline is potentially confusing and should be removed

On the one hand, it contains a term that is not clearly defined: "previous difficulty level". What does that mean? Let's say we are talking about an Overdose - is the previous difficulty level the next lower Overdose, or is it the Rain? This uncertainty is confusing and makes applying the guideline harder. If you need an example for a discussion about the guideline, check out the beatmap discussion link from earlier.

On the other hand, it just clutters the RC, making it even less beginner-friendly. The RC shouldn't be any longer / more confusing than necessary.


So my proposal: Just remove said part of the guideline.
Nokashi
First of all, first.

Now to cut to the chase. To Preface, AR being Equal to OD is a not a must but rather a convention used since the beginning of catch mapping to somewhat find a consistent use for the OD Setting. Our main concern and goal for handling the OD is to make sure the harder the difficulty is, the more score it gives. So hypothetically. A 6/7/8/9 AR but 5/6/7/8 OD is also perfectly fine since the goal is achieved, harder diffs give more score. Now I did see that you also did some number crunching on the scoring formula so correct me if im wrong here.

The main usage of this guideline is to disallow a lower difficulty level map to have more score than a higher Difficulty Level map.

But What's a difficulty level?
The Ranking Criteria consistently presents the term "Difficulty Level" as the set of a certain stage of the Difficulty progression. On the other hand, a singular difficulty is referred to as a "beatmap" or a "difficulty". For example, All the Overdoses of the set are one difficulty level, But a singular overdose of that set is a "difficulty" or a "beatmap". The guideline essentially, as its the only place this guideline breaks once in a blue moon, doesnt want a rain to have more score than an overdose, which is Valid.

In the beatmap discussion you linked, any OD larger than 9 is valid. The lower Difficulty level is the rain. The overdose can have any OD that's higher than the Rain. Now if you wanted to be a bit more streamlined, it should also have less than the harder ones and more than the easier ones. That's the ambiguous part, but its also not a necessity. The guideline is not broken as long as the OD is higher than 9.

To Conclude, I have personally enforced this guideline maybe once or twice since I started modding, but it feels like a nice fine line to have so a rain doesnt give more score than an overdose, which is the guidelines intention. If we can find a middle line to make it clear that score should increase consistently per difficulty level, then I'm all up for this, but currently the only thing that protects this mechanic is this guideline.
Spectator

Nokashi wrote:

So a rain doesnt give more score than an overdose
...
If we can find a middle line to make it clear that score should increase consistently per difficulty level, then I'm all up for this, but currently the only thing that protects this mechanic is this guideline.
Agree with Nokashi, especially about these parts. OD is there to properly represent the actual difficulty of the maps and removing the second part of the OD guideline would result in making the OD spread unorganized.

Also higher difficulty having lower AR than the lower difficulties is rather a rare case, which is the reason why I think that the possible confusion occurred by the guideline is little to none compared to the benefit it gives
JBHyperion
Removing the guideline and allowing the user to set whatever value for Overall Difficulty they like can create a misleading representation of the beatmap's actual difficulty, for no benefit other than arbitrarily making the map award a different score.

"Previous difficulty level" is clearly understood to be the highest rated beatmap in the difficulty level one below the current one. For example, "Easy" is referred to as the "Lowest level of difficulty" in the Ranking Criteria here and here.

The Ranking Criteria contains so much information ("cluttered", you might say...) because people wanted everything clearly defined, so that they can point to the RC to justify things in their modding. The absence of this guideline historically introduced more confusion than what we see now, which is precisely why it's there in the first place, so removing it is counterintuitive to me.

Each time this use case has come up (which is rare), it gets explained and we all carry on.
ZiRoX
As previously said, the "difficulty level" concept is presented elsewhere in the Ranking Criteria, so confusion shouldn't arise on that part. Even if it wasn't, it is possible to come to the same conclusion: when the guideline says "the highest Overall Difficulty value in the previous difficulty level", that's implying there could be multiple ODs in a single difficulty level. Obviously, a single difficulty can't have more than one OD, so the only logical conclusion is that "difficulty level" refers to a group of difficulties of similar difficulty.

In any case, if you can propose any better alternative for the guideline - because the discussion is going in the direction that the guideline should be kept, at least in spirit (which I also agree with), bring it on.
Mniam
As a creator of the difficulty linked, to me it was also unclear if I need to have a higher ar than lower star-rated overdose or the rain (like it's said in the map's discussion)

My idea to get rid of further confusion about this isn't removing the guideline, but linking the definition of "difficulty level" in the guideline (if it isn't done yet of course. I can't check the ranking criteria while writing this).

Anyways thanks to Nokashi I know I'm still free to use od9.2 in this case.
Topic Starter
Sanyi
Let's clear up some things.


JBHyperion wrote:

Removing the guideline and allowing the user to set whatever value for Overall Difficulty they like can create a misleading representation of the beatmap's actual difficulty, for no benefit other than arbitrarily making the map award a different score.
This might be a misunderstanding about the proposal itself, since I intended to keep the first part of the guideline:

Overall Difficulty should have the same value as the Approach Rate. This is just a standardised value, as Overall Difficulty does not affect gameplay nor does the number of fruits a spinner has. Overall Difficulty only affects the maximum score of a difficulty.

So the mappers wouldn't be able to use whatever value for OD they like.


Nokashi wrote:

In the beatmap discussion you linked, any OD larger than 9 is valid. [...] The guideline is not broken as long as the OD is higher than 9.
That's false, since the guideline states that it should be equal, not equal or higher. Setting the OD above 9 in this example still may follow the intention of the guideline, but since the intention isn't stated anywhere, your interpretation is arbitrary to a third party, which is not how the RC is supposed to be applied.


Spectator wrote:

OD is there to properly represent the actual difficulty of the maps
Is that really the case though? In other gamemodes, OD has an actual utility - it determines how accurate you have to click. Just because OD doesn't serve that purpose in osu!catch, it seems arbitrary to me to give it your meaning, especially since we already have a value across all modes that is supposed to represent the difficulty of a map - Star Rating.


Spectator wrote:

Also higher difficulty having lower AR than the lower difficulties is rather a rare case, which is the reason why I think that the possible confusion occurred by the guideline is little to none compared to the benefit it gives.
That's an illogical conclusion, since it can be benefitial only when it gets applied, so the "rare case" argument also goes against the benefit, not only against the confusion.

-----

The main (and only) argument for keeping this guideline seems to be, that it is supposed to prevent lower difficulties from giving more score than higher difficulties. I was aware of that reason before I made this proposal, yet I am confused by how much people defend this guideline for exactly that reason, when it actually isn't that effective at ensuring that.

Just take the following map discussion as an example: beatmapsets/1284565/discussion/-/generalAll#/2315655. In this case, BNs had to break the guideline for the exact reason the second part of the guidline exists, yet it failed to be applicable in this case. If the guideline was actually important, why wasn't the guideline fixed after that incidence? Maybe the value of the guideline is just overestimated in this discussion. An adjustment like:

If a difficulty would have a lower max score than one or more difficulties in the difficulty level below it, then the Overall Difficulty should be set to a value where the difficulty gives no less score than any difficulty in the previous difficulty level.

immediately comes to mind to prevent that issue in the future.

Also why isn't the RC stating the reason for the OD guideline, when it is that important? Yet another case of neglection which makes the more sense the less value you attribute to the guideline.

-----

I didn't make this proposal though to improve this guideline. I made this proposal, because I think focusing on a case that rarely ever happens and doesn't affect gameplay in any way is a waste of time. If you disagree with that mentality, then that's the way it is. If this discussion continues to be about adjusting/improving the guideline, I have nothing more to add to the discussion than what I already wrote.
Topic Starter
Sanyi
Ok so after my initial reaction I thought about it again.

Since everyone besides me (at least the ones who contributed here) wants to keep the guideline, but some people (including me) think that the current version is improveable, I think we should take this opportunity to improve it.


As I already pointed out, I think the guideline should include the reason to why it's there in the first place. So altering the first part of the guideline would be nice (bold part is the addition to the current version):

Overall Difficulty should have the same value as the Approach Rate. This is just a standardised value, as Overall Difficulty does not affect gameplay nor does the number of fruits a spinner has. Overall Difficulty only affects the maximum score of a difficulty and should therefore be scaling with the Approach rate to properly display the difficulty level with the maximum score.


Also a change on the second part would be nice to make it more functional:

If a difficulty would have a lower maximum score than one or more difficulties in the difficulty level below it, then the Overall Difficulty should be set to a value where the difficulty has a higher maximum score than any difficulty in the previous difficulty level.


Both suggestions are only meant to show the change in content and probably need an overhaul by a more proficient english speaker than me.

Let me know what you think.
ZiRoX
I don't agree with either of the changes. Reasons below:

First proposal:
The Ranking Criteria is already lengthy as it is, so adding the reasoning for the rule would only make this worse. You yourself said this:

Sanyi wrote:

On the other hand, it just clutters the RC, making it even less beginner-friendly. The RC shouldn't be any longer / more confusing than necessary.
In any case, after reflecting on some similar discussion that is taking place for the General RC (community/forums/topics/1425629?n=1 ), I think one good idea would be to make a "companion page" for the RC, in which the reasoning behind rules/guidelines is specified, so the RC itself focuses on what your map must/should follow and nothing else. That way the RC becomes more to the point and less lengthy, and whoever is interested in the reasoning behind stuff (because, let's be honest, not everyone will be) can refer to that companion page.

I'm all up for helping with this if people would consider it a nice addition (I don't wanna work on something that will end up being majorly useless), or maybe even leading it, considering I've been here for long and actually handled the RC and some of its major reworks for a few years.

Second proposal:
As indicated in the ScoreV1 documentation, the max score changes happens at fixed values of the sum between CS, AR and OD. Since CS and AR affect gameplay, they should be considered fixed when it comes to "manipulating" the max score of a diff, leaving OD as our only "control lever". Since the break points are quite spaced (6,13,18), in most cases it will be impossible to manipulate OD in such a way that it actually changes the max score so that it meets the proposed guideline.

Side note: After some brief testing, apparently only CS and OD are relevant for osu!catch in this regard. But the point still stands.
Greaper
I'm personally not against some form of removing the 2nd part of the OD guideline (as the reason for why this exists is mainly because ScoreV1 exists). Although giving people the freedom to put whatever is actually something I really do not want. A better approach is to completely remove OD from the gamemode itself and completely ignore its value like is done for ScoreV2. Because of this, I think leaving the rules as is should be done.

After a short discussion with the osu!catch NAT about this we've decided to move this proposal to finalized as this would overall create more complexity on both the RC and in-game. If you want to make the RC less lengthy then I would suggest creating a new proposal to only change the wording of this. I will be moving this proposal to finalized as not many people seem to be behind the proposal.
Please sign in to reply.

New reply