In reference to: https://osu.ppy.sh/community/forums/topics/756407

For reiteration, after quite a bit of time in the BNG and QAH, it was clear to me that there was a rather large gray area surrounding how to enforce breaktime guidelines. Since the rc rules and guidelines are written under the parameters of BPM=180, when the bpm is higher or lower than 180, the lines become very blurry about what is considered “exceptional circumstances.”

With that said, i believe the following should be amended in the taiko ranking critiera:


The added exception of counter-intuitive break times are to ensure that if a break is not inserted, the reason for not doing so is for the sake of quality and flow. In other words, putting a break would otherwise not play well and not match the music correctly. Furthermore, the exception of allowing more continuous mapping if the part is more forgiving allows mappers to use breaks such as 2/1 in a kantan as oppose to 3/1 so long as the 2/1 break(s) occur frequently. That choice is equally as forgiving as a 3/1 or longer break.

Along with that, there was some other “housekeeping” things i think should be added in the RC with times changing in taiko:



The reasoning behind the slider velocity tweaks is that at higher BPM, beginning players need ample time to react to the notes on the playfield. The speed at which the notes move becomes quite fast at 240, which is why it is a guideline, and then definitely unreasonable for beginners at 300, which is why it is a rule. As for the 1/2 patterns being changed from 4 notes to 5, using 4 notes as a limit indicates that the song will always contain an offbeat rhythm and that is not the case. 5 notes is much more versatile and can create a more fair spread with Muzukashii in the cases that the Kantan is not included.

special thanks to ono and oko for their opinions and help : )

Not toooooooooo sure on the base SV suggestions but it's all multiplicative in the end so it's probably ¯\_(ツ)_/¯

Breaks can be handled case by case, there's no need of a guideline, imo. No one want maps with unfitting or forced breaks and that's why mappers should be able to reason why they're putting shorter/less breaks. Also, the more wordy a guideline gets, the more confusing it gets to read.

About SV, is it okay for 299BPM, but not for 300? Better put it as a guideline.

And as a person who had no rhythm sense and spend a lot of time learning to play with Kantan and Futsuu, I support base SV regardless of BPM: usually faster songs have harder maps than the average, so low difficulties for those songs can be used as a "training" to learn to read faster maps while being light in note density.

Of course if the BPM go nuts (like 300BPM), small adjustments can be done, but always based on a guideline.

And there's the case that one could map a 150BPM as 300BPM, so ¯\_(ツ)_/¯

thing with sv sound like unnecessary duplication, only the guideline thing seems enough because no ones uses normal velocity for lowers difficulties on high bpm maps, it would be pointed as a issue already

Using rest moments less frequently is acceptable if either the pace of the music makes rest moments counter-intuitive or if the continuously mapped part is overall more forgiving to the player. This sentence is still too vague in my opinion.

May we should say something specific. For the exact norm like,


What I write here is only my opinion. Do not serious with it.
But I think that we should give some words to describe the word "less frequently "

I think anything faster than 240/250 bpm for a kantan MUST have lower slider velocity, for futsuu there could be some arguing but kantans are supposed to introduce players to the game. How can you be introduced to the game when there are notes flying by faster than even some more experienced players can read? It just doesn't make sense. A lot of the time I see people simply shrug off guidelines without exhaustive reasoning. I think to have it truly be always enforced in kantans it needs to be a rule. Also if the futsuu is the lowest diff in the set I think it should also have a similar rule.

Dake: stuff's already there, i believe the guidelines we currently have are good reminders to have, just that breaking them seemed to be so hard to reason that most people just didnt

usually when that happened explicitly softening the guideline is the way to go, hence being being more specific for this sort of thing seems harmful (Tyistiana)

The SV ~rule~ should be a guideline at most to allow being scalable

given how there's practically no debate going on, does this mean people dont care either way or are fine with the changes?

Leaving this here for like 1 more week

yes, my apologies for not contributing to the conversation for a hot minute, forgot to bookmark this on my pc

saying that these amendments are unnecessary additions is a bit false imo, although Dake is correct that break times are and should be handled case by case depending on note density and bpm, that is basically exactly what these guidelines say-i think it also clearly establishes that breaking these guidelines is ok given certain circumstances, which i'm sure newer mappers don't exactly understand

as for the SV rule, i think it's fair enough to put that as a guideline yeah

https://github.com/ppy/osu-wiki/pull/1839

didnt include the guideline about 300+ bpm because now that it's not a rule, it has the same meaning as the 240bpm one... i think? maybe i'm not reading it right, so let me know

Alright, guess it's time for us to draw. We basically share most of what Dake has written.

The clarifications to the guidelines are unnecessary. They are redundant, and simply make the guideline itself even more wordy and difficult to understand specially for non-English speakers. In "guideline" it is alraedy assumed that reasonable judgement is applied anyway, the so called "gray area" has never existed, but has been artificially created by people who were incapable of interpreting the RC. Making those changes to appease such people is not an intelligent move if you ask me.

With that in mind, my opinion on the rest are as follows:
- Lower SV on >240BPM: can go with it. But you must reword it to "You should use" and remove "simply" (I can suggest these on github directly). Also should not be tied to a specific difficulty and instead tailored to most lower difficulties (MINIMUM to Muzukashii). Remember that this would be a guideline and you would not need to fully apply it (e.g. you simply use 1.4 in a 250 bpm Muzukashii)
- The second guideline would be obsolete with the reword of the first (also, 300 is a very arbitrary number to go with)
- Third guideline is obsolete with the modification of the first
- Same with the fourth
- Can go with the change from 4 to 5 notes in the Futsuu-as-lowest-diff

Guess that's it

The elaboration on the guidelines should stay / be made more simple if it's too complex - the main issue that the OP takes with them is that the community treats them as rules and doesn't even think of reasons to break them so example sentences of what constitutes for breaking guidelines don't really do any harm in this case

the gray area you talk of is indeed people not using their brain after reading the criteria, but if using your brain after reading it was intuitive people would be doing that - if they aren't it can't hurt to point them in the right direction i think

lower SV should be a guideline at best, but its use of direct bpm shit is confusing with the "lol nerds look here it's all 180 bpm" in the beginning of guidelines - i dont think there's a neat solution to it though
the 300 being an arbitrary number is just as arbitrary as 180 or 240 is, it's just a threshold lol


otherwise agree

Basically what Raiden said. Regarding the rest moment guidelines, they are actually self-explanatory as they may be violated under exceptional circumstances and thus reasonable judgement is required. But if we still want to make it clearer for what exceptional situations that the guideline could be broken, I think "Using rest moments less frequently (up to 32/1 to 40/1) is acceptable if the pace of music becomes significantly faster to support longer patterning or if the continuously mapped part is overall more forgiving in drastically slower songs. Otherwise, it may put too much strain on beginners." is more concrete and easier to understand while keeping the terms "intermediate players" for Muzukashii and "players of this difficulty level" for Oni.

About the base SV on >240 bpm songs, the guideline to be added could be amended as follows and be put in the general section instead:


And I agree that the 1/2 limit for Futsuu-as-lowest-diff could be made more lenient from 4 to 5 for versatility of patterns.

can go with what Nardo wrote too, go ahead

Since bpm is talked about in relative manners in the general part, the 240 in that wording should be ~240 because the other thing is approx 180

----

where would you add the "Using rest moments less frequently (up to 32/1 to 40/1) is acceptable if the pace of music becomes significantly faster to support longer patterning or if the continuously mapped part is overall more forgiving in drastically slower songs. Otherwise, it may put too much strain on beginners." after each of these instances ? cause that's even longer than what was proposed or im just tired and reading shit wrong

As long as specific conditions are clearly included, it is fine to have slightly longer sentences.

Here is a summary of the proposed guideline amendments:





EDIT: Did some rephrasing to make them less chunky.

i do not really see the point of setting a hard line at 240 bpm, i mean 220 bpm kantan with 1.4x sv is still really fast for noobs (having watched noobs play 220 bpm kantans they complained about this), maybe the guideline can be more like "at higher bpms, sv should be adjusted to maintain easy readability for new players." rather than a specific bpm...

Nardoxyribonucleic wrote:

Regarding the rest moment guidelines, they are actually self-explanatory as they may be violated under exceptional circumstances and thus reasonable judgement is required. But if we still want to make it clearer for what exceptional situations that the guideline could be broken, I think "Using rest moments less frequently (up to 32/1 to 40/1) is acceptable if the pace of music becomes significantly faster to support longer patterning or if the continuously mapped part is overall more forgiving in drastically slower songs. Otherwise, it may put too much strain on beginners." is more concrete and easier to understand while keeping the terms "intermediate players" for Muzukashii and "players of this difficulty level" for Oni.

Or maybe, we can just skip the "exceptional circumstances" crap and just go straight by the discretion of the nominators. You as mapper do not have to clarify an "exceptional circumstances" in at least 80% of the time, because reviewing the map more than once helps to understand the mindset of the mapper more. In either case it will be fine; the map either gets disqualified because a certain diff lacks of rest time and receives post-DQ modding or the respective nominators don't have to rely on that max. 20/1 nonsense and judges the rest time distances properly and leave the difficulty/map in their natural setting.


also the BPM limit seems unreliable, because like Nao said it doesn't matter to a new player if the map was 240 or 220 - it is fast in the end. I agree that you should say that for higher BPMs you should use 1.2x, including to consider how dense the difficulty is, because a 240 BPM Kantan with half of the density from a 200 BPM Kantan would be easier in the end and may not need 1.2x after all.

holy this blew up, fantastic

i can agree with the majority of these changes, but i have 1 small note:
nardo wrote specific values of 32/1 to 40/1 and while i do enjoy those values, i think it's best we still don't use those numbers in case those need to be broken as well; leaving it open-ended is better in the long run imo to keep it in the guideline nature else it turns into a rule imo

everything else seems fair enough, though i am wary on allowing the term "higher bpm" since that is a little bit too broad imo, i understand that density is a factor to consider as Stefan said but i still think there is a bpm value at which becomes almost objectively fast for beginners, and it's possible that it's 220, or maybe a little bit lower. but "higher bpm" i think is too broad.

thanks for y'alls input C:

Gonna agree with people saying that a specific number for what "high BPM" is should not be added. If some beginners struggle to read those, some others use them to get into high BPM reading. When you go from low difficulties with purposely lowered SV to higher difficulties that use normal SV, the difficulty gap is incredibly high.

Raiden wrote:

300 is a very arbitrary number to go with

This number is as arbitrary as values used in break guidelines for the like of 180 BPM. Instead of adding the 32/1 and 40/1 precisions i would rather remove the 16/1 and 20/1 ones (as well as the 180 BPM precision). Writing something more general like "You should regularly insert rest moments (1/3 or longer) depending on the song's BPM and density. Continuous mapping may put too much strain on beginners." would open up a bit mapper's AND nominator's AND qat's minds about the topic, forcing actual "reasonable" judgements.

Though this change implies simplications on other guidelines and rules.

Nofool wrote:

Gonna agree with people saying that a specific number for what "high BPM" is should not be added. If some beginners struggle to read those, some others use them to get into high BPM reading. When you go from low difficulties with purposely lowered SV to higher difficulties that use normal SV, the difficulty gap is incredibly high.

incredibly good point, did not think about that.

edit: i saw it before, but like nofool said, i don't see how 300 is anymore arbitrary than 180-300 is a good cut off (i think) of what a majority of players no matter their skill level begin to struggle with readability, but with nofool's point that i quoted, that's why it's a guideline.

Nofool wrote:

Raiden wrote:

300 is a very arbitrary number to go with

This number is as arbitrary as values used in break guidelines for the like of 180 BPM. Instead of adding the 32/1 and 40/1 precisions i would rather remove the 16/1 and 20/1 ones (as well as the 180 BPM precision). Writing something more general like "You should regularly insert rest moments (1/3 or longer) depending on the song's BPM and density. Continuous mapping may put too much strain on beginners."

Except 180 is a factual average (maybe it is a little less, 150-160) of what is commonly mapped in this game, and the numerical values (16/1 and so) of the rest moment guidelines were actually tested with players of all skill levels, being these the ones which yielded optimal results.

While I would fundamentally agree on your second point, the current exact numerical values exist to streamline the guidelines, as leaving it to "depending on song's BPM and density" leads to newer mappers not knowing what to do and struggling to create a normal spread, something that we do not desire at all.

Nofool wrote:

would open up a bit mapper's AND nominator's AND qat's minds about the topic, forcing actual "reasonable" judgements.

Though this change implies simplications on other guidelines and rules.

Reasonable judgements are already being applied even with the streamlined exact numbers, though. They are merely a rough estimation for newer mappers to be able to discern what is excessive and what is not when mapping lower difficulties.

Raiden wrote:

the numerical values (16/1 and so) of the rest moment guidelines were actually tested with players of all skill levels, being these the ones which yielded optimal results.

I never heard of these tests, i believe this is something important and interesting enough to be made public ? Or is it already ? If not you can't really say "tests were made and this is the result" without linking the actual study. Showing everyone these informations would greatly help and speed up this discussion.

Had to fiddle with my Drive folder to find it hahaha

https://docs.google.com/spreadsheets/d/1JgDafWlWphaF9f578gECD5s7Ippxkg_vUHdf3pBRlN8/edit?usp=sharing

Blank boxes means "no issue with this", PC means playcount

Nofool wrote:

This number is as arbitrary as values used in break guidelines for the like of 180 BPM. Instead of adding the 32/1 and 40/1 precisions i would rather remove the 16/1 and 20/1 ones (as well as the 180 BPM precision). Writing something more general like "You should regularly insert rest moments (1/3 or longer) depending on the song's BPM and density. Continuous mapping may put too much strain on beginners." would open up a bit mapper's AND nominator's AND qat's minds about the topic, forcing actual "reasonable" judgements.

In fact, those specified lengths of continuous mapping in the guidelines serve as important reference values so that beginner mappers could have clear clues to make proper spreads accordingly. BNs and QATs have been applying their reasonable judgement in this regard to ensure respective difficulties are not made over-the-top or too underwhelming.

^ My point is that these "reference values" came out of nowhere from an external point of view, and you don't create references out of nothing.

Now about Raiden's spreadsheet :
- Map link died so we can't see the 1/1 chains length => i can't judge players results about them yet
- Chains were tested in Kantan and Futsuu only ("all diffs: 1/1 patterns" and "Diff 2: 1/2 chains" respectively) => no data for Muzukashii

Making suppositions about your results in Kantan (BPM~180?) :
- Out of 6 testers, none had issues with 1/1 chains (except for you mom literally, no joke intended).
- If these chains are shorter than 16/1 and 20/1 then we don't know tester's limits and thus can't judge what the correct length would be.
- If these chains are equal or longer than 16/1 and 20/1 then we already know that these values are not straining limits for beginners and thus not necessarily relevant.
We would need to know/ask them at what point (chain length) they felt exhausted and then deduct a proper shorter length.

I suppose you know more since you talked to them personnaly. Though, to be honest, the data sample is too small to draw conclusions, but definitely a good way to go.

removed the rest moment clarifications and bpm/sv additions from the pr until a consensus is met

lowest diff futsuu thing is still in though

well looks like a consensus won't be met, get a pt. 3 if you want it to happen