Tae wrote:

AsrielDr33murr wrote:

The boost would scale based on SR. Even if it didn't, a 10% boost on a normal diff that gives 23pp wouldn't be all that game breaking.

I never said EZ would be worth more than no mod, I just said EZ plays would get a slight boost to pp.

Okay but SR is a rather arbitrary system to base that off. Considering EZ also reduces SR immensely, your system doesn't mean very much, as you'd end up having to base it off a case by case scenario.

Literally any boost you're trying to give it makes it a little bit overpowered. Once you can read low AR it's free pp to some extent, due to it reducing every other difficulty setting too, if we are going by your idea. Perhaps think about this a little bit more first.

Tae wrote:

And yet that's still countered by the fact that using EZ reduces the difficulty in the first place, and thus makes it easier to do so after a certain point.

AsrielDr33murr wrote:

No, EZ is a million times harder to read. It takes a lot of skill to read EZ. I'm better than most and even I'm not that good at it

Tae wrote:

AsrielDr33murr wrote:

No, EZ is a million times harder to read. It takes a lot of skill to read EZ. I'm better than most and even I'm not that good at it

Practice makes perfect is my point. After a certain point it definitely becomes easier to do, like most things in this game.

Tae wrote:

DT objectively increases the difficulty of the map (AR/OD/HP), as does HR, EZ does not. As for HD, well that was rebalanced lately, wasn't it now?

AsrielDr33murr wrote:

Different things are difficult for different players. For some, HR is piss easy; for some, DT is impossible; for others, HD might be unreadable.

abraker wrote:

The issue with low AR is more complicated than visible note density. If you take AR 3 with 10 visible notes, then certain arrangements of those notes will make it easier or harder to read. Consider a stream pattern vs a random placement of notes. The stream pattern would be a lot easier to read because it is far more predictable than the random arrangement of notes.

AsrielDr33murr wrote:

abraker wrote:

The issue with low AR is more complicated than visible note density. If you take AR 3 with 10 visible notes, then certain arrangements of those notes will make it easier or harder to read. Consider a stream pattern vs a random placement of notes. The stream pattern would be a lot easier to read because it is far more predictable than the random arrangement of notes.

The pp system can differentiate streams from jumps tho. So it wouldn’t be too much of a problem. Just would take more code

abraker wrote:

AsrielDr33murr wrote:

abraker wrote:

The issue with low AR is more complicated than visible note density. If you take AR 3 with 10 visible notes, then certain arrangements of those notes will make it easier or harder to read. Consider a stream pattern vs a random placement of notes. The stream pattern would be a lot easier to read because it is far more predictable than the random arrangement of notes.

The pp system can differentiate streams from jumps tho. So it wouldn’t be too much of a problem. Just would take more code

It is not about differentiating patterns, but making sure it can handle anything you throw at it. It is not wise to have the algorithm judge difficulty via detection of certain patterns. Instead, it is better to have it calculate by measuring things, properties, that all patterns have. This allows to make testbenches which construct any significant pattern arrangement using those measurements as values in parameters, and would allow us to know for which extreme values the algorithm gives unwanted results. Otherwise, sometime in the future people will find some pattern which works best to increase pp which we couldn't think of now.

Endaris wrote:

From what I recall, AR below 8 already receives a boost in pp unrelated to SR.

f32 approachRate = beatmap.DifficultyAttribute(_mods, Beatmap::AR);
f32 approachRateFactor = 1.0f;
if (approachRate > 10.33f)
	approachRateFactor += 0.45f * (approachRate - 10.33f);
else if (approachRate < 8.0f)
{
	// HD is worth more with lower ar!
	if ((_mods & EMods::Hidden) > 0)
		approachRateFactor += 0.02f * (8.0f - approachRate);
	else
		approachRateFactor += 0.01f * (8.0f - approachRate);
}

_aimValue *= approachRateFactor;

Endaris wrote:

https://github.com/ppy/osu-performance/blob/master/Src/Processor/Standard/StandardScore.cpp

f32 approachRate = beatmap.DifficultyAttribute(_mods, Beatmap::AR);

f32 approachRateFactor = 1.0f;

if (approachRate > 10.33f)

	approachRateFactor += 0.45f * (approachRate - 10.33f);

else if (approachRate < 8.0f)

{

	// HD is worth more with lower ar!

	if ((_mods & EMods::Hidden) > 0)

		approachRateFactor += 0.02f * (8.0f - approachRate);

	else

		approachRateFactor += 0.01f * (8.0f - approachRate);

}



_aimValue *= approachRateFactor;

Endaris wrote:

Wrong, HD just gives a bonus twice as high for low AR compared to nomod.
It naturally doesn't award me any points but that is mainly because I don't have impressive AR7 aim plays in the first place.