It's good that u brought this up so quick peppy.

As technology is getting more advanced by time maybe there are other "tools" we should consider an fix on as early as possible.

There are certain companies out there that are currently developing eye-tracking software. What are you thought on this seeing as this may be an issue?

Wow the world is gonna end because oa f 900 pp play omg!!! Please panic erase all the pp

for real though I'm sure many people (myself included) would be willing to make a more balanced pp system, I doubt separating the leaderboards will make any difference in terms of people's satisfaction with both the pp system and the touchscreen issue

peppy wrote:

I do not see this as a good solution when at the end of the day the majority mostly care about making sure scores like this do not appear higher than tablet scores.

I really doubt there exists a solution that would ensure this never happens without completely separating the two, due to the way the current pp system deals with jumps. Having to consider touchscreen when balancing for mouse/tablet spacing and vice versa would just throw everything off. (we've seen this already, back when TAG4 maps were unranked btw please rerank)

Also, just a suggestion, but if touchscreen had their own leaderboards, they could have different mechanics as well. Maybe taking the streaming mechanic of osu!droid(aiming with one finger and tapping elsewhere with the other hand) and putting it into osu!std could make touchscreen more approachable. What's your opinion on that?

lolarisan wrote:

Maybe there could be an option on your profile where u select touchscreen and it will automatically be sent to another leaderboard?

Yeah but people could just not select that option and have it submitted regularly, and to prevent that there would always have to be people looking at the leaderboards and all plays for that, which is kinda impossible imo

lolarisan wrote:

There are certain companies out there that are currently developing eye-tracking software. What are you thought on this seeing as this may be an issue?

I felt sick for three days after trying to play osu! with eye-tracking. Two other people who tried almost vomited on my desk. Keep in mind that when playing osu! your eyes tend to read ahead. Trying to target hitcircles directly is hard. Sliders is puke (eyes are not made to smoothly track).

I completely agree with basically everything mentioned in the OP. The best time to solve this issue isn't now, but after the release of lazer. Knowing that it WILL be addressed is more than enough for now, while the issue isn't hugely damaging. It's better to inform touchscreen players of proposals sooner rather than later to help them decide whether the peripheral is something they'd still enjoy using.

Your temporary fix of applying a mod will work in the short term, but I think the only true solution to the issue is creating a separate gamemode for touchscreen when lazer drops. Touchscreen is a completely different game that needs different maps and different gameplay features and by extent, it's own leaderboard (like taiko/ctb/mania).

hi-mei wrote:

But for temporary solution I can suggest to nerf how much PP is coming from short maps, it is just not ok that 20 second map gives 900 pp. Its just wrong.

Completely incorrect. IF that map was a 20 second 300bpm full screen spaced deathstream, it definitely should give 1000s of pp. This shouldn't even have to be said.

fieryrage wrote:

I doubt separating the leaderboards will make any difference in terms of people's satisfaction with both the pp system and the touchscreen issue

Well I mean, it would completely solve the issue people have with touchscreen. PP issues are an unrelated topic.

What I don't really get is that changing the pp system doesn't work.
By nerfing difficultyspikes(Example: 5* map with one or two short 7* jumpspikes)
and improving pp gained from streammaps or technical maps,
the pp mappers, that would have mapped short longjump difficultyspiking maps would start to map different styles, effectively nerfing touchscreen indirectly.

Letting length, average difficulty, and accuracy play a greater role in the current system would probably shift the meta to a more balanced set of skills required for pp.

A few friends of mine who reached the top 1000 fairly fast, all have only one skillset in common, and that's long fast jumps.

No streams, barely any technical ability, no low AR reading, no complicated tap variations, barely any consistency on non-jumpmaps.

It's sad to see those playstyle slowly go away, they add so much depth to the game, but in the new mapping meta it's very rare to see stuff that isn't 1:28 jumps with the odd triple in there.

thank you peppy for listening to the community

Tune wrote:

What I don't really get is that changing the pp system doesn't work.
By nerfing difficultyspikes(Example: 5* map with one or two short 7* jumpspikes)
and improving pp gained from streammaps or technical maps,
the pp mappers, that would have mapped short longjump difficultyspiking maps would start to map different styles, effectively nerfing touchscreen indirectly.

Letting length, average difficulty, and accuracy play a greater role in the current system would probably shift the meta to a more balanced set of skills required for pp.

A few friends of mine who reached the top 1000 fairly fast, all have only one skillset in common, and that's long fast jumps.

No streams, barely any technical ability, no low AR reading, no complicated tap variations, barely any consistency on non-jumpmaps.

It's sad to see those playstyle slowly go away, they add so much depth to the game, but in the new mapping meta it's very rare to see stuff that isn't 1:28 jumps with the odd triple in there.

I completely agree.

hi-mei wrote:

I think its a long awaited time for a meticulous review of how PP score is calculated.

There is a very tangible difference between 4 min 7* HDHR map fc and 20 seconds HDHRDT haitai FC.

This thing HAS to be addressed at some point, I truly hope that the solution will be finded and we gonna shift from TV-size farming meta to something a bit different.

I am not a math-guru, but I can say that the formula of PP calculation is struggling when it comes to sliders.

Maps with lots of sliders gives less score, doesnt matter if the actual difficulty is higher than the common TV-size anime maps.

So the main thing that comes to my mind is that PP system doesnt care how you hit sliders, how much input sliders are having on the player during the gameplay, how it affects reading and such.

But for temporary solution I can suggest to nerf how much PP is coming from short maps, it is just not ok that 20 second map gives 900 pp. Its just wrong.

I think this is more or less what people should be focusing on. The problem is not strictly a "touch screen problem" because stupid plays like this can and will happen on any input device if a map is short enough.

Trying to separate the playerbase with different leaderboards is both unnecessary and abusable. The bigger problem lies in the system not weighing consistency properly which creates a "spam retry meta". The touch screen players can thrive on this since their accuracy varies a lot and all it takes is "one good run" to set the score.

The only real solution I can think of is introducing a soft limit on how much pp can be earned before a certain map length is reached. Anything below 60 seconds or so should start seeing some diminishing returns that scale exponentially with how short the map is.

You might say that this is a "band-aid fix" because "what if freedomdiver just gets better at playing longer maps" but let's be real, if you can sustain 85%+ acc on OD11 with touchscreen for over a minute then you probably deserve to set the score.

happening or not, im glad to see some action going on

peppy wrote:

lolarisan wrote:

There are certain companies out there that are currently developing eye-tracking software. What are you thought on this seeing as this may be an issue?

I felt sick for three days after trying to play osu! with eye-tracking. Two other people who tried almost vomited on my desk. Keep in mind that when playing osu! your eyes tend to read ahead. Trying to target hitcircles directly is hard. Sliders is puke (eyes are not made to smoothly track).

That is true. But this sickness might just be temporary like motion sickness. People might play enough to get used to it like changing from mouse to tablet.

And as for your eyes reading ahead, people might change a slight delay in the software making is go a bit slower when u look and then learn to hit after the circle.

Lets make osu! Great again together

I'm glad there will be a solution in the future. A touchscreen mod sounds like it might be a patchwork solution for now that doesn't break everything..

Grae wrote:

I think this is more or less what people should be focusing on. The problem is not strictly a "touch screen problem" because stupid plays like this can and will happen on any input device if a map is short enough.

Trying to separate the playerbase with different leaderboards is both unnecessary and abusable. The bigger problem lies in the system not weighing consistency properly which creates a "spam retry meta". The touch screen players can thrive on this since their accuracy varies a lot and all it takes is "one good run" to set the score.

The only real solution I can think of is introducing a soft limit on how much pp can be earned before a certain map length is reached. Anything below 60 seconds or so should start seeing some diminishing returns that scale exponentially with how short the map is.

You might say that this is a "band-aid fix" because "what if freedomdiver just gets better at playing longer maps" but let's be real, if you can sustain 85%+ acc on OD11 with touchscreen for over a minute then you probably deserve to set the score.

This is an emotionally fueled solution. Long maps are widely considered overrated, it's not an issue with map length. It's not an issue with low accuracy scores. It's an issue with touchscreen.

To imply that a touchscreen player can't do insanely hard maps for a minute straight is ludicrous. If a normal player can manage it, so can a touchscreen player.

This thread should be about the issues with touchscreen IMO, it's hard to discuss the standard PP system when so many people are trying to shoehorn in a fix for touchscreen. There's no system that can compare touchscreen scores with normal scores because they're completely different games.

Hey peppy, I guess this is a good opportunity to ask you about this:
What are your thoughts on moving the osu! pp system towards a multi-category kind of system that allows for different objectives (such as aim, control, speed) like the guys at http://osuskills.tk/ did, rather than a single leaderboard?

I think it would be a great refresher for both playing and mapping. Instead of mapping for ''pp'' like we usually see nowadays, mappers would have a lot of diversity to do so and maps that focus on other abilities such as reading would be far more valued as well.

Perhaps you've considered something like this already but I'm just curious on your thoughts. Personally I enjoy trying to improve my scores on that website from time to time, even if it's far from finalized. Thanks for looking to improve the pp algorithm!

please dont nerf touchscreen i was gonna get myself one for christmas ;(

MariahCarey wrote:

Hey peppy, I guess this is a good opportunity to ask you about this:
What are your thoughts on moving the osu! pp system towards a multi-category kind of system that allows for different objectives (such as aim, control, speed) like the guys at http://osuskills.tk/ did, rather than a single leaderboard?

Perhaps you've considered something like this already but I'm just curious on your thoughts. Personally I enjoy trying to improve my scores on that website from time to time, even if it's far from finalized. Thanks for looking to improve the pp algorithm!

pp already works like this under the hood, but aim/acc/speed are accumulated together for each score and presented as "pp". If they can't be separated for the current system, I don't know if we can expect them to be for the new system. If they can be separated, that should be done before implementing a new system to check which pp system actually performs the best in optimal conditions.

nibs wrote:

Touchscreen players can only exploit a small selection of maps as far as I know (ones that are almost entirely jumps/no streams, like yuima-ru world and granat), so maybe just have a thing in the ranking criteria that prevents maps that are:

very short and have very high star ratings (yuima-ru world is 43 seconds and it's 6 stars)

maps that are easy until the last 10 or 15 seconds and then it's just jumps (use something like this to see https://rarelyupset.com/oppai/)

from being ranked? Sorry if this isn't helpful, just my thoughts on this entire situation.

These criteria aren't set in stone. Touchscreen players can do longer maps, it's just easier to do the shorter ones. The only things really required are huge jumps and a low amount of short streams. It's also unfair on mappers to disqualify their maps just because a subset of people want to use a peripheral that isn't suited the the performance system.

jesse1412 wrote:

This is an emotionally fueled solution. Long maps are widely considered overrated, it's not an issue with map length. It's not an issue with low accuracy scores. It's an issue with touchscreen.

To imply that a touchscreen player can't do insanely hard maps for a minute straight is ludicrous. If a normal player can manage it, so can a touchscreen player.

This thread should be about the issues with touchscreen IMO, it's hard to discuss the standard PP system when so many people are trying to shoehorn in a fix for touchscreen. There's no system that can compare touchscreen scores with normal scores because they're completely different games.

Its not ok that AIM-based short maps are considered more valuable than long maps that are focused on reading/stamina/memorization.

Short maps has to be nerfed, ill say it again: 20 second map can NOT give 900 pp. No matter how you set the score, it is just makes people who has one strong ability to be higher (lets say Rafis) than people with other strong ability (lets say -GN).

The rewarding system should consider all the players equally. For now we are living in a racist world where "aim players" are better than "reading-focused players" and "stream-focused players".

Talking about touchscreen players - they will NEVER get a good score on any map longer than 1 min with lots of bursts, streams and fast sliders.

The problem was never in touch screen players, the problem is that the system thinks that jumps are way harder than anything else (streams, fast sliders, low ar maps etc).

hi-mei wrote:

ill say it again: 20 second map can NOT give 900 pp.

And I'll say it again. If you have a 20 star 20 second map and someone fc's it, you'd have to be braindead to argue that it should be worth less than 900pp; thus 20 second maps CAN give 900pp.

hi-mei wrote:

Its not ok that AIM-based short maps are considered more valuable than long maps that are focused on reading/stamina/memorization.

The rewarding system should consider all the players equally. For now we are living in a racist world where "aim players" are better than "reading-focused players" and "stream-focused players".

Different argument that I completely agree with. It also, sadly, has no easily implementable solution.

hi-mei wrote:

Talking about touchscreen players - they will NEVER get a good score on any map longer than 1 min with lots of bursts, streams and fast sliders.

True but you're relying on the touchscreen player to only use touchscreen. If the player uses keyboard/mouse/tablet to do the first few stream sections of a map and then swaps to touchscreen for an ending jump spike, it becomes possible.

hi-mei wrote:

The problem was never in touch screen players, the problem is that the system thinks that jumps are way harder than anything else (streams, fast sliders, low ar maps etc).

Completely wrong, did you read the OP and watch the video provided? It explains why. If a tablet player could set the scores that freedom diver set, they'd definitely deserve 800+pp. That play is FUCKING INSANE, regardless of whether jumps are slightly overweighted. PP is mostly not the issue, touchscreen is the issue.

osu!droid

jesse1412 wrote:

And I'll say it again. If you have a 20 star 20 second map and someone fc's it, you'd have to be braindead to argue that it should be worth less than 900pp; thus 20 second maps CAN give 900pp.

you have some misconception of what i meant there, we have a clear comparison of maps with kinda same star rating, right? so lets say, we have these two freedom dive hdhr and haitai 3mod. the first one demands all the possible abilities to get an FC (reading, speed, aim, stamina, memorization), its 4 min and u get a 800 pp for it.

the second one is ONLY focused on ONE aspect of the game, which is aim, and its literally x12 times shorter than freedom dive map.

x12 times, and its only requires one ability to have (FD requires at least 5).

so even if the HAITAI was 4 min with the same jumps all over the table, it would still be less valuable than FD score, cause it requires less anyway.

I know it sounds kinda strange but... Ill never be ok with the thought that less effort gives more rewards. its just not how life works.

True but you're relying on the touchscreen player to only use touchscreen. If the player uses keyboard/mouse/tablet to do the first few stream sections of a map and then swaps to touchscreen for an ending jump spike, it becomes possible.

Cmon, we both know its impossible.

Completely wrong, did you read the OP and watch the video provided? It explains why. If a tablet player could set the scores that freedom diver set, they'd definitely deserve 800+pp. That play is FUCKING INSANE, regardless of whether jumps are slightly overweighted. PP is mostly not the issue, touchscreen is the issue.

I still disagree, if the system would work normally, there wouldnt be a 20 sec 900 pp scores on a map that was made in literally 30-40 mins with no effort at all. People are ranking short maps because of the score in first place, not cause its fun and shit.

Talking about touchscreens again in particular, its just allows AIM-aspect to dissapear from the actual diffuculty, but it struggles with other aspects of the gameplay (reading, streaming etc).

Nerfing the touchscreen will not resolve the issue because people will find other techniques to abuse JUMPS, the issue is more in the map, in my opinion.

Shouldn't it be possible to reduce the aim pp awarded when the cursor teleports?

hi-mei wrote:

you have some misconception of what i meant there, we have a clear comparison of maps with kinda same star rating, right? so lets say, we have these two freedom dive hdhr and haitai 3mod. the first one demads all the possible abilities to get an FC (reading, speed, aim, stamina, memorization), its 4 min and u get a 800 pp for it.

the second one is ONLY focused on ONE aspect of the game, which is aim, and its literally x12 times shorter than freedom dive map.

x12 times, and its only requires one ability to have (FD requires at least 5).

so even if the HAITAI was 4 min with the same jumps all over the table, it would still be less valuable than FD score, cause it requires less anyway.

I know it sounds kinda strange but... Ill never be ok with the thought that less effort gives more rewards. its just not how life works.

This still isn't a better argument in my opinion. Freedom dive hdhr gives less pp despite having more balanced requirements because those requirements aren't as hard (subjectively) as the individual aspect of aim in haitai. Balanced maps maybe could be slightly buffed, but that's just shifting the problem further down the line. Eventually there's going to be a cut off point where balanced maps aren't anywhere near as hard as the imbalanced map, there's no solution to that.

hi-mei wrote:

Cmon, we both know its impossible.

Pressing pause button and swapping peripherals isn't hard at all, let alone impossible.

hi-mei wrote:

I still disagree, if the system would work normally, there wouldnt be a 20 sec 900 pp scores on a map that was made in literally 30-40 mins with no effort at all. People are ranking short maps because of the score in first place, not cause its fun and shit.

Talking about touchscreens again in particular, its just allows AIM-aspect to dissapear from the actual diffuculty, but it struggles with other aspects of the gameplay (reading, streaming etc).

Nerfing the touchscreen will not resolve the issue, the issue is more in the map, in my opinion.

You can't blame the map like that. What if a song literally can't support 1/4 notes without overmapping? What if the song is also short? Is that the maps fault? No of course it isn't. Are you going to blame the song then? Granat fits into both categories comfortably, as an example. The issue isn't the maps, it's touchscreen being incomparable to tablet/mouse.

hi-mei wrote:

Cmon, we both know its impossible.

Let's go back in the time let's just say....2 years maybe. Have you ever thought there will be a 900 pp play, let alone it has been achieved with a touchscreen, a peripherial people didn't really consider using to play osu! back in the days. You are assuming there will be no changes regards to touchscreen and how it is working right now. The future is, and has always been, not clear and what is maybe impossible to do now, most likely will be possible to do in the future.

hi-mei wrote:

I still disagree, if the system would work normally, there wouldnt be a 20 sec 900 pp scores on a map that was made in literally 30-40 mins with no effort at all. People are ranking short maps because of the score in first place, not cause its fun and shit.

We all can agree ppv2 is not the best system, and we all know it has its flaws. It is not a fresh system after all, and with all the systems out there, as the time goes by, people might and will find ways to abuse it. But being this one of the main issues, as mentioned in the OP, it is not about a whole new pp system right now, nor fixing/patching some aspects of the current one, but the nerf of the touchscreen as a device, as it neglects the aim aspect of the pp. Of course if you have an idea about patching something in the current system that does exactly that, then sure you can propose but assuming from your arguments you wanted to change the system from the git-go which is not the main priority right now.

jesse1412 wrote:

This still isn't a better argument in my opinion. Freedom dive hdhr gives less pp despite having more balanced requirements because those requirements aren't as hard (subjectively) as the individual aspect of aim in haitai.

I respect your opinion, but I have to disagree with this. It's just that flow-aim isn't considered as important as jumps. We should not even compare those plays, one is a play with the mindset of rhythm game, the other one clearly not. And I thought we had to "click circles to the beat".

hi-mei wrote:

you have some misconception of what i meant there, we have a clear comparison of maps with kinda same star rating, right? so lets say, we have these two freedom dive hdhr and haitai 3mod. the first one demands all the possible abilities to get an FC (reading, speed, aim, stamina, memorization), its 4 min and u get a 800 pp for it.

the second one is ONLY focused on ONE aspect of the game, which is aim, and its literally x12 times shorter than freedom dive map.

x12 times, and its only requires one ability to have (FD requires at least 5).

so even if the HAITAI was 4 min with the same jumps all over the table, it would still be less valuable than FD score, cause it requires less anyway.

I know it sounds kinda strange but... Ill never be ok with the thought that less effort gives more rewards. its just not how life works.

Probably the entire community agrees upon the fact that short, super aim-heavy maps are overweight. But this is not the issue brought up in this thread. It's the fact that an aiming peripheral that requires less practice and completely circumvents aspects of the game that your everyday tablet or mouse has to take into account is currently allowed and thus gives people an advantage in a way that was never "intended" by the global rankings and skews them because of that.

You can't create a pp system that balances a touchscreen and at the same time fairly judges tablets and mice, no matter how much you tweak with length or whatever else. The system was made for tablets and mice, and it won't adjust for anything else because it's a waste of time and is only a temporary solution in the end at best.

jesse1412 wrote:

Pressing pause button and swapping peripherals isn't hard at all, let alone impossible.

How do you imagine that?

So you play, lets say, Apparition, you fc the streams, then you pause, then you switch to the touchscreen to hit jumps, then you see the stream again and the jumps right after it.

You switch again, dont forget to hit that orange thing that appears after pauses.

And if you miss, you gonna start again and again and again and again. Its super inefficient and way less abusable than it is now. Why? because you have to be as good as your touchsceen aim to play these insane streams and patterns.

To summarize our convo:

I do agree that touchscreen is a "cheat", the "cheat" that can only work on specific kind of maps - very short jumpy map, other than that - its super bad.

I do believe that rewarding system should be more balanced, more rewarding for people who likes others type of maps (i.e. -GN, Gayzcmgee, idke). For now, the mapping meta are based around playing short maps with 1/2 spam, that affected the entire generation of mappers and players, and to be fair it has to be addressed at some point.

Its like in any RPG or MOBA, when one class is dominating over all the other classes for more than 3 years, everybody knows that, nobody do shit about it.

Xytox wrote:

Probably the entire community agrees upon the fact that short, super aim-heavy maps are overweight. But this is not the issue brought up in this thread.

Man, its been literally 3 years since this shit started.

80% of top 50 are DT tv-size maps abusers.

It affects players to focus on short low effort maps, also it affects mappers who wants their maps to be played, so they are finding low effort short maps are way more profitable to spend their time on than making long maps with variety of stuff in it.

I mean, while we are at it, isnt it the best time to fix it all together?

Canadian wrote:

Shouldn't it be possible to reduce the aim pp awarded when the cursor teleports?

pp works by giving it the achieved score and difficulty of the map. It knows nothing about the specifics of the replay and might require infrastructure change to implement.

Other than that, it sounds like a plausible solution.

hi-mei wrote:

I think its a long awaited time for a meticulous review of how PP score is calculated.

There is a very tangible difference between 4 min 7* HDHR map fc and 20 seconds HDHRDT haitai FC.

This thing HAS to be addressed at some point, I truly hope that the solution will be finded and we gonna shift from TV-size farming meta to something a bit different.

I am not a math-guru, but I can say that the formula of PP calculation is struggling when it comes to sliders.

Maps with lots of sliders gives less score, doesnt matter if the actual difficulty is higher than the common TV-size anime maps.

So the main thing that comes to my mind is that PP system doesnt care how you hit sliders, how much input sliders are having on the player during the gameplay, how it affects reading and such.

But for temporary solution I can suggest to nerf how much PP is coming from short maps, it is just not ok that 20 second map gives 900 pp. Its just wrong.

I'm personally of the same opinion that the PP algorithmn needs to award stamina more than it currently does. While longer maps do give more PP it becomes extremely hard to get any PP from them as 1 miss in the middle of the map and you don't get anything for it. If the song is 5 minutes long it is much more challenging to stay consistent for 5 minutes than it is to stay consistent for 30 seconds. Easiest solution to this would be to simply make combo irrelevant and have PP calculated from Accuracy and Number of misses. You'd still likely miss more times on a long map than a short map but the length bonus would counteract the number of misses and make long maps more viable.

Also the current score meta would need to be altered as well as combo is still very much relevant in that and in order for a # of misses system to actually have a 1x play where you missed in the center to beat a 10x play where you comboed the first 2/3 of the song and then started missing periodically in the last 1/3 it would have to be altered to this.

Also by having combo less important Accuracy becomes more important and helps to negate alot of the issue with touchscreen.

I don't think touchscreen needs to be on a different leaderboard or even get a mod reduction thing to scan plays to see if they used a touchscreen. I just think the problem with the PP and Score system has to do with combo based scoring. 1 miss should be 1 miss no matter where it occurs

Edgar_Figaro wrote:

I just think the problem with the PP and Score system has to do with combo based scoring. 1 miss should be 1 miss no matter where it occurs

The problem with osu! is that it introduces an aspect other rhythm games don't have. If you don't force an FC out of a player for performance points, you'll definitely see mapping and playstyles evolve into a new kind of meta where intentionally missing one or two really hard jumps for the sake of comfortably hitting the rest is common, ruining the intended way to play the game to begin with.

The maps to abuse pp in such a way might not exist yet, but neither did the extreme tv size jump maps back when ppv2 was first introduced.

I just wanted to comment in order to explain why I'm very against deciding on solutions by listing a couple of options and then having the community vote on them.

Voting works great for producing correct solutions when the entire voting population has full information about the problem and has heard all the arguments. Like guessing the number of marbles in a transparent jar: voting gets precise answers pretty reliably. But for problems like "touchscreen and osu! PP", it's not the right tool. The osu!team and/or a few dedicated volunteers can think through the problem and proposed solutions in way more detail with way more complexity than is possible to expect from the entire voting population. Most voters will probably not even read this entire thread before voting, which puts a very hard limit to the number of arguments that are accounted for when solutions are decided by popular vote. (Even if you think you will think through the problem in high detail, and read up on everybody else's arguments, do you really expect this of most of the other voters?)

(Note: This is not (like, triple-not, absolutely not, not even the slightest) an invitation to speak about US politics in this thread, only comment if you wish to talk about the merits of using voting as a tool for deciding on the best solutions to the osu!touchscreen problem.)

abraker wrote:

Canadian wrote:

Shouldn't it be possible to reduce the aim pp awarded when the cursor teleports?

pp works by giving it the achieved score and difficulty of the map. It knows nothing about the specifics of the replay and might require infrastructure change to implement.

Other than that, it sounds like a plausible solution.

This is the best solution I've heard so far. In fact, I think it's more than "best", I think it's a good solution. Except I don't know how to implement it. It probably takes more work than a temporary solution can afford... Though I personally don't mind living with no solution until the proper solution is rolled out.

Edgar_Figaro wrote:

I just think the problem with the PP and Score system has to do with combo based scoring. 1 miss should be 1 miss no matter where it occurs

I haven't heard this argument mentioned anywhere, so I thought I should mention it (even though this thread should primarily be about TS). I like the current system of awarding score for higher combo, where you get a higher score if you hold a combo but miss at the end rather than the middle. The reason I like it is that it adds suspense to the game. As an example: I love watching tournaments, but I think the tournaments would be less fun to watch if suspense of the type "can rrtyui maintain his combo to the end of Image Material or not?!" was lessened. : )

peppy wrote:

Touchscreens greatly reduce the difficulty of aim as players using them now have up to 10 localised input points (their fingers) which both need to travel less distance and can preempt where then need to be for future hit objects.

peppy wrote:

A proposed solution of a touchscreen mod which devalues touchscreen pp values across the board may be something we apply as a stop-gap measure, for instance.

interesting
would like to see the status of this ^ ^

hi-mei wrote:

It affects players to focus on short low effort maps, also it affects mappers who wants their maps to be played, so they are finding low effort short maps are way more profitable to spend their time on than making long maps with variety of stuff in it.

in my opinion, the best way to fix this is to change PP system, having it rewarding different skills independently, like aim/speed/accuracy/reading/stamina and others skills i may forgot, something like how PP+ does (with better wheighting of course, the actual one isn't really accurate imo), so mappers that absolutely want their map being played wouldn't be forced to focus on aim focused map (because actually the game values aim more than anything else), that would actually add diversity on mapping (even if it'll end up the same as the actual mapping meta, they'll try to do the easiest overwheigted maps in X skill, but since it'll add more diversity i think it's still a good thing).

Having different wheightings would make it so if touchscreens plays a really short map with heavy aiming, easy flow and full of 1/2 rhythms (let's say haitai), they would get full PP from the 'aim' part of the map, but negtlecting accuracy, stamina, reading, etc.

Then also this could create news leaderboards per skill (aim, speed, etc) creating new competitions and new goals for players. But i guess it would be pretty hard to implement, i have no idea.

Edgar_Figaro wrote:

I just think the problem with the PP and Score system has to do with combo based scoring. 1 miss should be 1 miss no matter where it occurs

then FCing a map wouldn't be as rewarding as it is now, if you miss on a hard part or a slow part would mean exactly the same. FCing is still something that has do be rewarded, osu! combo wheighting is still a unique part of the game (as any other rhythm game doesn't use), making it important is what is making osu! unique imo. It makes you really wants to FC maps, and is rewarding you.

Thank you for the shoutout peppy

smooches

Hi, great post!
Regarding the pp algorithm, I think that there might be a solution? Im not very familiar with touchscreen, but im assuming that its pretty hard to do streams on it, isnt it? Because I rarely see any tablet plays that involed a lot of stream parts( like ror, freedoom dive etc) and mostly jump maps like granat. So why not buff streaming pp? If you dont have the way i think it would be pretty easy solution, like what is streaming? Its when theres little to no time between pressing notes, so just check that(the timing between notes and the ammount of notes pressed after the timing threshold is at right value), im sure you can figure it out. And its not necessary that it should be increased a lot, a tad bit should be alright.

Good luck with this problem,
-Atago

it seems the touchscreen only leaderboards is the only way that works, maybe I'm stupid but I don't think there have any methods to balance the pp system between normal players and touchscreen players. some maps is extremely easier for touchscreen, so let's say if u nerf aim only pp on short map. so what if a tablet or mouse player did the same record like touchscreen but they can't get high pp cuz of this nerf? that will be extremely unfair for tablet/mouse player.

buff the stream pp is also not works, it will destroy the balance no matter how u change it. if u want to put touchscreen players on the same leaderboard, maybe the only way is touchscreen players get touchscreen only pp system but that's even harder so better make a touchscreen only leaderboards.

sorry for my bad english.

hi-mei wrote:

No matter how you set the score, it is just makes people who has one strong ability to be higher (lets say Rafis) than people with other strong ability (lets say -GN).

Yeah, because Rafis totally can't do anything other than TV-size DT maps.

Anyway, I don't agree with all the arguments being made that short maps are inherently overweighted. Yes, sure, there are more farmable TV-size or shorter maps (my own top plays are pretty saturated with them) than longer ones, but no one says Dopamine is "overweighted" despite being less than two minutes long. Length isn't the only factor, and people overestimate its importance without regard to any sort of nuance. The same could be applied to aim - no one says Gangsta is overweighted (at least not in the past 3 years or so) even though it's literally just fast DT jumps - other factors, such as angles and the lack of a short difficulty spike, also should be taken into account.

Actually, Gangsta is probably a good map to take into account here given that half of its DT fcs have been using touchscreen. Very few people would say Gangsta is inherently overweighted, but it's still abused by touchscreen - so the whole argument of "everything can be balanced if you nerf this specific aspect of the game" is overly optimistic and lacks nuance.

Thx for responding to this issue so quickly peppy. As a touchscreen player, the only worry I have is that this touchscreen mod idea devalues all plays no matter the kind of map. While I know it is difficult to know if a map is a jump heavy map or a stream heavy map, I still feel like a solution like this would do more harm than good esp on more modern maps. For example, if I do well on a map that's more stream heavy or has intricate patterns, I don't think my play should be worth less than that of another player. While I understand that changes do need to be made, I only hope that your longterm solution isn't so overarching.