I thought the polygons were made to measure the intensity (thus the difficulty) of the map. I agree with bwross, but BPM should not be ruled out as it still affects difficulty greatly.

The thing is that BPM doesn't directly affect difficulty at all. What matters is how it's used... a 320BPM map can easily be mapped like it was only 160BPM (or vice-versa). So Just knowing 320 or 160 can be deceptive. That's why the focus should be on the objects in the map, not the music. Which is why you want stats based on the density of objects... which is very much like the BPM (same units), but is actually things the player has to do.

Now, the BPM itself is still useful to know. People like to judge themselves against the BPM level of stream they can do well. But for that you want to know the exact value of BPM, not just an impression of how large it is... and that's something that's currently displayed (and should remain). However, the listing of a song at a BPM you have difficulty streaming at doesn't mean that it has any real streams that you have to go up against (or whether the streams in the map are divided into chunks of a size you can manage, or are put together into a single long death stream where you're going to eventually slide off). To know that, it comes back to the objects and how they're packed in the map. Right now that can be done to an extent by using the number of objects and the length (which can give objects/second)... but deeper analysis of the map and it's bursts could be far more accurate.

I don't mind if speed is added in a different form, but it seemed like some people in the thread were saying the speed of the map (notes and all) have no sway on difficulty. I think more people would be able to play the big black if it was 120 BPM xD. Note Density sounds good, but we still need a way of calculating it then.

I wholly support the idea of showing/visualizing more stats per map, but I can't say I like the "diamond" way. Simple - it takes up too much space, and map selection interface already feels cramped enough. I'd be fine with plain old horizontal bars - for OD, DR, CS, and AR, with colors ranging from dark green (easy) to crimson(insane). Pros: intuitive, expandable (stream intensity, spin rate and other things would be REALLY nice to see :3). Cons: plain, still might take up lots of space depending on implementation.

If they were to add this I would like the stats represented to be things that actually reflected the actual difficulty of the map. Things like average distance snap, approach rate, OD, circle size, stream length/ number of streams, and the stream density of a song. A large amount of factors can be lumped together in major categories like Accuracy, Technical, and Endurance ratings. Each map could also be given an intensity rating.

I just don't think just showing all the basic stats is any indication of a beatmap's actual difficulty, and we all know star rating can be poor at determining that. So a solution would be to detect things that do make a beatmap more difficult and represent those things as data values that we can actually ascertain valuable information from. The base stats are only a partial indication of difficulty.

Edit: A future idea would be a dynamic difficulty graph in which it takes your success rate for playing x difficulty at x average BPM, and it would then compare the stats of the difficulty to the proficiency of the player and then determine how difficult that map will be based off the number differences.

That would be very difficult to implement.

TheVileOne wrote:

If they were to add this I would like the stats represented to be things that actually reflected the actual difficulty of the map. Things like average distance snap, approach rate, OD, circle size, stream length/ number of streams, and the stream density of a song. A large amount of factors can be lumped together in major categories like Accuracy, Technical, and Endurance ratings. Each map could also be given an intensity rating.

so then maybe select maybe 4-5 categories. and those categories could be calculated using things that would effect the overall difficulty? like, as you said, average distance snap, approach rate, OD, circle size, stream length/ number of streams, and the stream density of a song

Timekiller wrote:

I wholly support the idea of showing/visualizing more stats per map, but I can't say I like the "diamond" way. Simple - it takes up too much space, and map selection interface already feels cramped enough. I'd be fine with plain old horizontal bars - for OD, DR, CS, and AR, with colors ranging from dark green (easy) to crimson(insane). Pros: intuitive, expandable (stream intensity, spin rate and other things would be REALLY nice to see :3). Cons: plain, still might take up lots of space depending on implementation.

The way I wanted it is an addon button that shows a page for it, from a default point you wont even see it on the song select until clicked upon...
So I really don't see what space your referring to unless its that tiny 16x16px icon

RBRat3 wrote:

The way I wanted it is an addon button that shows a page for it, from a default point you wont even see it on the song select until clicked upon...
So I really don't see what space your referring to unless its that tiny 16x16px icon

I'm referring to the large space that the diamond inside a circle takes up, regardless of whether it's shown by default :3 where diamond shows 4 stats, you can place 6-7 bar graphs plus some additional info like pass rate, actual bpm spread per song time and whatever else.

The "diamond" can show more stats if you want it to... you just add more arms to the star. Don't focus so much on the graphic mockup... the number of stats and what they are are up for discussion.

nic3 Idea support

Timekiller wrote:

I'm referring to the large space that the diamond inside a circle takes up, regardless of whether it's shown by default :3 where diamond shows 4 stats, you can place 6-7 bar graphs plus some additional info like pass rate, actual bpm spread per song time and whatever else.

Well it is a bar graph at its heart but adding more arms/legs doesn't take up anymore room. Theoretically its limitless but whether or not its discernible is another question XD...

The whole point of using one is being able to associate shapes with a maps difficulty attributes and bar graphs cant do this, well they can but its going to take a little more thought to get the association than you would with a shape and bar graphs tend to make you look at each individual bar rather than looking at it as a whole while a perceptual map allows you to do both.

As stated by bwross the mock up is well a mock up... The main issue is coming up with value sets that actually mean something to you and relate accurately to the map.

All that aside a graph is a graph is a graph... I don't see any reason not to slap a button on that page that will display these value sets in any applicable graphing format you wish after all it is just numbers with eyecandy

Well we just need to compile 4-5 different stats that house all the things you would want to know before you go into a map. Cursor speed/jumps should be one, hit density should be another, and approach rate/overall difficulty should be in there too.

RBRat3 wrote:

All that aside a graph is a graph is a graph... I don't see any reason not to slap a button on that page that will display these value sets in any applicable graphing format you wish after all it is just numbers with eyecandy

I always seem to forget that most things can be made configurable If choosing between diamond/bars/something else would be an option, I support. Here, have my precious star

As for speed and density what about averaging the amount of notes in 1 second slots or per deci-minute (1/10th of a minute / 6 sec)...?

AKA average notes per second or average notes per deciminute....

This would factor bpm changes although it would be averaged but it greatly influences the outcome being in 1 sec slots this way, If bpm changes shouldn't be averaged then average the notes per second in the bpm time sections keeping them separate....

You've got my support for this. This would make it easier for players of all skill levels (I know I still get taken by surprise by AR9-10 maps) in being able to tell exactly why a map is difficult before playing it without having to look at it through editor beforehand. Have you thought up of a comprehensive list of potential factors to be included in your graph?

I know there is talk already about BPM but I don't see why it should be included if the BPM of the song is already shown on the top-left of the song selection screen. If anything, if your Star Perceptual graph isn't a feasible feature in terms of how to fit it on or make a new tab for it, I can see the text stats of it being posted under where the current BPM and Objects text is displayed.

I would like to see these particular stats listed though overall so support+!

Yea, awesome idea. Spending my last vote for this.

I spended my one and only vote too...

it is a nice Idea.
why not make this configurable, choose you arms, your shape, or that you want bars...

You forgot something: Some old maps do have Circle Size below 3 and above 7 being set. This is one of Ranked example with Circle Size 0, and this is another Ranked example with Circle Size 8.

I haven't see anyone setting the map with Circle Size above 8 (maximum is 10) though, but it's possible to force-set it.

I think this is a great idea, however just out of curiosity, which one of these screen shots will be the final design because if it's going to partially cover up your own score, like the first picture, I'd be better off without it. I think the one under alternate looks best. It's a nice size and it looks like it is almost proportional to the scoreboard underneath and I like the fact that the black BG is as transparent as the scoreboard. This will really make it blend well. So.... In my opinion... Support!!

Kuro wrote:

I think this is a great idea, however just out of curiosity, which one of these screen shots will be the final design because if it's going to partially cover up your own score, like the first picture, I'd be better off without it. I think the one under alternate looks best. It's a nice size and it looks like it is almost proportional to the scoreboard underneath and I like the fact that the black BG is as transparent as the scoreboard. This will really make it blend well. So.... In my opinion... Support!!

None of them are a final design they're just concept... Im sure peppy would do his own take on the look.

yes but skinnable pls...

why is this even popular. it hardly displays all the necessary information you need.

I want my difficulty rating to be displayed as numbers, not visually like this. This takes up a lot of space.

t/92485
That is how we should display difficulty.

theowest wrote:

why is this even popular. it hardly displays all the necessary information you need.

I want my difficulty rating to be displayed as numbers, not visually like this. This takes up a lot of space.

t/92485
That is how we should display difficulty.

Because you failed to read high points, This can do both at the same time and it only takes the space of a 16px square.

Support. It should scale with for example HR and DT so when you add those mods the graph changes. No need for the difficulty changes the mod makes to be seen the entire time. If you add DT the graph for AR and Over all diff should just go up by it's respective values. It should also have numbers on it.
The hardest approved maps could have an average og 15-20 in over all diff value, while normal insanes could be 10-15 etc.

hmm

maybe up here

too small up there I Like the scoreboard option, coz I dont need it anyway...

My1 wrote:

too small up there I Like the scoreboard option, coz I dont need it anyway...

it should be small.

well, what are settings for...

Instead of the Difficulty Setting, why not use something like

Drain ( 0 - 10 ) - HP Drain Rate
Precision ( 0 - 10 ) Mix of OD and CS (since higher CS means higher aiming precision and higher OD means higher clicking Precision)
Stream ( 0 - 10 ) - How Fast and Long are the Streams? How many are present?
Jumps ( 0 - 10 ) - How Long are the Jumps? How many Jumps?
Speed ( 0 -10 ) - Depending on Average Notes Distance and Approach Rate

could be a way to upgrade the Difficulty rating system too.

HakuNoKaemi wrote:

Instead of the Difficulty Setting, why not use something like

Drain ( 0 - 10 ) - HP Drain Rate
Precision ( 0 - 10 ) Mix of OD and CS (since higher CS means higher aiming precision and higher OD means higher clicking Precision)
Stream ( 0 - 10 ) - How Fast and Long are the Streams? How many are present?
Jumps ( 0 - 10 ) - How Long are the Jumps? How many Jumps?
Speed ( 0 -10 ) - Depending on Average Notes Distance and Approach Rate

could be a way to upgrade the Difficulty rating system too.

i prefer that. but hey, they are going to upgrade the difficulty setting.

stamina would be cool to see too.

It shall be based on notes concentration and lenght?
Like an Endless 1/2 or a Deathstream producing higher Stamina and so?

the more notes after each other, the higher stamina. less stamina if there are more breaks, sliders, spinners, etc. less time between the notes (higher bpm) will increase the stamina difficulty.

It should be counted in ms instead of BPM. It all matters how you map after all.
Also, spinners (especially long ones) uses up stamina as well.

Btw RBRat3, I love your graphics


And of course, I fully support this

Idea looks nice! Support!

deadbeat wrote:

also if hard rock will play a factor in the star map, can i suggest having the cap at 11 instead of 10?

what cap do you mean with that? AR?

Saten wrote:

It should be counted in ms instead of BPM. It all matters how you map after all.
Also, spinners (especially long ones) uses up stamina as well.

very good point there. it should be taken into account

Stamina is what all the talk about object density is about. Breaks, sliders, spinners are areas with low density... areas with less time between notes are high density. However, if you're looking for information on how long and fast bursts are, that would be the realm of streaminess. It's best to keep things in the stats used as independent as possible... having stats that always move together means that you've screwed up and should have combined them into a single stat.

HakuNoKaemi wrote:

Instead of the Difficulty Setting, why not use something like

Drain ( 0 - 10 ) - HP Drain Rate
Precision ( 0 - 10 ) Mix of OD and CS (since higher CS means higher aiming precision and higher OD means higher clicking Precision)
Stream ( 0 - 10 ) - How Fast and Long are the Streams? How many are present?
Jumps ( 0 - 10 ) - How Long are the Jumps? How many Jumps?
Speed ( 0 -10 ) - Depending on Average Notes Distance and Approach Rate

could be a way to upgrade the Difficulty rating system too.

Things like Drain, OD, and CS shouldn't be part of a secondary chart. They're too important... they're not stats that tell you what the map is like, they're parameters that tell you the rules you'll be playing under. Besides, combining OD and CS is silly... the precision that affects OD isn't spacial, it's temporal. CS affects jumps and movement rate stats, because there the size of the target affects how difficult it is to hit. With OD, the change to timing windows means potential more 100s, and with a high DR, that can kill if you're not careful. But that's not important to combine... these are parameters that tell the player the rules of the game, they should be up front with the object and length information, always in sight, regardless of whether the player has up global/local scores or a star map. And that request has been made elsewhere.

Stream and Jump are pretty much universally consented. They're definitely important and what people want. The thing is that they refer to the burst rates of their domains (time for Stream and space for Jump)... they don't give a picture outside of that. Which is why there should be a measure of the base time and space rates as well. Base time is object density, stamina, effective BPM (which is to say, not the BPM of the song, but the BPM experienced in the map). Base space is a measure of the movement of a map, it's velocity, beat spacing, and slider speeds. Stream and Jump are the extremes that those go to.

Chaos has also been brought up a lot, but nobody seems to have presented a picture of what it would be. Well, if chaos is a measure of the entropy of a map, we need to consider what entropy is. Entropy is essentially a measure of the number of states of a system. Consider a glass, it's solid and has a certain shape. The way it's atoms can be arranged in that glass is large, but it's miniscule compared to the number of states if you throw the glass on the ground and break it... where the number of pieces and their shapes and there positions relative to each other adds a nigh infinite number of new positions. And so the glass has gained a lot of entropy by being broken.

What does that mean for osu!... well, consider a stream of 12 overlapping circles in straight line. When such a thing comes up in a map, it registers essentially as one object... it's like a slider, but you need to hit notes instead of holding them. From a stamina point of view, that adds something, but from a mental chaos point of view, it's the same, the stream can be processed as one thing. Now consider replacing one of the circles with a repeating slider. It has a different length than the rest, is a different type of object, and can had a variety of lengths (depending on the number of repeats), and could occur in any of the twelve spots... suddenly the processing has become more than one thing, it's gained some entropy. Take those twelve circles and have them bouncing around in a pattern, and you can get more entropy yet. Add a complex rhythm and you add more possibilities and entropy... add some emphatic spacing for that rhythm (ie jumps, changes in beat spacing) and you add even more.

So what I'm thinking for a way to measure entropy, is that you need to look at adjacent intervals between objects (under various size windows... I'd consider using combo groups as a width for windows (ie 1 group, 2 groups, 3 groups), because they should delineate patterns in the map to some degree). If they represent different notes or have different beat spacing, then entropy is definitely higher than a map of 1/2 beats that trudges around the map at the same beat spacing. Pattern entropy is a harder thing to measure. A simple way that might be worth looking into, is to consider changes in angle between intervals. A straight line is just 0, 0, 0... (note that we're talking change in angle (delta) so the exact direction doesn't really matter, only that each circle heads in the same direction). A slightly curved line might not be more complicated... say, 5, 5, 5 (degrees). Regular polygons (and polygrams) would also register as simple recognizable patters (ie squares would be sequences of 90s). Jumping back and forth from a series of radial points adds a bit more complexity: a 180 to get back to the center, alternated with the angle to the next poiint... as would a zig-zag. Random jumping around would be random. There are a number of ways to potentially analyze such things... one would be to look for cycles and other patterns, but a simpler way might just be to look at the numbers within a window and see if there are one or two dominant values. In the end, a lot of experimentation with real map data will be needed in any case to work out what the weights of any of the factors should be (and how they might combine... ie a sudden change in direction combined with a note of a different duration than anything around it combined with a change in beat spacing). Although, you could also try to apply machine learning to working out the factors... although, then you'd still need reliable expert opinion on what the values should be to teach it.

Anyways, there's a brain dump on "Chaos" as a stat for you.

I'm a derp

JeMhUnTeR wrote:

OMG BUUUUMMPPPPP

what have you done...

Well, since this has now been bumped, I suppose i'll contribute by saying that I think the important thing is to just keep the information shown in the chart fairly simple just to get it implemented. If it needs to be fine-tuned over time then great, but what i mostly wanna see is a more accurate representation of difficulty besides the .5-5 star rating, because lets be honest, the star difficulty rating should really go up to like 15.


tl;dr I support.

JeMhUnTeR wrote:

I'm a derp

how dare you edit my post >.<