I've been playing mania for a while now (about 3 month(I'm a retard, I don't know why I wrote a month originaly)) a the question that stuck with me to which I can't find an aswer is what is the correlation between scroll speed and the time duration between then the note appears at the top of the screen (bottom) and the perfect time to hit it. I am looking for exact values if anybody has them meaning for x scroll speed you have xx Ms. Also, does the hit possition affect the duration or does the speed changes accordingly to keep the duration the same.

Don't know what you're asking about but sound like you need hit timing help, this could help you.
Scroll down to the part that says: Fine-tuning mania to suite your hit-timings

but before you do this, you gotta choose what fixed scroll speed fit you visually and if you can react to it (not to fast but not to slow) (doesn't matter if you can hit perfectly with it or not). You going to need to make your scroll speed into fixed scroll speed. To do this go to option and go under gameplay, uncheck the box that say: Scale osu!mania speed with BPM.

razavana wrote:

I've been playing mania for a while now (about a month) a the question that stuck with me to which I can't find an aswer is what is the correlation between scroll speed and the time duration between then the note appears at the top of the screen (bottom) and the perfect time to hit it. I am looking for exact values if anybody has them meaning for x scroll speed you have xx Ms. Also, does the hit possition affect the duration or does the speed changes accordingly to keep the duration the same.

With a skin where the lanes span the whole game window vertically (no key images), with a Scroll Speed of 1 in the fixed scale (and a map with not SV changes), it is about 13720ms between the time when the first pixel of a note appears, and the time when that pixel leaves the screen (measured by recording the game screen at 200fps).

I measured changing the ColumnWidth of the lanes to 20, 40 and 120, and it gave similar results (13715, 13725 and 13720 respectively), within the error of the measurement, ColumnWidth doesn't seem to be a factor, even if bigger notes appear to move faster. I tested with 4K keycount (keycount doesn't seem to be a factor). I might be missing some variables that can change the result.

Changing the hitposition doesn't change the speed the notes fall down the screen.

To get the time between the note apparition and the time you have to hit, you have to consider the hit position and the scroll speed. For example, if the hit position is set to be at 75% through the screen (a hit position of 360), and you use a scroll speed of 25 with fixed scale, then the time is:
13720ms*0.75/25 = 411.6ms

Overall, the formula would be:
13720 * (HitPosition/480) / ScrollSpeed
Where HitPosition takes a value between 240 and 480.
Scroll Speed uses the fixed scale, to transform from BPM scale to fixed scale, multiply the value by BPM/100. For example: 10 BPM Scale in a 250BPM song is equal to 25 in the Fixed Scale

As Ayaya said, using fixed speeds is a very good idea if you're not already.

I don't think anyone (aside from maybe people who've worked on the o!m code) know the exact timing data for when a note appears at the top of the screen. The exact time that a note appears would also depend on where you set your judgement line.
Ninja'd. Nice work on the calculations FullTablet. Still not amazingly useful though.

Osu!mania allows players to set the specific location on the screen that represents perfect timing. That is, you can customize at what height on the screen a note must be hit.

For an example, if you watch jhlee play, you'll notice he doesn't hit a key until the note has reached the very bottom of the screen, not just above the key graphics like most skins default to.

Here's an example:


If you watch closely, he doesnt hit the note until after it's passed under the bottom graphics of his skin. By default, the skin is usually set so that you need to hit the note at just above the graphics.

Now, if you're switching from bpm scaled speeds to fixed speeds, you'll notice that the scroll speed is much slower then you're used to. That's because BPM based speeds multiply the scroll speed you pick with the bpm of the song, so at 150 bpm, scroll speed 10 is 10*150=1500 scroll speed. At 200 bpm scroll speed 10 is 10*200=2000. in fixed speeds, it always multiplies your scroll speed by 10, so at 150 bpm scroll speed 10 is 10*100=1000, and at 200 bpm, it's exactly the same.

As for picking a proper scroll speed, that's personal preference. Usually you will find that there's one sweet spot speed that works better than every other speed, and that's what you should stick with. However, don't be afraid to speed up or slow down if things feel slower/faster than they did before. It's very natural for newer players to slowly increase their scroll speed as they get better at the game.

For example, when i started, there was no fixed speed. Anyway, for 150 bpm, I would pick speed 15 when I started out. As I got better, I slowly moved up until I was using speed 18 for 150 bpm.

Thanks everyone for taking your time to try and answer me. FullTablet, you're right on spot with what I was asking for. I thought of the exact method that you used but my pc is rather poor and I can't record with a high enough yet stable fps to get proper results.
Also, I see rather interesting results comparing your results for a hit position of 360 and a scrollspeed of 25 it appears I still get a longer time to react(423 ms), my settings being 415 hit position and a speed of 28 which subjectively appears a lot faster, though to be honest 360 is fairly high and i doubt many people use it.

That would be because scroll speed 28 IS faster moving, even IF the time it takes to reach the judgement line is longer.

I am well aware of that as I'm not that oblivious, all that i am trying to say is that that it seems a lot faster then it actualy is.

Even if your hitposition is 200 or 600, 28 speed is still 28 speed. Speed doesn't change, if you have a higher or lower hitposition all it does is make the play area big or small so it appear to be fast or slow but it's still the same speed.

Full Tablet wrote:

13720ms * (HitPosition/(480 osu!px)) / ScrollSpeed

Interesting will have to check. I was looking for this and asked this on Slack earlier.

Edit: I did additional work to figure out the speed at which the notes fall at (for my curiosity) and used it to find the fixed speed for which the notes take 1 second to fall. A bit redundant, but eh. You can see the document here.


.
.
.
.