In the Ouendan series of rhythm games, and in rhythm games in general, it's common design practice that every action the player is supposed to do is communicated to them some time in advance. This keeps the game fair and less frustrating. This is why there are approach circles, circle numbers, spacing, and in some other games, sliding note markers that line up with stationary markers. To the best player, it should be possible to read a beatmap and play it right the very first time. No guesswork should be involved. (Mind, that this is supposed to be extremely difficult, at least on the harder stages, because the cues might be very subtle or too numerous, overwhelming most players, but that's okay and makes the game more challenging as long as these cues exist in some capacity.)
osu! suffers from one problem in this regard, and that problem is the current way 2× sliders are handled: A 2× (or ½×) slider is presented exactly the same way onscreen as a regular one. Effectively, the rhythmic slider tick rate is doubled/halved, correspondingly. The player has no way, aside from failure and memorization, to predict that one of these is coming.
Given the way that these are implemented in osu!, by halving/doubling the BPM, this behaviour is obvious and correct, but it's this very implementation which is the problem. BPM describes the beats per minute of the music, and should reflect that. Doubling/halving it is an abuse of this concept, and it shows when playing Taiko and the approach rate (correctly) increases dramatically and confusingly in these circumstances.
This virtual doubling/halving of the tick rate can also have an adverse effect on the map's scoring. Often on maps with half-speed sliders, the ticks are so far apart that the player doesn't even need to keep their cursor on the slider to maintain their combo. Likewise, the double sliders both boost the combo faster than the mapper may wish, and are much more strict in terms of mouse movement tolerance than might be desired by the mapper.
osu!'s model for slider speed alterations is also severely restricted. The BPM alteration model only allows for slider speeds to become double or half their usual value, or otherwise be multiplied by powers of two. (Any attempt to use a less conventional multiplier both violates ranking guidelines and makes the slider ticks sound strange and arhythmic.) Naturally, the double/half rate is often too excessive and a mapper will want to use much more subtle speed alterations such as 1.5×, 0.75×, etc., but this is impossible under the current system. (To be completely honest, the desire to use these smaller speed variations in my maps is the primary motivation for this request.)
I propose a very simple, straigtforward solution to both of these problems: Allow the Slider Velocity Multiplier to be specified separately for each Timing Section. To accommodate some of the Super Zoom Sliders which have become popular, the maximum value for the Slider Velocity Multiplier should be raised to at least 5.2×.
The ranking guidelines would then be adjusted such that 2×/½× sliders would not be permitted to be made the old way for the reasons described in the second paragraph, but must now be done this new way.
(tl;dr: bold text)
osu! suffers from one problem in this regard, and that problem is the current way 2× sliders are handled: A 2× (or ½×) slider is presented exactly the same way onscreen as a regular one. Effectively, the rhythmic slider tick rate is doubled/halved, correspondingly. The player has no way, aside from failure and memorization, to predict that one of these is coming.
Given the way that these are implemented in osu!, by halving/doubling the BPM, this behaviour is obvious and correct, but it's this very implementation which is the problem. BPM describes the beats per minute of the music, and should reflect that. Doubling/halving it is an abuse of this concept, and it shows when playing Taiko and the approach rate (correctly) increases dramatically and confusingly in these circumstances.
This virtual doubling/halving of the tick rate can also have an adverse effect on the map's scoring. Often on maps with half-speed sliders, the ticks are so far apart that the player doesn't even need to keep their cursor on the slider to maintain their combo. Likewise, the double sliders both boost the combo faster than the mapper may wish, and are much more strict in terms of mouse movement tolerance than might be desired by the mapper.
osu!'s model for slider speed alterations is also severely restricted. The BPM alteration model only allows for slider speeds to become double or half their usual value, or otherwise be multiplied by powers of two. (Any attempt to use a less conventional multiplier both violates ranking guidelines and makes the slider ticks sound strange and arhythmic.) Naturally, the double/half rate is often too excessive and a mapper will want to use much more subtle speed alterations such as 1.5×, 0.75×, etc., but this is impossible under the current system. (To be completely honest, the desire to use these smaller speed variations in my maps is the primary motivation for this request.)
I propose a very simple, straigtforward solution to both of these problems: Allow the Slider Velocity Multiplier to be specified separately for each Timing Section. To accommodate some of the Super Zoom Sliders which have become popular, the maximum value for the Slider Velocity Multiplier should be raised to at least 5.2×.
The ranking guidelines would then be adjusted such that 2×/½× sliders would not be permitted to be made the old way for the reasons described in the second paragraph, but must now be done this new way.
(tl;dr: bold text)