Show Posts
1
Предложения / Suggestions / Playlist shuffle function is not truly random
« on: June 24, 2022, 15:44:58 »
Hello,
I have been using AIMP for Android for at least 5 good years constantly in my car while driving and the way the shuffle function works is sometimes too predictable.
I have about 10 different playlists with 50-200 tracks each. The predictability is more noticeable in the lower track number playlists, no matter how many times I enable and disable the shuffle function (to try to randomize it more).
The unfortunate thing I noticed is that no matter how I change playlists back and forth, and change tracks manually (to hopefully re-shuffle them) it got very predictable as to which tracks come one after another.
Would it be possible to randomize the algorithm such that at every playlist change, or manual track change, the whole play order of the shuffled playlist change?
As a suggestion that would be more easily implemented, store 10 randomized lists generated from background noise, of each 1000 positions Then have the algorithm randomize combinations from these. Let's say I have a playlist of 282 tracks and one of 1620 tracks.
For the first playlist, it would fetch from list n 3, between position 300 to 582, then it would randomize it from a local algorithm that can be coded in the application.
For the second one it would get list n 5 + position 100 to 720 from list n 9, combine them then randomize it from a local algorithm.
If this is too complicated, just have one long list of 10000 positions to fetch numbers from, then randomize that interval further.
Maybe I am not understanding how the shuffle function works in AIMP, but it gets very annoying sometimes with shorter playlists. (maybe for longer playlists there is no time where I could listen to it for such a long period of time to notice this)
I have been using AIMP for Android for at least 5 good years constantly in my car while driving and the way the shuffle function works is sometimes too predictable.
I have about 10 different playlists with 50-200 tracks each. The predictability is more noticeable in the lower track number playlists, no matter how many times I enable and disable the shuffle function (to try to randomize it more).
The unfortunate thing I noticed is that no matter how I change playlists back and forth, and change tracks manually (to hopefully re-shuffle them) it got very predictable as to which tracks come one after another.
Would it be possible to randomize the algorithm such that at every playlist change, or manual track change, the whole play order of the shuffled playlist change?
As a suggestion that would be more easily implemented, store 10 randomized lists generated from background noise, of each 1000 positions Then have the algorithm randomize combinations from these. Let's say I have a playlist of 282 tracks and one of 1620 tracks.
For the first playlist, it would fetch from list n 3, between position 300 to 582, then it would randomize it from a local algorithm that can be coded in the application.
For the second one it would get list n 5 + position 100 to 720 from list n 9, combine them then randomize it from a local algorithm.
If this is too complicated, just have one long list of 10000 positions to fetch numbers from, then randomize that interval further.
Maybe I am not understanding how the shuffle function works in AIMP, but it gets very annoying sometimes with shorter playlists. (maybe for longer playlists there is no time where I could listen to it for such a long period of time to notice this)