You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Anim files after version 2.0 use lossy compression for certain tracks, which can cause desyncs in some games when used online if one the player's files is slightly modified. It's unclear if there is a reasonable way to achieve this given the current API design. ssbh_data considers anim compression settings to be an implementation detail, so ssbh_data_py also has no way of producing binary identical output.
It's unclear if the issue is the output not being binary identical or compressing at a lower quality setting than the original. It's possible to compute the error as a function of min, max, and bit_count. This function is not well behaved across its range of inputs, so there's no straightforward way to determine the optimal compression setting.
While not guaranteed, the compression <-> decompression process is 1:1 for floats in practice. This could be doable if the original compression settings were preserved. Attempts to modify the data would potentially require recalculating the correct compression settings.
The text was updated successfully, but these errors were encountered:
Anim files after version 2.0 use lossy compression for certain tracks, which can cause desyncs in some games when used online if one the player's files is slightly modified. It's unclear if there is a reasonable way to achieve this given the current API design. ssbh_data considers anim compression settings to be an implementation detail, so ssbh_data_py also has no way of producing binary identical output.
It's unclear if the issue is the output not being binary identical or compressing at a lower quality setting than the original. It's possible to compute the error as a function of min, max, and bit_count. This function is not well behaved across its range of inputs, so there's no straightforward way to determine the optimal compression setting.
While not guaranteed, the compression <-> decompression process is 1:1 for floats in practice. This could be doable if the original compression settings were preserved. Attempts to modify the data would potentially require recalculating the correct compression settings.
The text was updated successfully, but these errors were encountered: