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
In old AGS versions, before we switched to SDL_sound, beside using milliseconds two formats specifically supported an alternative way of using duration and seeking
MIDI - which used the beat number
MOD/XM/S3M - the pattern number (the "DUMB" library we used called it "orders")
This issue is more a question, if it would be acceptable or not to add some way to add format specific extensions in SDL_sound, and how it would work.
If such a way type of extension is acceptable, then the design could be done in a different issue, as I would like to add these two functionalities to SDL_sound to help us keep our backwards compatibility - looking at the dependencies in the repository it looks like the information is available in them, it's about devising an API to recover.
Currently the SDL_sound interface is opaque and there isn't difference when interfacing with different formats.
As such seeking, duration and everything else uses milliseconds. This is fine in general.
Some issues may arise though, when using "tracker-like" formats as mentioned in this comment here
SDL_sound/src/libmodplug/modplug.h
Lines 40 to 46 in fdcecaf
In old AGS versions, before we switched to SDL_sound, beside using milliseconds two formats specifically supported an alternative way of using duration and seeking
This issue is more a question, if it would be acceptable or not to add some way to add format specific extensions in SDL_sound, and how it would work.
If such a way type of extension is acceptable, then the design could be done in a different issue, as I would like to add these two functionalities to SDL_sound to help us keep our backwards compatibility - looking at the dependencies in the repository it looks like the information is available in them, it's about devising an API to recover.
Additional context
adventuregamestudio/ags#1695
The text was updated successfully, but these errors were encountered: