Thanks for the quick reply.
Fabrice Closier - 05 November 2014 09:25 PM
Best way to find out is to output both types and see how it goes!
Of course, live testing is the best method to check. I just wanted to know the opinion of the more experienced developers.
For example, the output *.awd file is larger by exporting animated model as vertex animation (1 character model, 1 run animation).
As I understand, the more polygons, the more storage and computation is required. For each frame the position of vertices is recalculated and algorithm of linear interpolation is used.
But what I do not understand how the algorithm works with skeletal animation?
I guess, that animation data are not stored for each frame in this case, but only the key states of bones and vertices.
All other intermediate positions are calculated by more complicated algorithm.
This takes up less memory, but obviously requires more computation.
Is that so?
P.S.: Sorry for my english…