@matt

I did a quick refactor to build a list of vertices and then touch the batch 
only after all the objects are updated.  I got very minimal performance 
gain.  I spend less time in _set_vertices but more time in the extend 
method of List and in other places.  I need to analyze the profile data 
more.  The old way was as fast as it could be and this new way probably has 
more room for improvement, but since I don't have textures and animations 
implemented yet, I'm going to leave it as it was which was more straight 
forward.

@greg

Its not very many vertices so numpy would't help.  If I could collect the 
data from all the polygons (center, theta, vertices and helper data) in one 
huge array, there could be some gain perhaps.  But building that array 
would be just as inefficient as updating the the polygons one at a time.

Thanks everyone.  Once I implement textures and animations, and know all 
the issues related to that, then I'll come back to this question of 
rendering efficiency.

-Elliot

-- 
You received this message because you are subscribed to the Google Groups 
"pyglet-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/pyglet-users?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to