On Fri, Oct 19, 2012 at 5:20 PM, Gabriele Lanaro
<[email protected]>wrote:
> Hi! I'm trying to build a molecular viewer with pyglet and I'm currently
> hitting performance problems but I can't figure out how to solve them.
>
> My test case contain 7 cylinders (100 triangles each) and 6 spheres (200
> triangles each). For each shape (13), I generate triangle vertices and
> normals and store those in numpy arrays. At the moment of rendering I use a
> call like this:
>
> draw(n, pyglet.gl.GL_TRIANGLES,
> ("v3f", self.tri_vertices),
> ("n3f", self.tri_normals),
> ("c3B", self.tri_color))
>
> I run a little test that simply rotates the camera, FPS are ~2. From the
> profiler output, sorting by time (time spent in the function, without
> subcalls):
>
> ncalls tottime percall cumtime percall filename:lineno(function)
> 7605 56.375 0.007 56.529 0.007
> /usr/lib/pymodules/python2.7/pyglet/graphics/vertexattribute.py:339(set_region)
> 169 0.891 0.005 0.895 0.005
> /usr/lib/pymodules/python2.7/pyglet/window/__init__.py:1209(clear)
> 8 0.520 0.065 0.720 0.090
> /home/gabriele/workspace/chemlab/chemlab/gletools/shapes.py:125(_generate_vertices)
> 2535 0.499 0.000 57.639 0.023
> /usr/lib/pymodules/python2.7/pyglet/graphics/__init__.py:168(draw)
> 41707 0.332 0.000 0.332 0.000
> /usr/lib/pymodules/python2.7/pyglet/gl/lib.py:85(errcheck)
>
> It seems that most of the time is spent in the set_region function but I'm
> not sure how to solve this problem. What do you think it's going on?
>
> The biggest win would be not to push all those vertices across to the GPU
every frame. Use batches instead.
--
Tristam MacDonald
Software Development Engineer, Amazon.com
http://swiftcoder.wordpress.com/
--
You received this message because you are subscribed to the Google Groups
"pyglet-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/pyglet-users?hl=en.
