Drew Smathers(e)k dio:
> On Sat, Jul 5, 2008 at 11:21 AM, altern <[EMAIL PROTECTED]> wrote:
>> hi all
>>
>> today i am trying to catch up with the newest pyglet API. It looks very
>> nice although a bit weird at the first moment. I am working on this
>> simple example where many instances of an object are rendered on random
>> locs on screen. On my Desktop this takes almost 80% CPU, i am pretty
>> sure i am doing something wrong because most of the CPU time is taken by
>> the render method, (If I comment render() out it drops to 4% CPU).
>>
>> any suggestions? I am on Debian, on a PIV at 1.7 gz . Below i paste the
>> script i am working in.
>>
>
> I would suggest using pyglet.graphics.Batch to store vertex lists and
> just modify the vertex data, rather than calling OpenGL transform
> calls for moving your quads - this might not solve you cpu utilization
> problem completely since you still have to update 500 quads for each
> frame (*), but your framerate should increase by a significant factor.
> There is sgood documentation on this in the graphics section of the
> programming guide for 1.1.
interesting... i attach a test, could anyone please have a look at it
and tell me if i am going in the right direction?. CPU usage has dropped
to 15% with around 20 FPS. Very nice indeed.
I add the vertexes into a batch and at the same time I keep the vertexes
into a list, then loop the list replacing the content of them every
frame. Not sure if this is a good solution.
How can i limit the framerate? in the previous pyglet release there was
some system to limit it with clock.set_fps_limit(12) but this does not
seem to work now any more. Or do i need to use my own main loop? I
checked the docs on this but I did not understand how to achieve this
> (*) If you have to update lots of things at a rapid rate, you're going
> to use the CPU. Unless you're doing something fancy like CUDA.
>
> Also, unrelated to your problem, but pyglet already sets up an
> orthographic projection for you in the window's on_resize() - you
> don't have to do this in your setup() function.
many thanks for the tip!
--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---
import pyglet
pyglet.options[ 'debug_gl' ] = False # increase performace when using opengl. disables error checking
from pyglet.gl import *
from pyglet import clock
from random import Random
seed = Random()
width, height = 800,600
count = 0
w = 10
try:
config = Config(sample_buffers=1, samples=4, depth_size=16, double_buffer=True,)
window = pyglet.window.Window(width, height, resizable=True, config=config)
except pyglet.window.NoSuchConfigException:
print "applying safe configuration"
window = pyglet.window.Window(resizable=True)
fps_display = pyglet.clock.ClockDisplay()
def rand_v() :
x = seed.random() * width
y = seed.random() * height
return x-w, y-w, \
x+w, y-w, \
x+w, y+w, \
x-w, y+w
def setup() :
clock.set_fps_limit(12)
glClearColor(1, 1, 1, 1)
global batch, stack
batch = pyglet.graphics.Batch()
stack = []
for i in xrange(500) :
v = rand_v()
vertexlist = batch.add(
4, pyglet.gl.GL_QUADS, None,
('v2f', v),
('c3B/static', (255, 0, 0, 255, 0, 0 , 255, 0, 0, 255, 0, 0))
)
stack.append(vertexlist)
@window.event
def on_draw() :
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT)
fps_display.draw()
batch.draw()
def update(dt) :
global count
count += 1
if count > 100:
import sys; sys.exit()
for v in stack :
v.vertices = rand_v()
setup()
pyglet.clock.schedule_interval(update, 1./12)
pyglet.app.run()
