On Wed, Sep 10, 2008 at 11:09 AM, Elmo Trolla <[EMAIL PROTECTED]> wrote:
>
>  > On Tue, Sep 9, 2008 at 1:27 AM, 3TATUK <[EMAIL PROTECTED]>
> wrote:
>  > > ...
>  > > At this point it's kind of pre-mature optimization because pyglet
> has
>  > > serious FPS issues as it is.
>  > > ( see http://code.google.com/p/pyglet/issues/detail?id=341 )
>  >
>  > Pyglet does not have serious FPS issues. set_fps_limit() is
>  > deprecated.  If you want an almost constant FPS, write your own event
>  > loop with blocking clock ticks.  Otherwise, schedule your animations
>  > and be happy.
>  >
>  > -Drew
>
>
> Writing a good main-loop is not simple - mine took many days and is
> still not perfect. I hoped pyglet will do this work for me, since
> everything else seems really well thought out. Thanks :) But
> considering main-loops - all examples that come with pyglet have some
> issues (at least under windows. pyglet version 1.1.1 2224 2008-08-26).
>
>
> Pyglet 1.1 has the new application event loop. Docs recommend using
>
>  clock.schedule_interval(update, 1/60.0)
>
> to get 60fps, and
>
>  clock.schedule(update)
>
> to get max fps. Neither works, as fps is not even near the desired
> fps. Trying to use the deprecated
>
>  pyglet.clock.set_fps_limit(60)
>
> results in jumpy animations (if vsync is off) and too high cpu usage.
> Now what..
>
>
>
> I've created four test-programs, each one a slightly modified version
> of the pyglet-1.1.1-docs.zip opengl.py example. WinMerge can be used
> to quickly see the differences between these versions.
>
> http://etm.blastnet.ee/var/list/opengl_clock_schedule.py
> http://etm.blastnet.ee/var/list/opengl_clock_schedule_interval.py
> http://etm.blastnet.ee/var/list/opengl_fps_limit.py
> http://etm.blastnet.ee/var/list/opengl_custom.py
>
> And here are the results I got: (copy-pasted from the py files)
>
> ----
>
> opengl_clock_schedule.py
> pyglet.clock.schedule(update)
>
> #
> # after startup                       :         64 fps       0% cpu
> # while moving mouse inside the window:   143..149 fps,      3% cpu
> # after window resize or move         : 3300..3500 fps, 60..70% cpu
> (and stays this way)
> #
> # (core2duo e8200, nvidia 8800gt, xp sp2. 50% cpu means 100% of one
> core)
> #
> # window.invalid = True has no effect on fps
>
> ----
>
> opengl_clock_schedule_interval.py
> pyglet.clock.schedule_interval(update, ..)
>
> #
> # param : fps  cpu  cpu while mousemove  cpu after move or resize
> #       :
> # 1/30. : 21    0%           3%                  20..23%
> # 1/60. : 32    0%           3%                  32..39%
> # 1/100.: 64    0%           3%                  58..63%
> # 1/200.: 64    0%           3%                  62..75%
> #
> # no change if i move the window, or move mouse inside the window.
> #
>
> ----
>
> opengl_fps_limit.py
> pyglet.clock.set_fps_limit(..)
>
> #
> # limit  :     fps         cpu
> #        :
> #     0  :  3400..3600   70..82%
> #  1/30. :    29..60         14%
> #  1/60. :    59..60         28%
> # 1/100. :    99..102        44%
> # 1/200. :   180..185        50%
> # 1/400. :   126..130        50%
> # 1/800. :   126..130        50%
> #
> # fps was NOT stable. even changing between 59..60 means jumpy
> animations.
> #
>
> ----
>
> opengl_custom.py
> This has my own main-loop. Compare cpu usage.. And animations really
> are as smooth without vsync as with (ok, some occasional tearing, but
> absolutely no jumping).
>
> #
> # fps_dt :     fps         cpu
> #        :
> #     0. : 3400..3600    70..82%
> #  1/60. :         60         0%
> # 1/100. :        100         0%
> # 1/200. :        200         0%
> # 1/400. :        400     0..20%
> # 1/800. :        800    50..76%
> #
>
> ----
>
> Sadly I chose a bad example. Problems would've been much more visible
> on a big rotating checkerboard. Well, until next time..
>
>
> Some of the jumpiness can be traced back to
>
>  pyglet.clock.tick()
>
> returning too big values ~4 times per second (if the program runs at
> 60fps), and too small values rest of the time. Using standard time
> module
>
>  dt = time.clock() - prev_clock
>
> is much more precise.
>
>
> I'll see if I can find something in pyglet source.
>

Hi, thanks for your detailed research.  I'll look into this in the
next week and let you know what I find.

Alex.

--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---

Reply via email to