wx.lib.graphics is pure pythonic. Cairo performs better on win/linux
but how do you get that performance out from wx.lib graphics?
The easiest way I can think of is to use cython(cdef & cpdef class
methods) and convert wx.lib.graphics into a .pyd module. I am not sure
how much performance gain I would get from this but it would be better
then pure python mode.
My app is using wx.GraphicsContext and Cairo seamlessly but there is
bug on win2k and linux where rendering an offline image using
wx.GraphicsContext on a wx.EmptyBitmapRGBA is producing alpha channel
with garbage pixels.
I have already posted about this problem long time ago.
Is there any other way to improve the performance of wx.lib.graphics?
I am not interested in any other features of cairo other then those
required in wx.lib.graphics.
Have you tried running the code under a profiler to see where the most time is being spent? Whatever that shows is where the optimization efforts should be focused.
···
On 9/24/10 9:37 AM, King wrote:
Hi,
wx.lib.graphics is pure pythonic. Cairo performs better on win/linux
but how do you get that performance out from wx.lib graphics?
The easiest way I can think of is to use cython(cdef& cpdef class
methods) and convert wx.lib.graphics into a .pyd module. I am not sure
how much performance gain I would get from this but it would be better
then pure python mode.
My app is using wx.GraphicsContext and Cairo seamlessly but there is
bug on win2k and linux where rendering an offline image using
wx.GraphicsContext on a wx.EmptyBitmapRGBA is producing alpha channel
with garbage pixels.
I have already posted about this problem long time ago.
Is there any other way to improve the performance of wx.lib.graphics?
I am not interested in any other features of cairo other then those
required in wx.lib.graphics.