In both solutions the turntable views are juddering. The machine I was
testing on is a g4 powerbook with 1.5 GHz and 1.25 GB RAM. I have an
GeForce FX Go5200 graphics card with 64 MB VRAM.
What is genarally the better solution regarding performance?
CoreImage sits on top of GL and, I think, GL shaders. Thus if you use
CI the functionality of CI is going to rapidly speed up your
development time. CI provides a higher level imaging API than GL
itself. So, choose GL only if you don't need the associated CI
functionality (which is significant). Actually I don't imagine that
CI adds a great deal to slowing performance at all...
Would it help to thread the drawing?
Threads don't necessarily increase performance. Unless you can
identify some true areas of parallelism then threads won't help. Only
one thing can ever draw to a gl context at any one time.
Additionally, unless you have multiple CPUs, threads might not pay
off for you either. Finally GL (and I suspect CI) is not thread-safe
so you'll have to mutex it all yourself. Sounds like threads are not
required here.
If threading is reasonable how can I
guaranty that the drawing in seperate threads is save?
Mutexes. However a common strategy is to handle drawing just in the
main thread (also commonly referred to as the UI thread).
Furthermore I made the observation that OpenGL reacts a lot better to
mouse events than CoreImage, is there a logic behind that?
I don't really see this relationship - input is different to output -
I don't see the two things being associated with each other strictly
speaking.
I suspect that your juddering can be cured with double buffering and
flush synchronisation with display refreshes. There's a lot on this
subject out there. Lookup Opengl animation.
Cheers,
-C
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Quartz-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/quartz-dev/email@hidden
