--- "Miller, Raul D" <[EMAIL PROTECTED]> wrote:
> Oleg Kobchenko wrote: > > there are three bottlenecks in new OpenGL: > > > > - blting: is OK for Win for medium window size, for full screen > > it's more than 3 times lower, but is much > > worse in Java possibly due to Java bridge > > (to test use full screen with simple model) > > I don't know about windows, but in principle for linux this could > be resolved by using a different opengl library. (This would also > require a different alloc and a different procedure for updating > the display.) Why diffent OpenGL library? Which one? The blting bottleneck is resolved by not blting, that is painting directly to the hardware. As I understand MESA by default paints to hardware, not off-screen buffer. > However, I don't know enough about the isigraph control's > internal implementation to know what this would mean for > the traditional "opengl gets rendered in an isigraph control" > approach. The difference is whether OpenGL is rendered to a bitmap, which is - then copied to a J array (possibly flipped) - then copied to gl buffer (with intermediate copies in Java) - then copied to the screen Or it's just drawn to the screen. > Note also that vertex arrays offer a very different sort of > abstraction than command lists. They seem to be suitable for > different classes of problems. AFIK, vertex arrays and display lists are orthogonal, so can be mixed and matched as you like. There may be an issue, that it might require the pointer to be valid after the call while it renders. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
