On 14. Apr 2008, at 4:31, rahul Kumar wrote: > > if we use equalizer, rendering module of VTK has to changed. > Before delving into it further, i would like to get some comments > about its > feasibility and efforts required
Hello Rahul, I personally have not delved into VTK, so my comments are quite generic. In order to get to run on a multi-GPU machine, you have to execute the VTK rendering from Channel::frameDraw using the contextual data given by Equalizer. This means using the frustum (cf. getFrustum, getHeadPosition), 2D viewport (getViewport) and stereo buffers (getDrawBuffer) to render. The GL context is made current by Eq, but sometimes you have to setup a per-context structure (i.e. per window) for your rendering toolkit to store GL-context-specific data. Another item is window and pipe initialization. Either you keep the Eq default routines to create a window (recommended) and pass the Window context and drawable to VTK, or you let VTK create them and pass them to Eq. I've done this work for an nvsg-based application, and my feeling is that it should be relatively easy to do for VTK. HTH, Stefan. _______________________________________________ eq-dev mailing list [email protected] https://in-zueri.ch/cgi-bin/mailman/listinfo/eq-dev http://www.equalizergraphics.com
