Monday, May 21, 2007

AVSynthesis


Some news after a long period.

I have been busy after my PC refused to boot, and I had to quickly replace it with a laptop, pre loaded with Vista. It was in March, and XP was not an option at this time (and this is no so widely available these days). The transition was not easy as the ATI display driver for the x1700 card that equips my new laptop has no OpenGL support!

Waiting for an update, I have done some tests with Ogre3D (a C++ 3D engine) and csound (worked fine together), as well as using vvvv and OSC protocol with csound (ok too). After a while, an updated driver supporting OpenGL was released by the laptop vendor and I was pleased to use again HPKComposer. The performance are right (sufficient). Core 2 duo (1.66) and X1700 class GPU have enough power for displaying three simultaneous layers of image transformation, and for running csound in real time.

After recovering my previous environnement, I have started the development of a new software derived from HPKComposer, AVSynthesis, with the following design points in mind:
  • No dependency on the Eclipse frameworks, for facilitating cross-plaform distribution.
  • Simple and multimedia like User Interface.
  • Random generation of composition made of Audio Visual layers (maximum of three at any time), for ever changing runtime interpretation . The software becomes part of the composition process.
Here is a link to a document that shows some screenshots of AVSynthesis.