> -----Original Message----- > From: Rolf Dubitzky [mailto:R.Dubitzky at Physik.Tu-Dresden.de] > > Ok, the following example works. > (except: > - 'size' is not yet dynamic, > - width,heigth are meaningless at the moment, > - effect in/outpoint are ignored) > - keyframe time is not relative between 0.0-1.0, but absolut localtime > - the overlayed bitmaps are rgb while the video is yuv, gives funny effects > ) > > <veml> > <scenelist> > <scene duration="2" > > <effect tracks="1"> > <inputs> > <input file="../samples/4x3-PAL.dv" inpoint="0" /> > </inputs> > <video effect="textmaster" inpoint="0.5" outpoint="1.5"> > <dtext>Some test AV ??</dtext> > <font>/mnt/win/windows/Fonts/spacetoa.ttf</font> > <size> > <keyframe time="0.0">0.06</keyframe> > </size> > <position> > <keyframe time="0.0" x="0.0" y="0.1" width="0.8" height="0.2"/> > <keyframe time="1.0" x="0.1" y="0.1" width="0.8" height="0.2" /> > <keyframe time="2.0" x="0.3" y="0.6" width="0.8" height="0.2" /> > </position> > </video> > </effect> > </scene> > </scenelist> > </veml>
Ok, let's see if I understand what this means : In this scene, we have an effect that takes a single track as input. A video effect is applied called "textmaster", which presumably draws some text over the top of the video clip. The effect has two parameters - font and size. The font is defined to be spacetoa.ttf, and the size (of the font, or of the entire displayed text?) changes over time. A few questions : Do we need to specify the number of tracks in the <effect> tab? This could be determined easily as we parse the <inputs/> section of the effect. I think this question is something that we haven't really discussed yet : what effects do we need, and how should they be implemented? I bring this up because, for example, it is possible that we would want to have text displayed on screen that is not attached to any underlying clip. In the interface, this could be shown by dragging a "text clip" onto the screen and filling out it's parameters. Displaying text could be treated as a "generate text clip" operation, followed by an "overlay" operation, and the position of a clip could be modified by a generic "translation" operation, or it can be special-case. Have you had any thoughts on this? Cheers, Jason
