On Tue, Aug 10, 2004 at 11:10:39PM +0800, Chris Leung wrote: > Dear all, > > I would like to know what is the meaning of difference display layer type > means?!
They are basically just hints. > e.g. if a layer have the type 'DLTF_VIDEO', will such layer benfit > anything if it use for render a video provider stream?!... > > I also found that a layer can still render a video provider stream if the > layer only get type 'DLTF_GRAPHICS'. So, what is the difference for a > layer get 'DLTF_VIDEO' and without it? As it stands DLTF_VIDEO seems to indicate YUV format support. DLTF_GRAPHICS probably indicates RGB/LUT format support. I'm not sure what DLTF_STILL_PICTURE stands for as some layers set it and some don't. We should probably review all the drivers and make sure the flags are consistent. And maybe come up with some guidelines for driver writers. > BTW, could I write a directfb program such that it will render a video > provider stream in layer 1 and then make the primary layer above the layer > 1 and display some transparent GUI in primary layer? Only if the hardware supports alpha blending between the layers. You must check the layer capabilities for DLCAPS_ALPHACHANNEL to find out if it's possible. With DLCAPS_DST_COLORKEY you could use an opaque GUI. > Since I found that I could do the same thing with primary layer only. e.g. > render video and display the GUI both in the primary layer. So, any > benefit if I use the 2 layer?! It really depends on what you want to do and what hardware you have. Usually if the layers' can do what you want you should use them. It reduces the amount of work the accelerator has to do. -- Ville Syrj�l� [EMAIL PROTECTED] http://www.sci.fi/~syrjala/
