[interfacekit] Re: server part of BDirectWindow

"DarkWyrm" <bpmagic@xxxxxxxxxxxxxxx> wrote:
> #1 would be my preference as long as it is done well. The only thing 
> I 
> can think of is something quite similar to Gabe's ideas of dynamic 
> software/hardware mode switching. A specific app_server message is 
> sent 
> (currently a private protocol) to make a cursor out of a BBitmap, so 
> figuring out when we're dealing with color isn't too hard. Rudolf's 
> problem is the only major one to the whole concept as a whole. 
> Unfortunately, it's a big one, and the only solution I can think of 
> for 
> R1 is using a hardware cursor (of some sort) whenever the pointer is 
> over a BDirectWindow's client area (or when there is a BWindowScreen 
> connected) and switching back when the pointer leaves the client 
> area. 

Hm, I'd guess that would work. Dynamic mode switching would be nice 
anyway, as it would allow better support for dual head cards like those 
from Matrox, which obviously only have a hard cursor on the first head 
(as Rudolf told me).

BTW is all that prevents BDirectWindow to work the ability of a 
hardware mouse cursor on VESA modes? If that is so, we might think 
about just turning it off when it is over them (because most of these 
apps will probably not need the standard BeOS cursor anyway).

Bye,
   Axel.

Other related posts: