[cairo] rookie question

[Dennis J.]

On 10/10/2008 09:03 PM, Carl Worth wrote:
> On Thu, 2008-10-09 at 00:45 +0900, KwangYul Seo wrote:
>> What is the current status of glucose? It is hard find information on
>> glucose. Is it still under development?
>
> I don't think it's being developed at all.
>
> It does seem strange to me that there always seems to be excitement
> about using OpenGL to improve our 2D rendering stack, (whether within
> cairo or the X server), but then the projects that attempt this end up
> stalling, (see Glitz and Glucose).

I think we are dealing with the problem of the chicken and the egg here. As 
long as no solid hardware acceleration exists as a baseline in the 
toolkits/x11 no application is going to rely on that and as long as no 
application relies on any sort of acceleration there is no demand for an 
accelerated 2D rendering stack.

> Meanwhile, there's *still* interest in this. Joe Drew plans to do a new
> OpenGL-based backend for cairo, (Joe, do you have a cute name for it
> yet?). And Eric Anholt plans to do a new OpenGL-based acceleration
> architecture for X, (it's got a nice name, but I'll let him announce it
> when there's more code than vapor).

Yet another OpenGL-based acceleration architecture for X. Don't get me 
wrong here I would love to see the come to pass but as long as there is no 
cooperation with a least the toolkit creators I don't really see this idea 
taking off. In order to get this on the road I think two minimum 
requirements have to be met: 1. The OpenGL architecture must be shipped by 
default with Xorg and always be enabled when the X11 driver supports 
OpenGL. 2. The toolkits (gtk, qt, etc.) must automatically detect this 
extension and make use of it.

Right now you can have a top notch card in your system and a working OpenGL 
driver yet I can still see all these redraw artefacts when dragging a gtk 
window around. In 2004 this was not so good, in 2006 it was bad but in 2008 
that's quite embarassing actually.

Regards,
   Dennis