21 |
*** Optimize compositing. Add support for different formats. |
*** Optimize compositing. Add support for different formats. |
22 |
??? Add support for reading TIFF images into GFImage-structures? |
??? Add support for reading TIFF images into GFImage-structures? |
23 |
|
|
24 |
|
|
25 |
|
|
26 |
|
* Surface server |
27 |
|
|
28 |
|
** libpowder |
29 |
|
*** Add hardware acceleration |
30 |
|
*** Update MGA-driver (add hw accel). |
31 |
|
??? Write KGI-wrappers to support more drivers. |
32 |
|
|
33 |
|
** Layers |
34 |
|
*** Look into the whole locking-business - it can probably |
35 |
|
be optimzed a lot. |
36 |
|
|
37 |
|
** General |
38 |
|
*** Double buffering should be optional. |
39 |
|
|
40 |
|
|
41 |
|
* General ideas |
42 |
|
|
43 |
|
** Create an interface for have "client" surfaces in video memory. |
44 |
|
This would increate performance alot (for hw accel). We should |
45 |
|
then don't have to move memory from system buffer into video |
46 |
|
buffer for bitblit. |
47 |
|
|
48 |
|
This will create add more complexity into the client library |
49 |
|
though. |
50 |
|
|
51 |
|
Another idea is to burst memory from system buffer into (possible) |
52 |
|
video buffer on surface update. This is a much simpler approach |
53 |
|
that will not add any complexity to the client . |
54 |
|
|
55 |
|
|