TODO
author Patrice Mandin <patmandin@gmail.com>
Thu, 19 Jan 2006 21:28:52 +0000
changeset 1257 448a9a64537b
parent 150 df1d68818edb
child 1361 19418e4422cb
permissions -rw-r--r--
[PATCH] SDL_GetVideoMode() does not find best mode, part 2

Following commit 1.51, I come accross a problem when SDL must choose between
several video modes that could suit the one asked.

If I ask 320x240 with this list:
768x480 768x240 640x400 640x200 384x480 384x240 320x400 320x200

The smallest selectables modes are 384x240 and 320x400. And SDL choose the later
in this list, but 384x240 is more suitable. So I added a check to compare
the pixel count (surface) of modes, and select the one which has the smallest
pixel count.

In my example, 384x240 has 92160 pixels, and 320x400 has 128000 pixels. So now
SDL will choose 384x240 for the asked 320x240 mode.
slouken@0
     1
slouken@0
     2
Wish list for the 1.3 development branch:
slouken@0
     3
slouken@2
     4
 * Use /etc/fb.modes, if available, like GGI does
slouken@0
     5
 * Add mousewheel events (new unified event architecture?)
slouken@0
     6
 * DirectInput joystick support needs to be implemented
slouken@0
     7
 * Be able to enumerate and select available audio and video drivers
slouken@0
     8
 * Fullscreen video mode support for MacOS X
slouken@0
     9
 * Explicit vertical retrace wait (maybe separate from SDL_Flip?)
slouken@0
    10
 * Shaped windows, windows without borders
slouken@0
    11
 * Multiple windows, multiple display support
slouken@0
    12
 * SDL_INIT_EVENTTHREAD on Windows and MacOS?
slouken@0
    13
 * Add a timestamp to events
slouken@0
    14
 * Use RDTSC for timer resolution on x86 hardware
slouken@0
    15
 * Add audio input API
slouken@2
    16
 * Add hardware accelerated scaled blit
slouken@2
    17
 * Add hardware accelerated alpha blits
slouken@2
    18
 * Redesign blitting architecture to allow blit plugins
slouken@0
    19
slouken@0
    20
In the jump from 1.2 to 1.3, we should change the SDL_Rect members to
slouken@0
    21
int and evaluate all the rest of the datatypes.  This is the only place
slouken@0
    22
we should do it though, since the 1.2 series should not break binary
slouken@0
    23
compatibility in this way.
slouken@150
    24
slouken@150
    25
Requests:
slouken@150
    26
 * PCM and CDROM volume control (deprecated, but possible)