Future of SDL
This email went out on the SDL mailing list today, and I thought it might be of interest to many of you.
* What's happening with SDL?
Over the past year and a half, I've been swamped with a newborn child
and working on World of Warcraft. I finally have time to move forward
with SDL development. I've caught up on the SDL list e-mail, and Ryan
Gordan is hosting an SDL bug tracking system based on bugzilla:
http://bugzilla.libsdl.org
(ignore the SSL certificate warning, it's a virtual host on icculus.org)
Here's what's happening:
SDL 2.0, the intended complete rewrite of SDL is officially vaporware.
Nobody has the year or so of time it would take to completely redesign
and re-implement the level of platform support that SDL has right now.
So, the ideas that were going to be in SDL 2.0 are going to be migrated
incrementally into SDL 1.2 and SDL 1.3.
SDL 1.3 is the experimental branch of SDL, where we will be breaking
binary compatibility and upgrading the API to fix a few shortcomings and
support new functionality. At one point SDL 1.3 had experimental render
to texture support, but it wasn't possible to maintain the same GL context
semantics on each platform, and it has been phased out in favor of the
OpenGL vertex buffer object extension. The original render to texture
code is in bugzilla as a patch relative to 1.3, for historical interest.
Right now, I'm in the process of integrating the SDL 2.0 build system
into SDL 1.2 and 1.3. This involves dropping automake and going to an
autoconf based build system. It's more work, but it avoids the constant
workarounds trying to stay compatible with three or four versions of
automake that are common on the platforms SDL supports. In addition,
I've set up enough C library support that SDL can potentially be built
without any C runtime support on the target platform. A proof of concept
was done on the Win32 platform. There is also a configuration header
SDL_config.h which makes it much easier to set SDL up for new platforms,
and other kinds of build systems. This work is almost complete, and you
can check it out in SDL 1.2 CVS (http://www.libsdl.org/cvs.php)
Other people are busy working on SDL as well:
Bob Pendleton is working on multi-window support based on the SDL 1.3
code, and has some proof of concept work in progress for the X11 driver.
Eric Wing is working on MacOS X 10.4 projects and Universal Binary
support.
Stephane Marchesin has contributed the glSDL code for inclusion in
SDL 1.3, and will be integrated once the build system revamp is done.
David Olofson is working on non-power-of-2 audio conversion in the
SDL audio code.
Once the build system revamp is done, I'll be doing a sweep of bugzilla
to catch important bug fixes, review compatible API improvements, and I'll
be putting together a SDL 1.2.10 release.
* So you want to help?
If you're interested in helping, you can grab the latest CVS snapshot
and build it and test it out on any platforms you have available to you.
http://www.libsdl.org/cvs.php
If you have a bug report in the current version of SDL, please make
sure it hasn't been fixed in CVS, and then submit a bug report to bugzilla.
http://bugzilla.libsdl.org
If you want to help with development, feel free to browse the bug list
and pick up anything that you'd like to help with.
See ya!
-Sam Lantinga, Senior Software Engineer, Blizzard Entertainment
i don't know what to say other than WOOT!!1!!11one!1!!eleven!1!!!1
a complete rewrite would be nice, but this is still great news for sdl users.
Quote:
SDL 2.0, the intended complete rewrite of SDL is officially vaporware.
Nobody has the year or so of time it would take to completely redesign
and re-implement the level of platform support that SDL has right now.
So, the ideas that were going to be in SDL 2.0 are going to be migrated
incrementally into SDL 1.2 and SDL 1.3.
a complete rewrite would be nice, but this is still great news for sdl users.
To be honest, I stopped reading the main SDL list over 2 years ago, and even then I could tell that SDL 2.0 was never going to happen. They should have concentrated on getting OpenGL rendering into what they already have and making incremental improvements a long time ago.
Awesome.
What's the ETA on these new features? I don't want my tutorials to be outdated.
and is there any news of force feedback support? I would like to some day make Linux from Scratch/SDL/OpenGL based arcade boards and consoles.
Man I wish I new lower level programming so I could help. [sad]
What's the ETA on these new features? I don't want my tutorials to be outdated.
and is there any news of force feedback support? I would like to some day make Linux from Scratch/SDL/OpenGL based arcade boards and consoles.
Man I wish I new lower level programming so I could help. [sad]
LazyFoo, since they're working on 1.2 and 1.3 rather than 2.0 I expect the API will remain largely constant. Once glSDL is integrated you should probably be able to switch to OpenGL rendering just as easily as you currently switch between DirectX and GDI now. (Which I bet lots of people don't know how to do, actually! But it is easy.)
Quote:Original post by Kylotan
...just as easily as you currently switch between DirectX and GDI now. (Which I bet lots of people don't know how to do, actually! But it is easy.)
Hate to thread-jack, but... You have me intrigued. [smile] If you could elaborate or maybe provide a link, that'd be great.
Vovan
Done through environment variables..
Example:
set SDL_VIDEODRIVER=directx
or
set SDL_VIDEODRIVER=windib
in a batch file prior to launching the app.
(you could also set them through system calls prior to starting SDL_VIDEO as well)
Example:
set SDL_VIDEODRIVER=directx
or
set SDL_VIDEODRIVER=windib
in a batch file prior to launching the app.
(you could also set them through system calls prior to starting SDL_VIDEO as well)
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement