Unity 4 to offer support for Linux
arstechnica.com
arstechnica.com
Unity 4 is also the first version of the engine to
offer support for "the estimated 10 percent of the
game-hungry PC market" that runs Linux, though the
feature will only be available in "preview" form
for the time being. Those that pre-order Unity 4
starting today will get access to a beta test for
the new engine, ahead of a full release that is
still undated.
-- from the articleI hope I'm wrong.
So even if it is not the Unity IDE... it's still good news.
Running Unity apps themselves on the server even with this announcement would be a big mistake, IMO. It is not a stable piece of software.
[1] http://www.kickstarter.com/projects/inxile/wasteland-2/posts...
Each graphical element (a "prop" in Moai parlance), when moved, rotated or scaled, defines its own coordinate system, and other props can be added as children and share some or all of the new coordinates. It makes it very easy to compose elements. People familiar with 3D graphics would know this as affine transforms and UV coordinates.
For example, you can tie the position and scale of a text label to a sprite, but not its rotation, and tie the scale, rotation and position of the sprite to the background. You can then zoom everyting in one command, rotate the wole game world or and keep the label next to the sprite and rotate the whole scene or the sprite relative to the map while leaving the label straight. You don't have to track the relative angles, the engine does it for you.
Similarly, every object whose properties depends on time ("actions") can be attached or detached not only to the root action (which provide a tick each frame) but to other actions as well. When you detach an item, its descendants are also paused. Actions include animations, coroutines, physiscs worlds, you name it...
Both trees are defined declaratively in Lua, and implemented in C++, giving the best of both worlds: expressivity and speed.
Moai's biggest flaw at the moment is the sparse documentation, and we're tackling it. Right now, the main focus of the dev team is the cloud offering (which is our revenue source, after all), so it does not proceed as fast as we would like, but come back in a few weeks and you should see a vastly improved documentation.
The part that was interesting to me. "the estimated 10 percent of the game-hungry PC market" that runs Linux
This kind of runs contrary to the popular philosophy that gamers are very unlikely to be running Linux, mainly because the games support has historically always been very bad.
Of course there may be a mass of people who are running dual boots and are just waiting for a load of games to be released in order to delete their Windows partitions (people like me).
On the other hand, I run a dual boot and mainly use Linux as my desktop but since Windows is only a reboot away whether or not a game is available for Linux is not a massive part of my purchasing decision for the game. So it's possible it won't really affect their sales figures.
This surely also brings DirectX into question in terms of it's future as a platform. If I were to develop a game today I would almost certainly use OpenGL for portability reasons even if DirectX was a better library and it doesn't make a load of sense to make seperate DirectX and OpenGL versions.
What's the other thing holding Linux back from the desktop in your opinion that ranks higher than either of reasons I mentioned?
>What's the other thing holding Linux back from the desktop in your opinion that ranks higher than either of reasons I mentioned?
You're shifting the burden of proof here. For what possible reason would anyone believe the market wants Linux on the desktop? I can think of no reason. It doesn't look as nice, it doesn't run the major apps people are used to (e.g. Photoshop, Office, etc.). What "killer feature" does Linux have that would make the masses want to switch to it, and don't say "open source" since I doubt that even a majority of people who know what that means care about that.