I have both a 512MB Model B and a 256MB Model B. My linux build takes up about 30MB currently when running. Given the system requirements of the original alleg, there's a good chance it'll just work. But I don't want to try and setup the tool chain on the Pi itself (would take a dog's age to compile), and I don't have any experience with cross compiling for an ARM target. The biggest unknown is how well OSG performs on the PI. But some people are working on that, and there are precompiled binaries available.
I'll probably get around to smoke testing this, but if you want to give it a whirl and get back to me, that'd certainly let me know how well the rendering library might perform. Functionally, the VideoCore IV chip has enough power in it, and with the release of the recent free OpenGL drivers it's an attractive target. We'd just have to be prepared for the tsunami of children and teenagers that would come flooding in.
"I make it a point not to chat with AP off... space is vast, but it's never vast enough for my scout."
I would only caution against trying to optimise for PI while sacrificing quality on other systems (i.e. model polygon limitations/etc) as that would result in Alleg not only becoming niche game, but also run "at its best" on a very niche platform. (and we are having hard time attracting anyone with current artwork already).
The only thing you'd have to sacrifice is visual fidelity, and having the ability to support multiple levels of quality shouldn't be an issue. The CPU load of this game is minimal, so the limits you're going to hit on the pi are in terms of Video RAM and fill rates. So maybe they don't get 1k textures and bump mapping...
"I make it a point not to chat with AP off... space is vast, but it's never vast enough for my scout."
Modern space games have ~5-20k tris on each model. Both OGS and OGRE support LOD, but even at close-ups R-Pi will struggle with that (Quake 3 runs at 30-40 fps, but you get ~40k tris at most on screen)
All I can say is bully for them. As it stands, I don't have art assets even approaching that level, and when I do, they can be user optional. Everyone wins.
"I make it a point not to chat with AP off... space is vast, but it's never vast enough for my scout."
Feel free to get hacking. Msg me here or find me on the RT IRC channel if you have questions. If we've offficially got a 2nd dev I'll do what I can to provide better documentation and stuff, get the Wiki page setup and setup implementation requesets, bug lists on the github account.
But the short list is:
Make an offline dogfight trainer
Make an online deathmatch version
Get conquest mode back in
Polish
"I make it a point not to chat with AP off... space is vast, but it's never vast enough for my scout."
factoid wrote:QUOTE (factoid @ May 1 2014, 09:29 PM) Feel free to get hacking. Msg me here or find me on the RT IRC channel if you have questions. If we've offficially got a 2nd dev I'll do what I can to provide better documentation and stuff, get the Wiki page setup and setup implementation requesets, bug lists on the github account.
But the short list is:
Make an offline dogfight trainer
Make an online deathmatch version
Get conquest mode back in
Polish
GO FACTOID!!!!!!!!!!!!
QUOTE ^cashto@Elem (all): yeah, i imagine if you're rusty, you could build op short for no reason, build a naked ref, then go two techpaths even though your mining is by all objective standards $#@!ed[/quote]
Since not everyone here has access to the development forums, and it's really not the best way to communicate updates to the project. I present the FreeAllegiance wiki article.
It's still mostly just me, but there's a lot of artistic and administrative stuff that I could use help with, even if you're not interested in or able to assist with coding.
"I make it a point not to chat with AP off... space is vast, but it's never vast enough for my scout."
