Slashdot Mirror
New Linux Game needs Developers
shaggs writes "Time City is an open source, linux, 1st person action game, that will have time travel, network capabilities and more. We are currently looking for volunteers for all aspects of this project, and we need you the community to help. If you have any talents to offer us, visit Time City and sign up to volunteer today! "
This looks pretty cool. It will use the Crystal Space engine and be openly developed. With all the positive
exposure Linux has gotten, gaming is one of the areas where Linux is not well developed yet.
Free, high quality games could easily turn this around.
I've been pondering a game project myself, lately, and while I'm a big believer in the power of Open Source to improve the quality of software, I have some reservations about opening the source to multiplayer games.
Specifically, once the source goes public, there is no feasible way to prevent end users from changing the code to give themselves unfair advantages -- faster movement, more powerful weapons, brighter gamma correction, whatever -- and so disrupting the balance of the game.
As a different model, I point to the distributed.net folks, who don't release their source for very similar reasons. I posit a game-development model that follows their model -- aggressive bug fixing, constant posting of updated binaries for as many OSes as possible, and generally the same speed and responsiveness as an Open Source project, just without the actual source out there. Something much like id's scheme, but faster and more multi-platform.
Thoughts? I'd like to think that Open Source is generally a Good Thing(tm), but I'd like to hear feedback on how to make sure everyone's on a level playing field if your client code can be modified.
From the premise on their story page, this has potential to be one weird game.
...
You probably expect me to say something about causing major historical changes. You're waiting for me
to tell you not to touch anything, talk to anyone, or break anything. Balderdash! Just the opposite. We
want you to attempt to cause rifts in the timestream by any means necessary. Kill whomever you want.
Blow up a building if you can. It's only by testing our technology (and the universe) to it's limits that we
will learn from this experiment. Keep in mind that other historians receive the same advice; it will be just
as important to defend yourself against them, as well as the Chronopolian natives. Since we're a
scientific expedition, we're not licensed to issue weapons of any sort. You will need to find them yourself.
If this means that the game will actually try to keep track of events that happen at different times, and the way they affect each other... Well, it's hard to imagine how they'll manage it, but it could really be something. The problem is, since we can't really time-travel, how would it resolve cases like, "I think I'll beam a BFG-9000 back to the room where I had that fight last week. Then I wouldn't have gotten killed, so I would be able to do it."
They do seem to imply that that's the sort of thing they want to make possible, but it could get ridiculous, like that scene in "Bill & Ted's Excellent Adventure" -- "After I win, I'll go back and put a gun here that I can use to kill you (and win)." "Oh yeah? Well, after we win, we'll go back and make sure your gun isn't loaded, so we'll win."
It'll have to have some way of resolving these things: you can't have something you do take effect before the subjective time that you decide to do it, and even then it needs some way to be sure that you actually will do it before it can take effect.
This is all probably way beyond what it'll actually be, but it looks like they are going for some way of having actions and events affect each other non-sequentially. Maybe like forcing a rematch if you change the circumstances of a previous fight, and cascading the effects if the outcome is different. I hope they can pull it off.
That Crystal Space thing looks pretty interesting, too. I hadn't heard of it before. My #@%*% modem just hung up at 90% on the download, so I won't be trying it out for another few hours, though. (Why can't Netscape continue interrupted downloads instead of starting over? It does so occasionally, but it seems to decide pretty much randomly. I know, I know, read the source and fix it, right?)
David Gould
David Gould
main(i){putchar(340056100>>(i-1)*5&31|!!(i<6)<< 6)&&main(++i);}
I think it is *extremely* possible to make Linux into one righteous gaming platform.
First key is to realize that most modern BIOSes allow one to boot from CDROM. No need to have a seperate boot partition any more, let alone install Linux.
Second is to work on making the kernel use hot-loadable device drivers -- mix-n-match the components as needed for a given machine. Only load the drivers needed for a particular hardware config, etc.
Third is to have Linux able to safely read and write to other OSes partitions. Save game data and config setup can be stored on the hard drive without needing to partition it.
I'm going to quit trying to count, because I'm so freaking tired. I'll just babble on... if Linux can be designed to have a bloody fast kernel and device drivers with an excellent gaming-centric API, as well as excellent memory management functions, I'm sure game designers would find the platform extremely pleasing. As always, identifying what the customer (ie. game author) *needs* and *values* would be the trick here.
Plop in a CD, reboot the computer and, shazam, up comes the game. Quick, stable, awesome. Requires less computer hardware "power" because the OS isn't dragging things down... it'd be cool.
Gahd. Off for a nap, now...!
--
Don't like it? Respond with words, not karma.
This announcement seem a little too PRish for me. They haven't even got the game engine they want to base it on complete yet, and storyline and plot are cheap. (Hey, I've got a million idea lying around.) What am I trying to say? Show Me The Code, and then an announcement is appropiate, so all we little eager Linux beaver can download, compile and bitch about bugs to be improved.