Sauerbraten Open Source Project?

by Aardappel on 01/21/2004 04:40, 94 messages, last message: 03/18/2004 20:41, 43896 views, last view: 05/18/2024 21:13

This discussion is for determining whether there is interest, skill, and determination available in the Cube community to make Sauerbraten an Open Source project... as in, one with many contributors.

You probably all have seen Sauerbraten on my main page, if not: http://wouter.fov120.com/sauerbraten/index.html

It really is a wonderful concept if I may dare say so myself. It also requires quite a bit more work than cube did and involves significantly more complex algorithms. I have been torn what to do with it for quite a while. Between Cube (which I want to continue) and my new gamedesign projects and my job, I am not sure whether there will be enough time for me completing it entirely on my own.

Normally I simply write software, and then open source it once I am finished, with sauerbraten I will consider a new direction, and make it a open source project all along. If this is succesful I may even consider making cube an open project, though I don't know how yet.

I would like to hear everybodies input, especially that of serious potential contributors. And here lies a problem, as you are all aware of the many projects that have people apply to do great amounts of work for and then absolutely nothing happens. I don't want to have to deal with that. Not everyone will be able to commit to the project, only very serious, committed and qualified people. If no such people are available, Sauerbraten will remain closed.

Also I still am an arrogant bastard, and will expect those working on the project to follow my guidelines for the design of sauerbraten closely. It will thus be more of an open cathedral rather than a bazaar :) But hopefully those who see the quality of Cube & Sauerbraten see that as a good thing.

let the ideas flow...

   Board Index   

#63: ..

by Aardappel on 02/28/2004 19:29

staffv02: its good to hear sauerbraten even does well on ancient machines. For those who do get bad performance on sauerbraten however, it will get a LOT better

fsmunoz: of course, eventually once sauerbraten matures, it will kick cubes ass and make it obsolete. But we're quite some time away from that moment. Right now cube has a lot going for it that sauerbraten doesn't have yet :)

email me the new make file and other linux changes required, and I will put them in CVS

sirlivealot: congratulations, you just made the first user-made sauerbraten map ;) But I think more can be done...

reply to this message

#64: Haven't looked at it yet ...

by spentron-postcrash2 on 02/28/2004 20:48

... but sounds like everyone else is, must be ready for a look. Some thoughts simmer already ...

Cube vs. Sauer -- Cube works better than anything else for certain things, it's just we've already seen basically what that is. For example, a grid-grained complexity such as a "cubic cave" like the puddly one in Conundrum would be all but impractical in some engines.

Cube emphasizes simplicity and relies on it on a certain cerebral level for its appeal. I don't think adding a bunch of complex widgets to it would make sense. Another advantage is expediency. I have said I go through maps like food, so this originally appealled greatly to me. However, Cube doesn't happen in a gaming vacuum, and we mappers all tend to obsess over our creations, so this advantage is in fact limited.

Sauerbraten is, philosophically, a shift from that. Compared to existing full-featured engines, there is a slight limitation in creating geometry in exchange for an increase in directness of creation. The difference is far smaller. As long as added widgets have a certain "sauerbraten" flavor of also being not over-complex or indirect, adding complex features does not conflict with the character of it.

Now this "RPG" thing sort of bugs me. Saying "like Morrowind or Gothic" tells me nothing, I haven't played those games and don't plan to. I don't think this is bad if it added in the spirit of the philosophy I just stated -- this new engine is less about limitation and therefore no old feature, if still useful, must be given up in exchange for new.

staffy: I just told someone Cube wouldn't run on 810, find out something new ...

reply to this message

#65: Re: Haven't looked at it yet ...

by Aardappel on 02/28/2004 21:15, refers to #64

I think you have the wrong impression of sauerbraten... its both simpler and more powerful than cube at the same time. It isn't complex at all. You will find that what you appreciate in mapping for cube to be even more so the case in sauerbraten.

The RPG is just an idea I have. The order of things is first an engine, then getting the base old cube gameplay working again, then looking forward to a new game. So there will definitely be a place for those preferring to stick with FPS gameplay.

reply to this message

#66: Re: Haven't looked at it yet ...

by spentron-postcrash2 on 02/28/2004 22:59, refers to #65

Maybe I should have said Cube does just fine on some things. Also I agree the game aspect expansion is appropriate and shouldn't be considered a replacement, my point was expansion makes more sense for it than for Cube. And even that is sort of a misconception, possibly some RPGers could see Cube as ideal for limiting the 3D world (some sort of text/2D graft?).

Anyway looked at it, interesting.

reply to this message

#67: sf cvs

by hungerburg on 02/29/2004 16:09

is missing enet/include directory.

reply to this message

#68: linux patch

by hungerburg on 02/29/2004 16:31

fsmunoz hints to compile sauerbraten on linux as a patch: download it from http://arton.cunst.net/cube/sauerbraten-init-linux.patch

to apply, change to project root directory - where eg. the src dir is - and then patch -p1 </path/to/patchfile

the patch is made on files via cvs, with unix linefeeds - so it may mess up the files extracted from archive...

reply to this message

#69: ..

by >driAn<. on 02/29/2004 20:55

looks great!

reply to this message

#70: Re: linux patch

by D.plomat on 03/01/2004 09:52, refers to #68

Congrats fsmunoz and hungerburg, you were faster than me having it work on Linux ;)

The speed of the engine is good, and the overall aspect excellent

Looking at the test maps, this shows that this engine'll allow for immense maps and much more "vertical" ones :)

reply to this message

#71: I like!

by Lethedethius on 03/08/2004 02:41

I like it, when you release this aard, what else do you plan on releasing? :), perhaps give cube it's own format for skins & weaponz ;), make it more unique :D

reply to this message

#72: Re: I like!

by dcp on 03/08/2004 05:31, refers to #71

if he gives cube an unique model format, no one will be able to make models for cube anymore.

imho the .md2 format is absolutely sufficient, as there are some good modeling apps which support this format.

reply to this message

#73: large maps

by lurch on 03/08/2004 23:26

Will sauerbraten be able to have large maps(like sidewinder in halo)? I get the idea that would be inpractical in cube.

Also I don\'t think you will get much work on the engine out of the open source comunity, Most open source fans are shitty programers, but are quite good at modding/map making/etc. So if you want us to help, finish the engine yourself and make it modable.

reply to this message

#74: large maps

by lurch on 03/08/2004 23:26

Will sauerbraten be able to have large maps(like sidewinder in halo)? I get the idea that would be inpractical in cube.

Also I don't think you will get much work on the engine out of the open source comunity, Most open source fans are shitty programers, but are quite good at modding/map making/etc. So if you want us to help, finish the engine yourself and make it modable.

reply to this message

#75: Re: large maps

by fsmunoz on 03/09/2004 00:05, refers to #73

Lurch... "open source fans" come from a variatie of backgrounds. Some of the more competent programmers I know develop free software.

reply to this message

#76: ..

by Aardappel on 03/09/2004 03:20

model formats: the problem with own formats is that you need to write exporters for a variety of packages. I have a nice skeletal animation exporter (for 3dsmax) that could be worth adding to cube/saurbraten at some stage. Alternatively the md3 format is the logical progression of md2 (less quantisation problems etc).

yes sauerbraten supports huuuuuge maps, its not scale dependent like cube.

We'll have to see how this idea of making sauerbraten open source goes, if it turns out no work on it gets done then its just a failed experiment and I'll see what to do with it next.

reply to this message

#77: Re: ..

by pushplay on 03/09/2004 05:43, refers to #76

Open source projects rise and fall on their leadership. I know you like to take a laid back stance, but sometimes you have to beat the drum.

reply to this message

#78: .

by Gilt on 03/09/2004 16:37

Well, I might be able to help out. I won't have any time until at least the end of the month, but we'll see.

I haven't touched C/C++ in a while (been mostly playing around in VHDL etc recently), but I figure programming's pretty much all the same, and I suppose I'm a fairly competent programmer.

Anyhow, this is more of a "Don't pack up shop until I get to check the wares" type post; no guarentees that I'll do anything, yet.

reply to this message

#79: Re: large maps

by lurch on 03/09/2004 22:57

I know, I said "most"....
I think.

reply to this message

#80: VHDL

by pushplay on 03/10/2004 01:37

You just brought back a lot of horrible memories for me.

reply to this message

#81: Re: ..

by Aardappel on 03/11/2004 21:12, refers to #77

I don't mind "beating the drum"... that's what I keep doing.. but if I had time work on it actively and regularly myself I wouldn't have open sourced it. So some programmers will have to take initiative in doing some of the missing pieces in getting sauerbraten into a usable state.

reply to this message

#82: Re: .

by Aardappel on 03/11/2004 21:14, refers to #78

good to see you back Gilt... if not your coding skills, then certainly your mapping skills will come in handy at some point :) One of your quake maps was practically a "cube" map already ;)

Have a look at the documentation and code that's already there, and try out if you can manage to get something useful done. If so I'd happily add you to the commit list.

reply to this message

   Board Index