[tcaudilllg]

Now I know you guys are probably gonna be laughing, maybe even scoffing, but what would you guys say to organizing a general purpose C++ library with specific focus on game engines? Now by engines, I don't mean the graphics, nor the music/sound playback. I mean the algorithms that link all of those together. Now I know I'm going to hear "but isn't the freedom of the design one of the best parts of the experience?" and to those people I would like to emphasize the reality that will be possible if this library is assembled. We think of the instructions in our engines as the atoms of game development, but in truth these instructions alone are but unions of smaller instructions that we do not readily perceive. Everything is a union of smaller things, and if we define the atom of our work as a larger union as opposed to a smaller one, then we can produce larger works more quickly, if only in a draft state, like the work of a sculpturer on a slab. We can have a huge variety of complex game engines practically overnight, and each part itself a concrete, stable whole. We can produce worlds the like of which we can scarcely imagine now, any more than we imagined the 3D worlds of today in the days of the NES. We can create worlds at will, and without the hamstrings generic creation tools force on us.

I don't have the ability to do this myself. I don't think any one person could in any reasonble length of time, unless they were devoid of any commitments at all. I have commitments myself, as does everyone here, and I think that if we work together we can produce something truely historic and amazing.

It would be a long haul, definitely something that wouldn't be done in a few months. It would take little bits of effort, in our free time, from all of us. But only little bits. It would be better to make this a very broad opensource coalition, in which more emphasis was placed on organization of existing resources than on code writing. Those engines that do not exist would have to be created, of course, but we could do much by negotiating with existing institutions and individuals and groups to get the work that we need for our system, without having to re-invent the wheel.

Does this sound like a capable idea? A workable idea? A worthwhile idea? I want your help to make it happen, not today, not tomorrow, but someday.

Who is with me?

[RuneLancer]

What you fail to understand is that no two game engine is alike and that these will always vary, both in terms of needs and implementation, with the features it will implement and how it uses these features.

Such a library would be useless for all but basic tasks, at best.
_________________
Endless Saga
An OpenGL RPG in the making. Now with new hosting!

[tcaudilllg]

Well I didn't expect you to understand, RuneLancer, because you are a constructionist/INTP. You didn't try to fill in the blanks of what I was trying to say, and you should always, always do that whenever I say anything, because I *always* expect people to fill in blanks when I'm talking about something very broad in a brief statement.

Assume I know what I'm talking about, and we'll have less misunderstandings.

Anyway... I expected such a response.... I really don't think that forums are a good way for anything useful to be discussed, except for posting purposes, so I'm going to stop using them. The anonymity is apparently too great of a shield, and attacks are freely exchanged between words, not people. Message boards alone cannot sustain a community, that is clear. Personality just doesn't come through, except along strict typology lines. The INTPs are always saying "hold on" or "this will never work" or "Maybe that's not such a good idea". The ENTPs are always attacking anything that isn't a part of the establishment they have aligned themselves with; the INTXs assail anyone who isn't using the "cutting edge" to do something; the ENTJs spin their wheels with pipe-dreams of massive community organizations; the psychos are just psycho; the IXFPs try to back off at any sign of disagreement; the ENTXs plot and scheme behind people's backs; ...and well, the ENFPs are pretty much in league with the ENTPs.

This is impossible. Nevermind.

[Nodtveidt]

I think it's a pretty decent idea, myself. I'll help you out with it, I'll make some time available after the eROSE closed beta run (starts tomorrow and lasts about 2 weeks).
_________________
If you play a Microsoft CD backwards you can hear demonic voices. The scary part is that if you play it forwards it installs Windows. - wallace

[RuneLancer]

This is a little off-topic, but have you ever considered dropping your pre-chewed personas and preconceptions about how there is nothing wrong with your own ideas? Maybe people aren't flaming you after all, and only really bring in the big guns when you start treating them like they are?

I can't say I've been here nearly long enough to make statements about other people on the boards, but it took me little more than a few days before I could become close-minded enough towards you to label you negatively and feel predisposed to disagree with you. I may only be speaking for myself, but maybe you'd find more people agreeing with you if you stopped trying to shape the field of programming into your own personal image of what you feel it should be, and give the tried and true methods a chance before trying to invent your own?

That being said, I'm not trying to flame you. Like I said, it's not my place to judge: I'm a noobie, essentially, to these boards. :) Just some advice. Take it or leave it, it's none of my business wether you label this post as an ADJX reply and ignore it or whatnot.

Back on topic, the problem with this idea is that there's no way to cater to every project's needs without being vague and treating only of global problems. For instance, given a speed X and a degree of friction Y, along with a gravity Z, a function could return the amount of acceleration an object falling would receive. Handling the actual object, however, would still have to be coded by someone. There's only so much a library can do without becomming a game engine and restricting the programmer's control over their creation.

The other problem is that such libraries already exist. There are literally millions of code libraries out there. Take OpenGL, for instance, which is a VERY simple graphics library that can render a scene with little effort than specifying a few polygones and telling it "turn on this feature" or whatnot. It'd be far easier to cumulate a database of libraries along with their uses and features than to reinvent the wheel.

Of course, there ARE domaines that aren't given much coverage yet, and plugging in a few holes here and there would be beneficial. But to reinvent the wheel is as counter-productive as it gets.
_________________
Endless Saga
An OpenGL RPG in the making. Now with new hosting!

[Gooseman]

[Sirocco]

[tcaudilllg]

VERGE -works-? You're kidding, right?

Really, aren't you kidding? :? I completely disagree with your "ecosystem" analogy to game engines. We're not making the ecosystem; we're making the fundamentals that -make- the ecosystem, right? Whereas, the ecosystem is the balance of the gameplay, and the style of the graphics, and other completely design-related topics.

What is important is that the engine be intuitive. VERGE is not intuitive. It's half an interpreted game library and half of half a tile-based engine. VERGE was made to make Final Fantasy III, lest you forget. Since the original version, it's original focus has not underwent a serious makeover. The most impressive VERGE project I've seen was the Final Fantasy remake that fell into limbo some years ago. Granted it was probably the stoke to the fire that prompted Square to completely remake its earlier games (because the demo was GOOD), but even that could have been made with a C++ library as easily as with VERGE. The amount of custom-designed code that must accompany VERGE products is telling, and that's what I'm aiming to reduce in the game development process.

To be honest, I've already completed my own 2D engine to my satisfaction. In JS. Now rather than just port my own engine over to C and have some people take a look at it, and maybe use it, but have many others completely confounded by it (as we have in all cases today with non-profit work), I want instead to use my understanding of what is needed in a 2D engine as an organizing principle about which I can work with others to create a game engine that is good for everyone. Although I have completed very little of a 3D engine (because it is not plausible today to conduct one inside a browser, and SVG is a performance hog), I think and believe that I have what it takes to successfully communicate with others about the design of a grand-coalition library that people of many different types and persuasions can enjoy and use to create games that focus more on design of mechanics than on design of instruction code.

I believe that if I work with people one-on-one as opposed to trying to get everything done on my own, then I can be more effective in my goal of increasing the quality of independent games. :) That is what I am working to accomplish with this project. I'm not good at getting stuff set up, like for a community and such, because I am usually oblivious to the social cues and miscellaneous other details of social organization. I need help. :) I know that DevX has all kinds of things on his plate right now, and he's already done lots for Worldyne, so I don't feel it would be right to ask anymore of him for this project. However, our work on Dream Destroyer is considered by me to be a part of the pool from which code may be drawn. Indeed, it is the model: Dream Destroyer was made with Allegro. We organized Allegro functions with C/C++ algorithms. Our engine is quite flexible, although primitive in many respects. (menus come to mind, for example) What this project is aiming to do, is bring the algorithmic organization to a minimum. Let's have our algorithms broadly predefined. Put more emphasis on the planning of the way the algorithms run into each other--that is, the game design--rather than on the way the algorithms operate. Less attention to the details at the code level, and more attention to the details at the game level.

[Mark_Y]

[RuneLancer]

And it would be of no surprise that he'd ignore the two posts directed towards him directly as opposed to his idea. :)

Consider looking into game engines. Many allow you to skip the tedious coding and get down to business. Most are configurable enough to allow quite a lot of freedom.

No, I'm not talking about bloody Verge or RM2k, or whatnot. Those are for newbie developpers and lack power.

http://nexe.gamedev.net/directKnowledge/default.asp?p=DirectX%20Engines
_________________
Endless Saga
An OpenGL RPG in the making. Now with new hosting!

[tcaudilllg]

I looked into them. They aren't what I'm talking about.

For one thing they are 3D. I think it a good idea to just focus on two axes for now. We'll call 3D the step up, once we've got a decent 2D engine in place.

[tcaudilllg]

I've begun searching for candidate libraries from which to negotiate the withdraw of code for this project. 3D libraries are not the focus at this point, so although I would appreciate the mention of as many *free* libraries as possible to me, libraries that focus on 3D interactions (beyond the strictly simple gouraud shading/texture mapping classic stuff) are not of interest to this project at this point. Admittedly, I know very little about the latest innovations in 3D technology, and until I do I am ill-suited to ascribe its relevance to games.

[Bjorn]

[tcaudilllg]

Well I'm working to make connections, make collaborators.

I'm not sure whether I should start a SourceForge project or not. I don't want people randomly making changes to the library at will. Rather, I'd rather it be done in a very open-ended community agreement. On the other hand, I need a base of operations about which to coordinate efforts.

I could set up a message board somewhere. Does anyone who believes this project is necessary have space for a message board available? My employment situation is bad, so it would be detrimental to the project's organization for me to schedule any meetings. Impromptu access to collaborators would be much better for me, like through ICQ/AIM/other IM. I think all that I really need to organize the project is a message board. Bjorn, considering the convention board got nowhere, and no one is showing interest, why not take it down and set up a new board for this new project, which shall be known from here on as "The Coalition Game Library".

[Rainer Deyke]

Aside from the very vague and all-encompassing goal for this project, I think there is one very fundamental problem that will prevent it from ever getting off the ground:

I do not trust you to manage any technical project. Therefore, and I think I speak for the vast majority of potential contributors here, I will not submit any code to a project managed by you.

Sorry if this sounds like a personal attack, but it's the truth as I see it.