About those attempts to make an opensource or indy civ3 clone

[WildWeazel]

So if we make a game, it should be "inspire" by CivIII, not a clone. This means that we should make a new full design document to define what we want to do, and more importantly what we don't want. The danger in this kind of project IMO is to go in too many directions at once, be over ambitious, and not be able to finish it.
So it's better to start relatively simple. But not too much if you want a game with enough depth.

Completely agree here. We need something more than a clone and less ambitious than say SGF, which would have been a fantastic platform if it were feasible. Something along the lines of FreeCiv, able to closely emulate Civ3 (perhaps as a mod) but with more room for change, fewer assumptions about how mods should work, and maybe a few most-wanted features.

Wasn't there that whole thing a short while back where Antal1987 had found a way to access the source code? I'm not sure if he's still around though...

He didn't access the full source code, but he found a way to slip in references to his own code. So I don't think he could make performance improvements because that would require changing the code that's already there.

 

[Bengal Tiger]

Ah, I see. Thank you both for clarifying.

 

[nick0515]

So how many people do we have here who can code in the same code(s) and are willing to put in at least a small amount of their free time to start a new project?

If it was deemed that there are enough coders they could start work on a plan such as steph is proposing and seek feedback from the rest of us when they deem it necessary.

I'd say using crowd funding is only going to be viable for getting small amounts of money as a small incentive but not to pay someone to work on it as a job.

Keeping the graphics as close to Civ 3 as possible would mean that a dedicated artist would not be needed for success. Ie reusing as much as possible from the civ 3 fan created art.

I would personally want something similar to civ 3 but with all features working and a small set to of realistic new features plus the removal of arbitrary limits.

 

[Steph]

A dedicated artist IS needed during development phase.
Because
a) the dev may have to try something a little different from Civ3, small improvement (ex: add canals), and so they need a minimal arts to get started
b) even if we reuse civIII style, for copyright issue, we need to provide a minimal set of graphics. It may not be new graphics, maybe graphics from an artist here is enough, but he needs to be in the team and agree to donate his work for this project.

As for programmer, I suggest anyone interested write here a very short summary.

So:

Steph
- Can program C++/C#, prefers C#, good at algorithm/ data model / WindowForm.
- Limited graphics programming skills (now GDI+, a bit of XNA)
- Good knowledge of current CivIII internal structure (biq files).
- No experience yet for network/internet programming for a multiplayer version of the game
- Good knowledge of gimp, so can do some limited 2D arts (but no unit animation).
- Can do project management
- Very limited free time until next year after I finish my NTW mod.

the others who could contribute can do the same. Then we can see who can as opposed to who could.

 

[general-jcl]

I'm not a programmer and can't do very good graphics but i can help if needed
I will send to Steph some ideas too (in french and to long to translate here)

 

[WildWeazel]

For graphics I was thinking the game should keep the same perspective at least, but probably use different file formats, so we could have a tool to convert files and whole directory structures from Civ3 to this. That would allow us to easily reuse community artwork with the authors' permission or even distribute the tool so that people could simply import their own Civ3 art folders. We would still need a dedicated artist for new features and things like UI that would not be a direct port.


WildWeazel
- Java and C++ are my strongest languages but I like C# best, good at design/architecture and testing
- Would also be interested in a scripting API with a language like JavaScript, Python
- Very limited gfx/audio/art skills, programming or otherwise, but decent at GIMP for mockups etc
- Some background in UX/HCI for the GUI design
- Very little networking experience
- Reasonably good knowledge of Civ3 internals
- Experienced with remote teams, can do project management and engineering process stuff
- If I'm going to contribute code I strongly prefer open-source
- My free time is somewhat flexible, but currently it is committed to another gamedev project which may continue for several months or more

 

[nick0515]

Oh yeah I see so you do need a dedicated artist. I was speaking from a place of very little knowledge about what will actually be involved.

 

[Steph]

About open source: I'm not for it at start, why no after.
I think this kind of games needs several phase.
-1) Design: write a design doc explaining the main features and concept we want in the game, at a relatively high level, and define the big limitations. For example
Map : will be isometric grid, no notion of altitude, only type of terrain, each of the 8 direction can have an independant road , etc.
Unit: will be single graphics units, OR will allow up to 6 guys (munit concepts).
Many other topics (faction, techs, buildign, etc)
-2) Definition of the main coding structure: language, type of graphics, animation, structure of the data (inheritance or no?).
-3) Definition of work package, like "animation engine", "map engine", "random map generator", "movement", "combat", "technologie", "trade"...
-4) Detailed specification of each workpackage
-5) Coding of the work package

1 to 3) should be done more or less in common with brainstorming. 4) can be assigned to a member of the team, with review from the others. 5) should be assigned to a member of the team.

1) Can be public at start, 2 3 4 and 5 limited to the initial team at start.

Once we have the main structure in place, and we have a working mockup, then we could consider open source, but I'm not in favour of open source from the start.

For me, open source allows refining and expanding a good base, but it needs the base or we would go in any direction and "get lost" too soon.

However, we can set up a svn for the project so all the members can easily share the work and access it.

I won't have time to contribute to 5) before quite sometime, but we can at least start with 1) to get a clear roadmap. Question is : who will actively contribute to that.

 

[general-jcl]

I completely agree with Steph and i think i can contribute to 1)

 

[Steph]

The question is : do we have enough motivated people to start with 1), how can we discuss it?
Since the would be coders are all busy at the moment, maybe we can take a few months to finalize 1) and start discussing 2), and then see from here what we can do.

 

[WildWeazel]

I am ready to contribute to (1) at least.

Once we have the main structure in place, and we have a working mockup, then we could consider open source, but I'm not in favour of open source from the start.

For me, open source allows refining and expanding a good base, but it needs the base or we would go in any direction and "get lost" too soon.

I see where you're coming from and I can agree to this much. You're right about having a good base first, there's no need to publish something that is far from usable. But as long as the source will be released with the game I'd be ok with that. Open source doesn't mean you have to open up development to everyone though, you still retain control of your own source or can have gatekeepers who filter incoming contributions to keep it on track. It just allows others to see and use the code on their own if they want.

 

[Redback]

You need a project manager and some 'workers' (with all respect )

 

[Kyriakos]

^And warriors too. This is an ant-colony level organisation future project

I think we should keep as many of the graphics usable from what community works (with the creators agreeing they can be used) we already have

Of course some changes seem very logical, but maybe they are easier to organise/agree upon/find people to provide for them once we have a stable starting point (and the thread is about using those web-funding sites for backing, eg kickstarter)..

 

[Ozymandias]

Long Ago & Far Away, I was the IT project manager for a $32 Billion Wall Street system.

I can - and am certainly happy to - collate (and otherwise organize) any "wish list" into a master program document (as I'm sure could Steph!)

Steps:

1. Roundtable and finalize features & desired methods (e.g., how specific aspects of the AI would work.)
2. Organize those into "functional lists" (e.g., everything to do with "Roads & Railroads" cross-indexed with other hypothetical lists like, "Worker Jobs" & "Movement.")
3. Re-assess and define the workings of the final game.
4. Formalize those into a, "Functional Specifications" which combines all of said workings into a single, coherent document, which would then be used to write:
5. "Technical Specifications" of discrete blocks of coding to be done, what coding standards to adhere to (for those of you who are C programmers, e.g., no arrays of pointers to arrays of pointers )
6. Then the actual coding.
7. IDEALLY, while all the Specifications are being written, the Truly Dedicated will be devising scripts to test (a) individual "chunks" (a technical term ) of code (b) expected behavior (e.g., does AI artillery behave as it should ) and, finally, "How the complete consort dances together."
8. Test ... Test ... Test ...

"Project 'Management'" TBD

 

[WildWeazel]

something something outmoded waterfall paradigm

 

[Wolfshade]

8. Test ... Test ... Test ...

Right up my street.

 

[Ozymandias]

something something outmoded waterfall paradigm

Agreed, it was what I was taught in the '80s - but it has one distinct advantage: it's the equivalent of an AK-47 - it might be an old design, but it's utility is still very high, the weapon almost never jamming etc. because it's designed to take a beating. And I know the metrics by heart (e.g., take a "chunk" - again, that technical term of code to be written in C (equivalents can be readily devised for other languages) divide these into "virtual modules" approximating 100 lines of C each, and, in a professional (i.e., full-time work) environment, each 100 lines will take 6 weeks to complete over the course of full systems development (and not all those 6 weeks are the programmer's time) divided in a 1:1:2:1:1 ratio of functional description/design; technical design; actual programming; unit testing (i.e., making certain each coding function works on its own) and integration testing (making sure the game itself works.)

Managing up to 25 programmers as "recently" as the very late 1990s (yeah; agreed: but, as Neil Young sang - yeah, again - "It's old, but it's good") I never once had a project go over time or budget, and that extended (albeit wielding a "whip-hand" which would obviously and certainly be utterly out of the question with this hypothetical effort) to my management of project managers.

Lastly, a question to all, perhaps most especially Steph: Why tiles & not hexes? - And, BTW, if the answer is to use all those non-proprietary graphics in the db etc., I'm 100% in agreement.

- z

 

[Steph]

Personnal opinion: hexes are much better to handle the edge (like rivers, roads, since every side is equivalent, while with square we have different behaviour with corners and sides).
But the graphics can be reused. For units, we could almost use 6 of the existing animation, it's not really noticeable except for long narrow units such as a ship.
But the map graphics are definitely not reusable.

But anyway, with even with square (actually diamond) tiles, I'm sure struggling to keep the graphics at all cost is the best move since it means accepting the constraints coming from it.

A word about the map here.
If we go for "ugly fonctionnal map" (like having a tile fully paint in see/grass, etc without nice smooth transition), it would be possible to have a working engine much faster, and with a lot more option. Making it look nicer / smoother is a lot more complex.

 

[Steph]

I'd like to stress a little something. Look at the number of people who posted here, look at the number of posts.
Compare that to the team needed and time needed to make a clone of CivIII...

 

[Kyriakos]

True, but if there can be a fund-raising project (Kickstarter etc), then it is far more possible to find also other people to help, even from CFC or other game-forums

And since we won't term it a 'civ3-clone', maybe there could even be a CFC announcement/group for that. You were once a mod here, you know, and Plotinus is a high-Admin

[In case that happens it would be cool if we unofficially set some sort of team/volunteers for programming/gfx/testing/other. Personally i do not seek payment for helping with this, but i also am not working full-time on it either (there is simply no time now on my part, but i can gift all existent gfx by myself which can be used, and likely make some new ones too)].