[C4:AC] Keeping Civ4 changes?

[wgabrie]

Originally posted by Chaos Theory
If all you want is Civ with SMAC flavor, you could just do it on Civ 3.

Acually, Civ IV offers Greater modability and has a social enginering option which is essential for SMAC flavor.

[Blake]

And if the team decides not to keep one of your desired cIV aspects, what are you going to do then?

Why, I'll code in the interceptor crash bug and make automated formers retarded out of spite
(that IS sarcasm! See the rolleyes smiley??? nevermind...)

Also, do you not support
our mission statement, or do you not consider yourself included in our group?

Ultimately, I will code what I like and you may take it or leave it, you are welcome to exclude me but I suggest not doing so out of spite, I certainly wont refuse to donate code just because I refuse to tow the party line. However you should understand that mostly I am stating I am not going to re-implement any relatively inane feature, or support doing so. I'm also not going to invent any entirely new systems or modify systems to be like neither SMAC nor Civ4 - that isn't on my agenda! you may quote me on that (okay to clarify I will certainly implement new systems but I wont push for them to be included in C4:AC).

There will be certain SMAC features I'll be very eager to implement (orbitals, ecodamage/worms, unit workshop, atrocities, social engineering etc), a few I'll want to forget (features that only really cause micromangment, a rehauled growth/happy model would be quite nice, should Civ4's be decent) and some I'll be a bit undecided on - Wonder build style, combat style and such. There are also some features I have a personal vendetta against - especially crawlers - but I would support reimplementing such systems if they are very characteristic SMAC and add interesting dynamics.

If it proves nessecary the project will fork early into the realists who support keeping Civ4 features and the diehard clonists, one fork will no doubt wither and die (hint: it'll be the second one as it's a tad harder to be enthusiastic about). However I doubt this will happen, as once people play Civ4 they'll like some Civ4 features much too much to even consider reverting to the SMAC way, or atleast to seriously question why they want to revert. "Just becuz" will start to look weak.

In the long run I'll also fully support a comprehensive options system to chose from 2 or more implementations of certain systems (SMAC style, Civ4 style, New Style(s)) - this would be a much, much better alternative to forking (as forking causes dilution of talent). In practise some options will pretty much become defaults for multiplayer games and such. For example Crawlers could have the Options Civ4 [none], Classic SMAC [overpowered], New SMAC [rebalanced]. In practise Classic SMAC crawlers would probably become the defacto default for multiplayer (because most people like crawlers as they are) but the New style would cater for people like me in SP.

[binTravkin]

As I already said once, clone is theory, praxis is an approximation.

I for one am not going to program stupid AI/formers or any bugs in.
If I see the cIV system on the issue works better, I might modify it, but not copy+paste SMAC bugs/inconsistences.


And this is getting too hot IMO.
Let's face two facts:
1.We have limited resources (which means we can't copy+paste SMACX, we will have to use at least some of the cIV base code).
2.The more you ramble here the more displeased with the project people get. Let's just learn Python and C++, read reviews and do other 'while we are waiting' stuff, but not shout at each other just because our opinions differ.

[Senethro]

BinT is right, and the mission statement was developed to reduce conflict, increase group coherence through a simple goal that will be attractive to as many people as possible. The people deciding if a feature from classic SMAC should make the final version are the coders. If they say something is not possible, we cut. Otherwise, we keep it the same as classic SMAC and log it carefully so that if someone wants to undo it, they can.


drekmonger: I think you're being too negative. Just to take an example, sea bases are probably as easy as enabling settlers to walk on water tiles and enabling them to use the "build city" function in water. The unit workshop will be much harder...

[binTravkin]

If they say something is not possible too hard or long to do with our limited resources, we cut.

Edition..

[drekmonger]

drekmonger: I think you're being too negative. Just to take an example, sea bases are probably as easy as enabling settlers to walk on water tiles and enabling them to use the "build city" function in water. The unit workshop will be much harder...

Apologies for being unclear. I was suggesting that sea bases would be something that probably could be easily modded into the game. You get a cool SMACish thingy for an afternoon or two of coding. (plus whatever time is required to do the modelling for the seabases--but placeholder art can be used in the meantime.)

Ditto Fungal squares, Monolithes, and the unique terrain squares (like the Borehole Cluster, or the Unity Crash Site.)

I'm suggesting stuff like the unit workshop and the Smac combat model should not be included:

* the Civ4 conception of units and combat seem adequate for a game that *feels* like smac
*both systems would require extensive work to implement
*it would take extensive work to "train" the AI to act intelligently. Don't underestimate the amount of work required to build and test an AI script that isn't rock-stuipid.

Sorry to be harsh, but:

Your mission statement means nothing to me, as I find it unrealistic. (and yes, that means I do not consider myself, nor will I ever, a part of a team who's stated purpose is to clone SMAC). You simply haven't considered the effort required, esp. in terms of the AI scripts. Question is, do you want a playable version of the mod done in less than six months? Less than a year? Less than three years?

Remember that most of the help you'll recieve on this mod will come from people working in their spare time, and often from people who are not professionals. (...if I'm wrong about this, then ignore all the above....)

Personally, I'd want to see a playable version in a month, or at most three. If you plan for a longer time frame on projects like this, things tend to get....out of hand. I'll offer up Stella Polaris and my own team's FreeOrion as examples.

[edit: I mean, a month or three after the release of the SDK in Jan.]

The big advantage of working off of Civ4 is you can use a bunch of stuff already implemented in Civ4--don't throw that advantage away by trying to recode everything. Keep as much as you can while still retaining the special Smac flavor.

My suggestion: have one tyrant, your most experienced, valuable coder perhaps. He makes a checklist of what is to be kept and what is to be changed in the "official" mod. In other words, One Linus to Rule them All.

Or you can work on this thing for a couple of years, and have Firiaxis deliver something better before you're half-way done. Up to you.

[binTravkin]

Regarding the time and playability issues I think our mod will start with something like v.0.1. which will be a fully playable cIV with first SMAC thingies implemented, then it will advance in stages ultimately coming to v1.0 where we should be done with most reasonable SMACXish implements.
After that we move on to harder-to-implement things and probably something different(if you know what I mean)..

[Senethro]

Though discussion of this is going to have to wait for several months, I think a timetable where graphics is second last and AI very last is best

[binTravkin]

Well, you now fail to understand one thing:

There are different specialities.

Do you offer the graphics guys to wait up until we finish all coding?

That'd be complete waste of time and their eagerness!

I even dare think the first things that'll change will be exactly graphics. Probably not 3D, but all those background patterns, menu styles and probably even some of the terrain.
Those should be the easiest to change things - recognise format, draw graphics, change format to needed, rewrite originals.

[drekmonger]

Originally posted by Senethro
Though discussion of this is going to have to wait for several months, I think a timetable where graphics is second last and AI very last is best

Saying the AI is very last is same as saying "never." It's a nasty, huge job, and scaring up anyone with the skills and desire to do it (sans hot coals or fat paychecks) is next to impossible.

Call the mod mutliplayer only, if that's your intention. You gain quite a bit of leeway if you don't have to worry about the AI.

Graphics should be concurrent. I see that at least one person is already working on a way to rip assets out of Smac. It wouldn't hurt anything to start modelling now, as you can be fairly sure there'll be a way to import popular modelling formats into Civ4.

[Senethro]

BinT + drekmonger: While concurrent graphics would be ideal, I wasn't even aware we had an experienced modeller aboard yet and that was the main thing shaping my thoughts.

Perhaps we should open a discussion on this. Previously I was hoping that small and early success would generate further and greater success and attract more people to the project. With the professional programmers that offered their support early I was expecting the programming to lead the graphics. Is it more realistic to think that with a 3 month headstart on the programming that our 2D artists can make something sufficiently SMAC-like for an early first release?


As for the AI, will it really be that hard to make a bad AI, which is what SMAC has already? What will be hard is making an AI on par with what some people are anticipating Civ4s to be, I agree there.

[binTravkin]

Is it more realistic to think that with a 3 month headstart on the programming that our 2D artists can make something sufficiently SMAC-like for an early first release?

It is sufficent to think that if for example Wgabrie got cIV now, he'd be ready to replace quite many of the pattern and other simple 2D graphics in a couple of weeks before programmers have even investigated the existing code properly.
It takes very little to replace 2D patterns - you just need to know which ones are they and that is pretty obivous usually and you need to draw the ones you want to be and just paste over.

[binTravkin]

As for the AI, will it really be that hard to make a bad AI, which is what SMAC has already? What will be hard is making an AI on par with what some people are anticipating Civ4s to be, I agree there.

Initially we wont be coding anything for anew.
We will be adjusting the existing code so that it fits our setup.

Once we get to features which have no prototypes in cIV, like Design Workshop and Supply weapon type, only then we'll be coding something new and chances are that even for those features >50% of the code will be just a copy of the old.

[Chaos Theory]

Originally posted by drekmonger
Saying the AI is very last is same as saying "never." It's a nasty, huge job, and scaring up anyone with the skills and desire to do it (sans hot coals or fat paychecks) is next to impossible.

Bull. It's easy to write a rule-based AI that, whiile quite vulnerable to anyone intimately familar with the rules, will still be a match for the SMAC/X AI. From there, just write a learning AI (yes, I know how), and allow it to learn. That last part will involve considerably more effort, but yield a very good AI eventually.

[Senethro]

Originally posted by Chaos Theory


Bull. It's easy to write a rule-based AI that, whiile quite vulnerable to anyone intimately familar with the rules, will still be a match for the SMAC/X AI. From there, just write a learning AI (yes, I know how), and allow it to learn. That last part will involve considerably more effort, but yield a very good AI eventually.

I imagine you suggest using whatever routines Civ4 does for forming attack groups and attacking, and otherwise giving the CPU a braindead route to pop-booming with tree farms?