How about a Mod Made Easy...

[gdijedi7]

Greetings, ya'll!

I was just thinking, has anyone made or had an idea to make a "Mod Made Easy" thing?

I mean, have something like the Civ2Gold editor that took the unit, and added it fully to the game with a click. It had an event editor that made it very easy to add events. You just clicked on the event and the effects and it was done. I understand that it may be more difficult, as we are dealing with animated units, not bitmaps, but surely there is a way to make it easy for idjits like me to do it?

[DougM]

i suspect the sort of person who would do this is waiting on the SDK before they proceed.

[Locutus]

Now why would they do that? The SDK has absolutely nothing to do with modding XML properties or events whatsoever. There is nothing in the SDK that would be useful for that.

People keep seeing the SDK as an essential part of Civ4 modding, as if you can't do anything without it. They just assume that if you want to do anything at all, you need the SDK. That's bull. There's very little for which you actually need it. 95% of the ideas anyone could come up with to mod into the game can be done with XML and Python only. Just look at the CtP community: none of the mods there use the source code, yet they completely transform the game. And Civ4 is considerably more moddable than CtP. And even if you do need the SDK, you have to be an experienced programmer to be able to use it, people just read a Programming for Dummies book are gonna have a very hard time doing anything meaningful with it.


More on topic, this particular idea is great but would take a LOT of effort to implement. I think there are one or two editors out there, but I haven't tried them yet, and I don't think they're terrible advanced yet. It's gonna take a while to develop such editors. But if you want it to be realised, it's probably a good idea to see what's there and to support their development by giving your feedback on them.

And it's not as if 'manual' XML modding is hard. You may need a day or two to get used to it, but anyone who can read and write, can -- after a short learning curve -- crank out new units or civs or whatever in very little time (assuming you have pre-existing graphics of course -- developing new graphics takes more skill and effort, but there are plenty of pre-made graphics available). Just because it's not point-and-click doesn't mean it's hard.

Python events are harder to do, but with XML alone you can come a long way.

[DougM]

if for no other reason

You might assume that the SDK has access to the XML loader code, and that along with the next patch we might expect widespread changes to the XML. I mean, so much of the XML is so inconsistent, one can only hope that they're improving it as part of this release. I think everyone expects it, so why work hard on anything now when you can expect big changes soon? In any case, the SDK is clearly likely to be a big piece of leverage on attempting a project like this, why do a ton of work today that might turn out to be comparatively trivial tomorrow.

[Bhruic]

Originally posted by Locutus
People keep seeing the SDK as an essential part of Civ4 modding, as if you can't do anything without it.

In some ways that is the case - but mainly for AI issues. I have dozens of mod ideas that I'd love to try out, but I haven't. Mainly because I know the AI wouldn't be able to handle the rule changes.

I suspect that is true for many mod designers.

Bh

[Locutus]

I'm not sure about the loader code, that may or may not be part of the SDK. Would be somewhat of a time safer but only somewhat. It's really not that hard to write, I have a fair amount of experience with this with other games and applications, and most of those didn't even have any kind of standard (like XML), for which there are in most languages (certainly in Python) packages that can help you with this.

If you're counting on major XML changes you're going to be disappointed. I think some values have changed (but obviously any editor should be able to deal with that), maybe there are 1 or 2 structural changes, but that's as far as it'll go if I'm not very much mistaken. Certainly nothing that should stop anyone from working with on an editor.

AI modding can for a good part be done through Python, and clever design usually negates the need for major AI reforms (as long as you're not completely overhauling the game, but we haven't seen anyone do that yet). But of course you're right that the SDK will give you infinitely more options in that regard, that will be the thing it will be most useful for.

[Bhruic]

Originally posted by Locutus
AI modding can for a good part be done through Python, and clever design usually negates the need for major AI reforms (as long as you're not completely overhauling the game, but we haven't seen anyone do that yet). But of course you're right that the SDK will give you infinitely more options in that regard, that will be the thing it will be most useful for.

I can't really think of many ways the AI can be manipulated via Python. Just as an example, one thing I wanted to do was modify the way culture spread. I was thinking of having military units apply culture to the square they were on. But that's a pointless change, as the AI won't know that its military units spread culture, so won't try to use them to do so.

So I think the reason you haven't seen anyone do that yet is because we can't - yet - without the SDK.

Bh

[Locutus]

Yeah, that would be a rather complex thing that would probably be easier to do in the SDK. But you can rewrite unit movement orders with AI_unitUpdate in CvGameUtils and unit production with AI_chooseProduction, so technically you can completely rewrite the entire unit control section of the AI in Python.

[Bhruic]

I guess only if you know that you can do that. There's no documentation I've seen that says anything about AI_unitUpdate being used in the game, or what it's used for, if it is.

Bh

[Locutus]

It works the same as all other functions in that file, I assumed it would make sense (although I did intend to post a few examples which I never got around to due to RL issues) -- they are 'pre-events': they're called just before certain events occur (e.g. isVictoryTest is called just before the game declares someone a winner, AI_chooseProduction is called just before city production is set for an AI). If these functions return False, as they do by default, nothing happens and the event is executed normally. If you make them return True, the whole event is canceled and you can use the rest of the function to redefine the behaviour that that event would normally do (e.g. by pushing unit orders manually for AI_unitUpdate).

[Bhruic]

Ah, that makes sense. I think it was mainly that some of the function names weren't terribly descriptive that was my main problem. I mean, isVictoryTest is pretty easy to guess. AI_unitUpdate isn't. I still think some of the things I want to do will require the SDK, but I might be able to do some Python pre-preparing.

Bh

[Locutus]

The names pretty much made sense to me, but I guess for an outsider not familiar with Firaxis nomenclature (which is largely hidden in the SDK) that might be a bit trickier

I think if you want to completely rewrite the AI, it's best done in C++, then Python's just a waste of time 'cuz you'll want to star over when you get the SDK anyway. If you have specific stuff you want to address now though, you can probably make a start on that...

[Addled Platypus]

Originally posted by Locutus

People keep seeing the SDK as an essential part of Civ4 modding, as if you can't do anything without it. They just assume that if you want to do anything at all, you need the SDK.

Yep

look at the completed mods Apoly members have done