A chronically broken epic mod. Am I the only one to notice?

Look at this picture.



Are you even aware what you see?

I tell you what you see. This is the turn 236 of my snail giant earth map game. I have been button of the points list since I have seen any rival's points. Owning to me having chosen the monarch difficulty. And still I am far out regarding any rivals military.
I could rush Egypt next to me with no trouble.
I have not focused on my military build up for a second. I focused on development! All the time. On monarch, severely lacking points, I literally CAN NOT HELP outclassing anyone military NO MATTER WHAT I DO. Except playing for loosing I suppose.

I am making such a fuzz because this has been such a major history in this mod. Again and again I find the AI ridiculously inept in handling any kind of military capability.

And now I am downloading the latest version and AGAIN - the military efforts of the AI are not even a joke - there are teletubby clowns just waiting for someone with half a brain ripping through them.

And my question is loaded: Loaded with praise! Because I like sooo much you do. The building list is out of hand. Of course. But I love the depth and complexity you add to civ4. I love the prehistoric age. I love all the grandeur development. The intricacies. The future era!
Short - I love what this mod stands for.

What I do not understand, however.

is how you can keep on piling such a CRAP combat AI on it - RUINING the whole thing. Your mod is so stuffed it is necessarily clumsy, too clumsy for delicate play. What is left is combat play. And then you UTTERLY ruin that.
And you do not even seem aware.

I am tired of this crap. I do not understand. How the frack is this possible. Such great work in such great detail. And so little fundamental concern for the very basics! This sucks. You are great. But you also suck. You drag down your whole project and do not even seem aware of it.

How the hell is this possible???

We are aware, but stuff takes time.

I understand your frustration, and maybe one day, it will get better on that front.

First of all:

I have not focused on my military build up for a second.

Click to expand...

You sure have a lot of units for someone who hasn't focused on military build up. Maybe most of what's inside the city is subdued animals? We're not seeing that. But of course, it's taking building wealth to be at +5. Staff a decent army out in the field and economically you'll have trouble.

Have you tried to rush your neighbor yet? You might find that it's not going to be as easy as you thought.

Mind you, I do understand there is a strong early war capability with spiked clubmen that others have noted before. Of course, having the AI be completely prepared for that would mean delaying the AI's development quite a bit, therefore making it so that players that opt more for a development focused approach rather than early conquest will find the AI wouldn't keep up because it's focused so much on trying to defend itself.

It appears to me that many human players get away with the risk of keeping very few city defenders at first. Sure it's a risk but it's also a way to pull ahead so that before the first war comes to their door they have better units to present at that point.

All part of the 'interesting decisions' being fairly difficult for the AI to please all players.

Yeah, they could use a little more defense. Specifically, they would benefit greatly from a defensive response stack or two that moves to where the incoming invaders are most likely to attack, swelling the defenses in the targeted city.

Yes, they certainly have some severe strategic flaws in hundreds of ways. I've just spent the day working on resolving the manner in which it handles property control units so poorly on multiple issues. So work IS being done on the AI and much much much more is coming.

However, to say that such AI work is 'the basics' is the farthest thing from the truth. AI is actually the most complex portion of the mod and only a few modders can do much about it.

I am one of them, and the only one that spends much time on the mod. I'm also still learning a lot and diving into AI work very infrequently so as to protect the integrity of the code. Though I'm beginning to overcome this reluctance.

Personally, my philosophy has always been, if you're going to arrange and perfect the AI, then do it under the game rules you want it to be working under. So development of game rules have come first on many occasions.

But as a model of how development is taking place, this version is exemplary so far. Game rules for stealth units and those that counter them have been fleshed out. The units involved in this style of conflict have been fleshed out. The promotions have been fleshed out. Now, the AI is being fleshed out. Pretty sure by the time I'm done with this version, while core military will still need a lot of addressing, the AI will have more than enough capacity to surprise and disturb as well as counteract the strategies involving these kinds of units that have historically been so easily used to undermine them.

Next version I'll be committing efforts to the core military units. Might seem silly to have put the stealth stuff over the basic military units but the most glaring flaws were actually on the stealth side, among other major balancing issues to address there.

This is a work in progress. But we're finally making real progress. So before criticizing too loudly, realize we've not yet stated this is a mod that's anywhere near 'perfection'. Releases are made as regularly as they are (a regularity which has diminished in frequency dramatically of late) purely so that the playtesters of the mod can give it another run and see what flaws they find this time. Not all flaws, however, can be fixed overnight. Some take years of daily effort to resolve.

The mod is still becoming what it has been envisioned to be. Unfortunately for some, the AI is a last step or you continuously must redesign the AI with each change.

As for most of the development that may make the mod seem like the 'basics' have been overlooked in favor of 'loading up' the mod, much of that work has been done by the majority of the team who don't have the capability to work on the AI at all.

One of the problems is unit cost vs city production ratio. There is tons of buildings which increase raw production or help training units that building huge army in few turns its not a problem. Its huge advantage for a player because he know what to build and he can build it very fast. But AI builds only siege rams and healers (or similar units).

Sparth said:

One of the problems is unit cost vs city production ratio. There is tons of buildings which increase raw production or help training units that building huge army in few turns its not a problem. Its huge advantage for a player because he know what to build and he can build it very fast. But AI builds only siege rams and healers (or similar units).

Click to expand...

Shouldn't be quite as bad now as half a version ago. According to some adjustments and recent reports.

Certainly much more work to do to improve AI army compositions. Some of that will be done here in this version but much of it will wait til 38's review cycle.

AI may seem like a basic for a player who needs a good AI to have an interesting game at all, but for a developer it is by far the most complex, intricate part of all. Of course, we are not speaking about an artificial general intelligence, where people so far haven't been able to come up with a system rivalling the mind even of a 5 year old, but it is still general enough to be a valid point.

We have more than 30 (I think) boolean choices on the custom game screen, making for more than a billion different combinations and some of these choices (Revolutions, anyone?) change the game completely. The AI must be able to handle that. We have hundreds, if not more than 1000 different units, the map can contain tens of thousands of tiles, these tiles have different properties, research changes the way you play the game and you can create completely new units.

As opposed to, say, chess, where you have a single set of rules, 6 different units, 64 tiles which are all the same, there is no research and no creation of units. And the AI has only now reached the level of humans, barely. After decades of research of literally thousands of people.

Edit: Here's an idea: When you feel the need to ask "Am I the only one to notice?" it is high time to take a step back and think carefully. Of course there is always the possibility that you are a lone candle in the dark world, but that's just one possibility. OTOH it is perfectly possible that you have not seen enough of this mod, and saying you could steamroll your enemies is easier than showing you can do it. Good luck with the enemy's Tribal Guardian (spoiler, I guess?).

A chronically broken epic mod. Am I the only one to notice?

Click to expand...

You are 236 turns into a 6000 turn game (.039% of game played) and you are fuming about the Combat AI on Monarch level?

And you give No Details on what Options you selected to play this game under, or what version it is. Is it stock v36? Are you using the SVN?

You're crying over such a small sample of what is yet to be in this game as to be, bluntly, ridiculously puffed up.

Just because the Empire list shows you to be 2.0 to 2.3 times stronger militarily could be based solely, at this stage of your game, because maybe you have more subdued animals in your city than the AI that is represented on that list. Or that perhaps your 3 clubman group has survived animal attacks better than the AI that shows a "military power ratio" to you.

You've made some seriously bad assumptions over a list that virtually means "spit" at this point in the game in regards to Combat strength. You Are listed at the bottom of that list and that means there are many factors Besides combat strength ratio that places you At The Bottom. Yours is Not the superior Empire that you swear you are because your number of Mil units is more than your "known" neighbors.

Maybe after a 1000 turns you may have a better perspective, maybe. Right now you come across as a rant/troll because the AI has not steamrolled you after less than 4% of the game played. Do you even realize how ridiculously over simplified that is?

"If" you are playing stock v36 then I Strongly Suggest you get the latest Patch in the v39 Crime Patch thread. Oh and If you play another game, leave Rev Off and play at Emperor or above.

And finally if I've hurt your feelings with my response that's just how I am, frank and blunt. It don't mean spit to me that you've been here at CFC since 2006 or that you have 9000 posts, not when you come in and post this kind of thread and your involvement over the last 8+ years in C2C's development has been nil. In fact your OP is just that, a rant with some sprinkles on top and nothing in it is constructive or informative.

Have a nice day.
JosEPh

Good post Joe! So true...

Nice Joe

The A.I. at the first 2 era's do tend to lag behind a little with Tech when you do focus primarily on advancing through the tech, but at the Third Era they are catching up and surpassing me in tech, even though i got my entire empire geared towards Tech and construction, so that is nice,
I Rarely "play" the first stage of the game, while i do love and have turned on the Minors option, i generally keep to myself in the first stage, meanwhile steadily increasing my platform of expansion and power, Writing does enable the A.I to trade Tech between each other (even if you have Brokering turned off and i do not like Brokering), making the tech race much more complicated

JosEPh_II said:

if I've hurt your feelings with my response that's just how I am, frank and blunt.

Click to expand...

You should add this to your sig. Joe.

Harrier said:

You should add this to your sig. Joe.

Click to expand...

Yeah I probably should. Something like this perhaps:

"If I've hurt your feelings with my response that's just how I am, frank and blunt and... and Slow......Watch Out!. It's Not Y'uns Turn!!!"

EDIT: Done!!!

JosEPh

I'd hate to see this guy playing Ver35. He would have had fit then for sure, especially playing with ruthless or aggressive AI on. The AI has come a LONG way since then and is in good to great shape in my opinion with the latest SVN. In my latest game one of the ai's isn't far behind me at all and with me having my hands full with barb cities that have tons of units in them, that Civ may very well catch or pass me soon. And by the way this is on noble difficulty so the ai should not have any advantage. I'd probably be getting left behind on Monarch or higher.

I really do recommend Terrain damage. It makes the Sahara rather impassable and that only really leaves you with Egypt as your enemy. Who, at the moment, has about double your score and I suspect that his Tribal Guardian is still intact, and good luck with that. Probably has some minimum defensive capabilities as well.

So you're probably going to lose your 'army' because its brutal to do early game assaults. Usually so much so that I avoid doing so because its too costly. Both in the long range support of a new city and the amount of production used. (That's just me though though, I tend to be a builder/trader... and if I can't get it by trade, then my imperialistic tendencies pop up).
Also the same reason why the AI isn't bothering.
If you still are having issues with the AI by mid Ancient to late classical in terms of it military capacity... then up the difficulty. Past that, the wars just get better.

Acularius said:

I really do recommend Terrain damage. It makes the Sahara rather impassable and that only really leaves you with Egypt as your enemy. Who, at the moment, has about double your score and I suspect that his Tribal Guardian is still intact, and good luck with that. Probably has some minimum defensive capabilities as well.

So you're probably going to lose your 'army' because its brutal to do early game assaults. Usually so much so that I avoid doing so because its too costly. Both in the long range support of a new city and the amount of production used. (That's just me though though, I tend to be a builder/trader... and if I can't get it by trade, then my imperialistic tendencies pop up).
Also the same reason why the AI isn't bothering.
If you still are having issues with the AI by mid Ancient to late classical in terms of it military capacity... then up the difficulty. Past that, the wars just get better.

Click to expand...

All very good points.

Additionally, the AI will want a minimum of 2 additional defenders IN all cities now. This should extend the game's difficulty a bit where it comes to trying to rush them right away. There's some more AI to come that will further improve their defenses. On Size Matters, the base of 4 city defenders will very likely create a situation where they will have merged defenders up to the limit of merging. We'll see if they can manage this once units are made a bit more expensive in general but for now it will definitely be more challenging since they shouldn't have a hard time handling it at all.

Thunderbrd said:

All very good points.

Additionally, the AI will want a minimum of 2 additional defenders IN all cities now. This should extend the game's difficulty a bit where it comes to trying to rush them right away. There's some more AI to come that will further improve their defenses. On Size Matters, the base of 4 city defenders will very likely create a situation where they will have merged defenders up to the limit of merging. We'll see if they can manage this once units are made a bit more expensive in general but for now it will definitely be more challenging since they shouldn't have a hard time handling it at all.

Click to expand...

I actually had a good game recently. The AI isn't perfect but it was doing a respectable job looking threatening.
I did have revolutions off, which is a shame, because conceptually its interesting. Unfortunately I think as things are, its not really possible to fully implement because Civilization is a 4X game at its core with Grand Strategy elements and not a Grand Strategy with 4X emphasis. (Pride of Nations and Distant Worlds have my favourite way of representing State and Private capital in an economy.)

Buildings take too long to build really. I kind of wish there was more auto created buildings at the expense of building 'infrastructure' like buildings that take a long term investment. Limit the speed of production to city size. (Maybe introduce Private capital as a concept, taxes are based off it... )
That's probably not the vision for C2C, just more of a personal wish. (I still rather enjoy this game - mod)

P.S
Maybe just cheapen economic buildings, but keep more serious infrastructure expensive.

Civilization is a 4X game at its core with Grand Strategy elements and not a Grand Strategy with 4X emphasis.

Click to expand...

How would you describe these concepts, these game 'types', individually?

Thunderbrd said:

How would you describe these concepts, these game 'types', individually?

Click to expand...

Well, I think it differs as a whole from people to people. Feel like an essay could be written about the finer points.

However, I think what it comes down to is the end game.

4X games - Is when you get down to it, Imperialism simulator XXXX . (Civ4 and by extension, C2C.
eXplore, eXpand, eXploit, eXterminate.
So all features are typically designed to be around these concepts. We'll use Civ4 as an example. I do know that there are victory types that don't solely revolve around conflict.
Scientific, cultural, diplomatic, space race. (Although you could argue, outside of diplomatic, warfare can play a key role in keeping an enemy from achieving any of the above.)
Which is why I say that Civ4 is a 4X game with elements of Grand Strategy. Diplomacy and espionage for example are just additional means to stave off potential unwanted conflicts, or weaken an enemy for easier conquest.
Conflicts are typically only external, there are no internal discourse of any real consequence other than maybe a malus to the continued efforts of expanding the war machine. [Aka. The steamroll effect.]

Grand Strategy. Basically a game that tries mimic the real world but in an abstract way to be fairly palatable.
Tends to play well with 'what if' scenarios and sandbox games.
War is just another means to aid in victory. (Pride of Nations - the goal is to win prestige wise. This can be achieved by a stellar economy, expert control of the colonies, war, diplomacy, and dominance of world trade). Victoria series by Paradox Interactive is similar in the regards that the key metrics are prestige, industry and military might.
Your own Empire can be your own downfall. Civil Wars are an actual serious threat to your imperial conquest, sometimes even more devastating than the actual wars you've fought.
The economy is much more interconnected. So devastating an enemy might be detrimental to your economy in the long run. (Distant Worlds - destruction of the private sector is crippling on state revenue. PoN - The world market)

So... if I had to summarize.
4X games focus on conflict, and any other means outside of direct conflict only exists to perpetuate the agenda of further conflict. (In times of peace - prepare for war)

Grand Strategy. War is just another means. Sometimes the risk of waging war is just as devastating as the gains of it. (Someone's enemy today - is an ally tomorrow.)
Pride of Nations - Building up your army takes HUGE investments and losing it will set you back severely.
Victoria 2 - You're losing actual population. Your manpower as soldiers could easily have been working the field or working the factories, partaking in everyday life.

So that's my take.
There are probably differences in opinion, and maybe I am outright wrong in how I describe either. However, this is generally how I feel either types actually work when you get right down to it.

Conflicts are typically only external, there are no internal discourse of any real consequence other than maybe a malus to the continued efforts of expanding the war machine. [Aka. The steamroll effect.]

Click to expand...

This is why I like the revolution mod so much. I realize it's kind of broken, and the AI has a hard time with it, but it is a super amazing feature which gives the player some internal problems that they have to do it. It would be super cool to see the revolution mod fixed/made in a simpler form for the AI since it is such an awesome feature and a shame to not be able to play with.

That's a lot of good food for thought. I'm sure we would like to have C2C be a balance of the two really.

@Giuseppe: I do think the rev mod is an inspiration. Eventually I'll get into some similar functionality and may blend that project in with an effort to dramatically improve Rev itself as it is manifest in C2C.

@Giuseppe: I do think the rev mod is an inspiration. Eventually I'll get into some similar functionality and may blend that project in with an effort to dramatically improve Rev itself as it is manifest in C2C.

Click to expand...

That sounds pretty awesome! I cant wait until you get around to it! (Even though every one of your billion projects is super amazing)