Discussion in 'Civ6 - General Discussions' started by Steve Eric Jordan, May 18, 2019.
Well, the topic has been up a few times. And whilst I agree on the AI I still think that modding at a more fundamental level is the only working path forward.
Given a patch or two per year the AI would always be lagging. The pace of improvement is too slow. And the community creativity and skill is somewhat impressive compared to FXS.
The AI is never going to be up to what consumers consider standards. I could write an essay on this, but that's not the point. It's an image thing.
Don't get me wrong, I'm saying the AI should always be improved.
But you're vastly overselling the fact that people play it in spite of its AI, which either means it's not a critical blocker from "greatness" or it's not a blocker at all. Kinda impossible to tell. Certainly, here on CivFanatics, we have a habit of overrating the issues we suffer from more. AI is always one of these. "We want challenges higher than Deity" (fair, not sure how possible it is). "Why don't we move the default difficulty higher / why don't we delete lower difficulties". I've seen these suggestions in the past. I'm not saying they're common, or even that they're popular (honestly, memory-wise I couldn't, hah).
At the very least, after an expansion or three, this is the best place to improve the AI. Each expansion shakes things up, but the base gets more stable. So I share your hope that it continues to be improved. I don't agree with comparisons to games as simplistic as Advance Wars, though
Agreed, once you know how to play the game you're essentially just waiting to win, nothing is going to happen in the current game that will change the outcome.
It's the same thing over and over, deity difficulty, AI starts off way ahead, AI stays ahead in science for most of the game, I eventually fly past them & I just wait to win.
I can only talk with 100% certainty about myself. And I stopped playing 6 a long time ago, and went back to 5 + Vox Populi. Main reason: the "AI". In 6 it's basically non-existent (you know what I mean), while in Vox it packs a punch as of now, a very fun-masochistic punch. From that, I can extrapolate: 1) that it can be done, if it was done by two passionate fans in their spare time, one of them a history teacher (not a programmer); and 2) that it is very hard for me to play 6 under the almost-certainty now that they will never make their "AI" better.
Again, I am talking with 100% confidence only about myself, but those are my conclusions.
> It's impossible to make good AI for civ games
> modders keep doing it
I mean, there are many reasons why this happens, but....
To summarize: good AI development time/cost vs modders free time.
To counter-argue: modders do it for free, and many of them are not anything close to programmers, or if they are, they are self-made (@Gazebo ), and they don't program for a living... developers do. Or at least some, in any case.
So many AI threads.
Sigh. Again, for the viewers at home, it isn't just the AI that's the problem. See my post about Difficulty in my signature.
Just to give one example: the AI builds a lot of Spearmen and Seige. Is the problem actually that the AI is stupid and so builds these useless units... Or is the problem that these units are useless, and so the AI is sensibly designed to play the game as it's intended to be played but the actual implementation of that intention is undermining that strategy?
Also, maybe the reason Modders are better at tweaking the AI is because they have more time to iterate and actually get to play the game more and so understand the current meta better.
The AI is actually useless at using siege, most of the time. Catas will just sit there doing nothing until they are killed.
I think if they wanted to improve the game some more, they could nerf coastal civs, maybe reduce the utility of navies, and make global warming happen earlier and more gamey.
Besides the obvious AI thing which will never really be up to snuff... the biggest way they can make Civ VI great is probably actually bringing the expansions to the Switch version.
In second place, make that "Civ 5 graphics in Civ 6" mod the default.
Well, to be clear, I do think the AI is mostly rubbish. The computer is not good with units.
I'm just saying the AI is not an isolated (or the only) problem. The game has real balance problems that need to be fixed in parallel.
To give another example, at higher levels the AI probably needs discounts to resource requirements if it's ever going to be competitive. It also probably needs extra free units at various points in the game (based on difficulty) - eg granting the AI free copies of their unique units. eg Norway would be much better and more distinctive if the AI's hands if it had free Longships and free Beserkers.
So far, the only adjustment FXS has made to how difficulty levels work since release is to give City States walls for Immortal and Deity. I mean, come on! All these new systems - loyalty, ages, resources, power ... and no tweak to the AI's bonuses to help cope with these systems beyond some extra starting units and some +% yield boosts?
Like I said; we're CivFanatics (even me, however laid back I can be at times).
There's no way of knowing for sure how many are dissatisfied particularly with the scaling of the AI, right? I don't know that, I can't pretend to know that. It could be tons. All I was saying is based on how popular the game is regardless, that it's unlikely that the AI in particular is holding it back from greatness. Like all great games (and I think this version of Civ is), I think there are various things Firaxis should continue to focus on (including AI).
Though a note on the modding. A lot of modders often drift into programming. I did Don't assume that modders can't also be professional developers. Or that it's uncommon. Modders do do it for free. But speaking as a modder, we have time that professional developers can't even dream of. I mean, I have less time now I have a child, hah. But I could spend six hours on a thing, and only that thing. I figured out a way around the texture compression in Beyond Earth because without Photoshop I had no easy way of getting the specular layer right. Took me a couple of evenings! I can't do that at work. I have six things to do on average, per day, at work, and eight or so hours to do them in. I don't get to spend four hours investigating a workaround without sacrificing something else on my stack. Modding's completely different in that regard
I don't know what to say. I shouldn't have to search for posts when you can do the same thing, if you don't agree with me so strongly (and even then, that wasn't really the point of what I was getting at).
Yeah, they need to balance the units. The anti-cav line should be much stronger, and only get hard countered by flanking and siege.
Giving the AI free unique units would just make the early warmongers even more powerful. An increasing number of free units granted to them when the world hits a new age would be better, especially if it included builders. The AI seems to suck at improving and repairing its territory mid-game onwards.
I like the Era idea. And I agree you can't just give Gilgamesh more Donkey Chariots. It would require a bit more tweaking than that.
That said. Starting next to Monty is "fun" - in it's own way - precisely because he does have unique units from the start. So maybe GilgaBro having Donkies at the start is not actually a bad thing...
I had a game recently where Ghengis declared early war on me and came at me with ... Slingers and Spearmen. Bit of a damp squib. I mean, Ghengis. Dude. Horses. Where are the horses? You're Mongols. Horses dude.
Anyway. AI need to field more of their unique units. That seems an easier thing to fix than getting the tactical AI to think like Skynet.
I don't think if someone programs as a profession correlates in to them being a great programmer, it just means they have some skill as a programmer. Just in general - I'm not implying anything about Firaxis with that statement,
I would like to think that if the AI was perfect I'm smart enough to find something else to complain about.
You may find if the AI put up more of a fight there would be less people playing it. Do not assume your agenda is everyone’s.
I disagree, many people say there is a lot wrong with the game excluding the AI.
The cost of the art team alone is probably enough to sink a better AI into place but they need the art team and the music and all the other things or the people that like playing for the scouting side, the building side, and the development side would not be happy. Civ VI is so much more than the wargame where people want deity to be so hard they can barely win it. Try multiplayer.
For months in college a favorite pastime of mine was playing the board game Go. My usual opponent was an order of magnitude better player than I, so in all the games we played, I never won one.
Enjoying that is not usual, but then, I went on to become an over-educated First Sergeant in US Army Combat Arms, so by definition I am not normal (or, probably, entirely Sane) . . .
A Good AI would beat the pants off most humans. The problem is, the game does not have a minimally competent AI, which is not the same as a Good AI at all. And from a marketing standpoint, trying to sell a game that will for many gamers be a series of lessons in humiliation is a Really Bad Idea.
Lord knows, I'm one of them . But if I had to boil down all of Civ VI's problems into a single sentence, it would be that the Design Team is much better at meeting Marketing Goals than designing a historically-based game. They have a game that sells, in a 'franchise' that is one of the most successful of all time. This particular iteration of the game sucks on many levels, but it does not matter as long as it looks good, Sells, sounds good, Sells, gives the impression of being a historical game to people who don't know history (i.e., the population of the USA), and Sells.
Since everyone's idea of a good game to play is subjective, whereas a Game That Sells is self-defining, we are going to continue to get Games That Sell that are not particularly what we like to play in all, or most, aspects.
The Mod Community is, therefore, our only real source for the Game We Want to Pay, and all or many of us will have to become Modders to realize that. This does not make me happy, because I'm having enough problems adding Cursive Cyrillic to my repertoire without having to start in on XML, LUA, PYTHON, C++ and all the other alphabetically-challenged computer languages . . .
Separate names with a comma.