Learning the Game

[Eternaly_Lost]

One thing you might also want to try, but it is in a way a very different game, is the Minor faction, Fallen Spire. Unlike in the traditional game, that one if you come to the end you end up basically in a full out war where it not only reasonable, but often required that you take every planet in the galaxy.

I had played AI wars before it, not really liking it that much, but at the Christmas sale, I decided to give it another try, found Fallen Spire and fell in love with how it plays. It plays a lot more like a traditional 4x game where you take the Galaxy, rather then AI war's design of only having to take what you absolutely need, but to me that is just another positive point in AI wars favor. Want to take a very few planets and do strikes against an overwhelming foe? You can do that. You want to take the Galaxy an field a fleet that would give even the Cookie Monster pause (if they only could kill it...) you can do that. Decide mid game you want to go from few planets to conquer the Galaxy? You can do that too. In theory you could go from conquer the Galaxy back to small raids, but after you gone far enough into fallen Spire, it almost impossible to do so.

[Wanderer]

Wow, thanks Wanderer! That's very helpful. I don't think I have any questions on that at the moment. Except for one extremely pertinent question, how did you know about the tacos?! Seriously I've been craving mexican food for like two days now, and there's a little taco place a couple of miles from where I live that I've been wanting to try. So thanks for torturing me. :-)

Glad to help (munches down on a Carne Asada burrito from a local hole in the wall called Rita's).  *burp*

The rest comes with time.  It's an intricate game, I'm still kind of a newbie at it.  Oh, I've got some tactics that work because I drilled into the core of the game, but I lost my first 10 5/5 and 6/6 AI Diff games.  A few really badly.  You will have to experiment.  You'll savescum to try new tactics.  You'll savescum again an hour later when you realized that you started a domino effect that wrecked you.

After you've done that and restarted eight or nine times and finally get a handle on the game after investing about 50-60 hours in the game, you'll understand just how intricate it gets, and suddenly you have an epiphany... you don't have to care about a LOT of it.  The game does it for you as long as you've made the right strategical moves and responded tactically to the changing board.  But of course, if you don't understand the finer tactics which you're getting bogged down in, it's hard to make assumptions about what's the right strategical move.

My only recommendation: Screw it, try one, or ten...  There's always a savegame you can fall back on when THAT didn't work, either... No, I've never said that playing a game, really...  Heck, one of those AARs I savescummed 4 times in a row trying to find an effective strategy.  Blasted Raid Engines....

[zoutzakje]

I think the good thing about this game is that it can always surprise you, or overwhelm you if you're not careful. There are always new things to discover (even more if they're going to make another expansion for AI war again after AVWW hits 1.0). A while ago I played a 7/7 game, which normally isn't that much of a big deal to me anymore (8/8 or 9/7 is standard for me). Unfortunately one of the random AI types turned out to be a Raid Engine, a type which I had never played against yet. I got over 500 hours of experience with this game so far but still it was giving me a serious challenge. everywhere I'd go I would trigger a raid engine, sometimes even 3 or 4 at once. I sure did savescum quite a lot lol.
But i did enjoy that game a lot nonetheless (also because I managed to beat it ), and now I'm strongly considering playing a 7/7 double Raid Engine game on a crosshatch map type... that should cause some serious problems lol.
My point is that, once you have figured out a bit how this game works, and you decide that you really like it, you will most likely never get bored of it. Each campaign is different, and there is always a tougher challenge you can work on achieving.

EDIT: woohoo, 500 post count

[PokerChen]

If you want to go down the X-resources and Y-ships route of judging your progress, here's a couple of rules-of-thumbs:

- It's probably better to use the fire-power of your fleet to judge how good it is, or how much knowledge you've spent unlocking things that can attack the AI homeworld. The player's assets is based around caps of ships, which are somewhat balanced so as to give overall a similar boom for your dollar (dollar being knowledge for most things).

- In that sense, you could do with a starting figure of 70,000 knowledge with 20 planets. A question to ask is: "What would you spend that on?" Let's take a wild ball-park value: I'd suggest that ~40K worth of unlocks in fleet-ships and star-ships will give you a reasonable chance at eliminating an AI homeworld that has <250 defending ships. The mark-IV factory is absolutely essential, and a fabricator or two will help. In this scenario, I'll be taking about 5 sets of mark III+IV ships and a gaggle of starships to add firepower.

- It'll be useful to map out a list of planets to take, which ones that you'll want to keep, and how to get there. So, starting at 5/5 will give you a sense of the basic lay-out of the path ahead, without the additional complications like guardians or excessive enemy waves.

- Here's a satisfying alternative to being afraid of what the AI *might* do and minimising the AIP : As a player, I value mobility, and stand by razing AI-worlds and paving a highway to anything that I want to defend (and nothing else), rather than skulking around with Transports. This puts me at about 100,000 knowledge in total, which I cash in to get a powerful fleet that I can divide in both offense and defense. (Note that Metal and Crystal resources don't factor much into the calculation.)

- The scale of AI-retaliation is a bit of an O(N²) thing, so you have some lee-way to experiment. Do so!

[apophispro]

Eternaly_Lost: That's an interesting idea. I'll have to try that out at some point, but I do want to really get a handle on the main game first.

Wanderer: Yea, I definitely get the inclination that a lot of losing will be involved on the way to learning how to play. I play piano and read a book on technique with playing songs. One of the things that it emphasizes is that many players will make a mistake and then keep playing the song the same way. They never stop to evaluate why the mistake happened so certain mistakes in their playing become engrained habits instead of just things that could have been corrected early on. Obviously there is no mistake here in the sense of hitting the wrong note, but I'm trying to avoid the situation where I'm making mistakes, losing, and starting a new game with no particular idea of what I did wrong in the last one. For instance I might interpret that the large wave I lost against in late mid-game was the result of what I was doing at the time when I could have been doing the same thing just fine if I had done something a little different in the beginning. That sort of thing.

Zoutzakje: The surprise factor of the game is one of the things that attracted me to it. Also the lead developer talking about how he found other strategy games frustrating because it was easy to just learn tactics that worked every time. I found the same thing as well. I got to the point a few years ago where I could look at the box for a lot of strategy games and already understand how it was going to work. I hadn't really touched strategy games since until this one. Although I realized there were things I had missed out on at the time and along the way. I also want to try Sins of a Solar Empire and Anno 2070. In my larger feats of blasphemy in strategy gaming, I've never played through Starcraft or Warcraft 3. I have played a few sessions of both of them, but I would like to pick up copies of the games (especially now that they're so cheap) and play through them. The surprise factor is very attractive for me though. I love things with open-approaches and dynamic variables that don't stay the same. Even chess frustrated me because I know that most of the game is based on just a process of internalizing certain moves. The variables are too rigid for my taste in that sense. I'm really hoping that this is a game that will always remain a challenge which is why I'm putting so much effort into learning it. Based on what I've seen of the game and what people are saying about it, it seems that this could be a real masterpiece of strategy gaming and one that will genuinely stand the test of time regardless of what else happens with the genre.

Zharmad: Thanks for all of this. What you're saying makes a lot of sense. Knowledge equates more or less directly to ship caps except for the ARS which seems to be the only major way to circumvent that. Thus why knowledge gathering is so important. I'm also still a little confused on knowledge raiding. From what I gather most systems have AI command stations on them. Thus on pretty much any system that I haven't taken I would need a Mark 3 Science Lab to steal knowledge from it. I'm not sure exactly what I would need to also build in that system in order to get one, but it seems like it would be a lot. Is knowledge raiding actually a common tactic then, or do you usually have enough from the systems you're taking? The figure of 70,000 knowledge and 20 planets makes a lot more sense though. I was so focused on keeping AIP down that I wasn't really taking enough systems to build my own forces. I assumed the goal was to take as little systems as possible, but I was taking so little that I wasn't a threat to the AI at all both in AIP and literally. It's interesting about the balance between your power and the AIP. You really have to keep both in your mind at the same time. It seems that you can't make one or the other a goal because then you'll end up screwed on the other end. If you make power the goal, you'll drive the AIP way up and be demolished. If you make AIP the goal then you won't have enough power to do anything. The way you handle that balance becomes preference of player strategy.

I'm still a little shaky on prioritizing systems to take. I'm getting an idea of what the priorities are. ARS will be necessary. Certain systems that hold offensive positions will be necessary. I guess what confuses me is the balance between offense and defense. A lot of systems that get taken seem to be offensive. Let me define what I'm talking about here, and you can correct me if my definitions are wrong. It seems to me that a system taken defensively is one taken almost exclusively to protect my core system or a valuable resource. So I most likely take the surrounding systems to my home system in very beginning game as a defensive maneuver regardless of what else is on them. Then as my scouts are moving outward, I begin to evaluate the rest of the map. There's an ARS two hops from one of my outlying defensive systems. There's a high-resource system one hop from a different one. I take both of those systems for offensive reasons. Taking a system offensively would be any system that will help me build myself up in order to fight the AI but will not provide a better defensive barrier. So taking that ARS system two hops away will help me build up my fleet but will also split my resources into defending a system now surrounded by hostile systems. It seems I might want to neuter the system that was one hop away for defensive purposes in that case. There also might be instances where I might want to take another system defensively to protect a resource from a large local threat. However there seems to be a large balance between offensively taking systems, defensively taking systems, and then maintaining appropriate levels of defense for your strategy in general. I think that's why the "islands" thing confused me. I felt every system should have a defensive gain as well, but it's not going to. The "turtles" then would be the people who build up a high level of defenses while attempting to minimize their offensive maneuvers down to only the absolute necessities.

[Wanderer]

Wanderer: Yea, I definitely get the inclination that a lot of losing will be involved on the way to learning how to play. I play piano and read a book on technique with playing songs. One of the things that it emphasizes is that many players will make a mistake and then keep playing the song the same way. They never stop to evaluate why the mistake happened so certain mistakes in their playing become engrained habits instead of just things that could have been corrected early on. Obviously there is no mistake here in the sense of hitting the wrong note, but I'm trying to avoid the situation where I'm making mistakes, losing, and starting a new game with no particular idea of what I did wrong in the last one. For instance I might interpret that the large wave I lost against in late mid-game was the result of what I was doing at the time when I could have been doing the same thing just fine if I had done something a little different in the beginning. That sort of thing.

Understood, however using your analogy, this is more like trying to pick up what the violin was doing wrong in the symphony.  Eventually you'll learn to hear it.

I guess what confuses me is the balance between offense and defense. A lot of systems that get taken seem to be offensive. Let me define what I'm talking about here, and you can correct me if my definitions are wrong. It seems to me that a system taken defensively is one taken almost exclusively to protect my core system or a valuable resource.

Basically, but what's a valuable resource?  Once you've gotten the K out of a system, what reason are you holding it for?  Are there unmovable fabricators there?  ARSs once captured are just glorified research stations, the research on the new ship is yours forever. Do you need the econ?  Is it worth the defense from random drifters compared to the ease you can rebuild your econ station there with?

There's a high-resource system one hop from a different one.

I never evaluate systems on this basis, though with the new harvester changes I might start.

So taking that ARS system two hops away will help me build up my fleet but will also split my resources into defending a system now surrounded by hostile systems. It seems I might want to neuter the system that was one hop away for defensive purposes in that case.

...or neuter everything surrounding it so you can ignore the satellite system except during major events.

I think that's why the "islands" thing confused me. I felt every system should have a defensive gain as well, but it's not going to. The "turtles" then would be the people who build up a high level of defenses while attempting to minimize their offensive maneuvers down to only the absolute necessities.

It CAN, but then you're talking a completionist campaign.  Those are fun, actually, but you have to be prepared for the equivalent of World War III.  My personal 'generic' strategy is harden the core, protect more heavily important worlds with fabricators, give any other systems a reasonable chance at survival without intervention except during major events, and then kill an AI.  Some worlds are simply taken in my games as step-stones.   I don't even bother trying to get research stations/cmdstations/etc on the world, because they'll have things I don't want to capture and possibly lose (Human Prison Camps come to mind) and I just need the world to continue advancing from.

[Nodor]

There's a business term that seems to apply to success in AI War. 

Fail Faster.

The idea is that because you don't know what will cause a product to succeed or fail, the importance skill is figuring out how to develop an interative testing process until you find one that leads to success.

In other words, grab a mark 4 or so AI foe, and start.    Once you win at that level, kick it up a notch or 2 and try again. 

Eventually you will find the setting(s) that fit your ideas of fun and provide you with the challenge level you crave. 

To quote the Dev's.  A win vs. 10/10 is filed under "bug reports".

[keith.lamothe]

A win vs. 10/10 is filed under "bug reports".

If it's not cheese like two non-wave-sending AI types, etc

[Wanderer]

A win vs. 10/10 is filed under "bug reports".

If it's not cheese like two non-wave-sending AI types, etc

Baaaaah!  I resemble that remark!

[Minotaar]

Hi Apophis! Some thoughts in response to your last post:

I would definitely say that this game just gets better and better the more you play  On not learning from mistakes, if you don't start on too high of a difficulty, (I started on 6/6 after reading lots of stuff, like you did), the strategic sort of mistakes probably won't kill you. But it doesn't mean you can't learn from them. Every little hole you leave in your position tends to become a Chekhov's Gun in this game, but on a low difficulty, it's not going to kill you outright, it's just going to point out that maybe leaving this mark IV planet nearby wasn't a good idea (as it proceeds to build up to epic proportions), or maybe that going out of your way to capture that fabricator wasn't worth (as it gets wiped out by a cross-planet attack). The lessons to learn will be there, and once you've played a bit, it'll be easier to spot the difference between the things you and someone else are doing and learn from them.

Also, what zharmad said: as long as you're getting something out of the AIP you're paying, it will be fine, that is the purpose of AIP. So experiment, by all means 

On capturing systems: you can surely split those into offensive aka. getting goodies and as "stepping stones", and you may or may not want to hold these systems, as Wanderer pointed out, and defensive, which create a buffer before your homeworld or strengthen your defense in other ways, such as decreasing your surface area. Thing is, you can achieve your defensive goals via neutering to some degree, allowing you to spend more of your "AIP budget" on offense (since a neuter usually costs you 5-6 AIP at most). Of course, there are planets which can fulfill both roles, and those are the planets you'll probably want to build your strategy around.

Lastly, if you want to talk some strategy, you can find me on skype (I'll PM you the details). I'd love to help, maybe we can even fire up a game together 

[Diazo]

A win vs. 10/10 is filed under "bug reports".

If it's not cheese like two non-wave-sending AI types, etc

Baaaaah!  I resemble that remark!

And one day I am going to file that bug report.

I've got my forum title for a reason after all......


However, for the offense vs. defense topic, keep in mind the only systems you have to defend are your homeworlds.

There are actually two questions you have to ask your self about an enemy system:

1) Is this system worth taking? (The offensive question)
2) Is this system worth keeping? (The defensive question)

It does not usually happen to my until mid-late game but I will take a system and not bother to hold onto it depending on the structures present.

As always, depends on the map layout and your playstyle but your only defensive imperative is keeping your home command alive. Not to say losing colony systems is a good thing but they can be sacrificed if needed.

And you can always leave empty systems floating around if they would over extend your defense.

It boils down to the fact you only have 2 objectives: Keep your home command alive and destroy the AI's home command, everything else is a means to one of those two goals.

D.

[apophispro]

I started a new 5/5 Random Easier 60 Planet game last night and played for a couple of hours. I'm feeling much more comfortable with the game now. I feel like I've got a handle on what I'm doing and developing some basic strategies. I started with a system that had four wormholes so I scouted out at least two hops in every direction. Two of the four systems near me had high resources so I took those first to build up my resources. I actually have maxed out resources now. My ships are almost maxed out as well. There's a Mark IV system one hop from one of my outlying systems. One hop from that system (and only immediately accessible via that system) is another system with an ARS and a data center. So I took the third of the four outlying systems, and then I'm going to take the fourth to build up my knowledge so I can unlock Mark 3 of either fighters or bombers depending on what the Mark IV system has a lot of. Then I want to take that system so I can also take the system beyond it. There's one other system one hop away from the fourth system I haven't taken yet that has an ARS as well so I'm planning to take that one. I think that will be the end of the systems I take around my home system.

I already did some more scouting and have identified two Factory IV systems as well as varying shield generators (two Bs, an E, and an A) in my nearby vicinity. After I finish the early stuff I have planned, I'll take one of the farther systems to start looking for the AI. Anyway I feel like I have a pretty solid idea of what I'm doing now. Obviously there's a massive amount of the game that I don't understand or am not familiar with, but I feel like I have some of the basics down enough to be able to approach that now.

Thanks for all of your help! The statements on whether I should actually keep a system helped as well. For some reason I was taking it as a given that if I took a system I would need to keep it. I've completely re-evaluated that now and am looking for opportunities to take systems that I won't need to keep. The statements on evaluating systems, neutering them, and that really the only defensive requirement is keeping home alive have been very helpful as well. Basically everything you guys have said has been helpful, and I plan on revisiting this thread a few more times as I keep playing beginner games both to review what has already been said and add to it when new questions arise. Thanks again!

[Nodor]

As a note, some targets, like data centers, are good things to kill.

You do not have to take the system to kill them.   

[apophispro]

So I needed more knowledge and didn't want to take an additional planet. I picked a relatively weak planet one hop away from one of mine, and I built a mobile builder and small fleet to go over and establish a temporary beachhead with a science station. As soon as I did this the AI on the planet started mass-producing ships. I managed to hold them off long enough to get the knowledge from the planet, but I have no idea why this happened. The wiki article on establishing beachheads doesn't say that if you do it, the AI will start throwing a never-ending line of hundreds of ships at you. How would that make beachheads a good strategy in taking the planet? It would seem to make it far more difficult. Did I just do something in particular that triggered that? Is this what it means in the knowledge raiding article when it says building a mark 3 science station enrages the local AI? That could have been a little more specific.  My threat is 225 all from one planet now.

[PokerChen]

Science-IIIs are what triggers them. In the alerts on the top-left hand corner you should have been a line that says the AI is spawning ships in response to your hacking.