Newb with some strategy issues...

[keith.lamothe]

Yeah, and I really need to clear all my logs too, because they are getting pretty messy.

There's a handy-dandy list for your reference in the patch notes for next release!

... which I wanted to get out tonight, incidentally.  Just finished a marathon of Valley2 stuff, and another one's coming tomorrow.  Hmm, this release-tonight thing may or may not happen

[mmobley]

Neutering doesn't include destroying the Warp Gate. That's Gate Raiding.

What's Gate Raiding?

[keith.lamothe]

Neutering doesn't include destroying the Warp Gate. That's Gate Raiding.

What's Gate Raiding?

Destroying the Warp Gate



(so you don't get waves from that planet; the intermediate tutorial covers this though I'm not sure if it calls it "Gate Raiding" specifically)

[Hearteater]

You Gate Raid a planet by killing everything that is a Warp Gate.  Typically this is just the structure called a "Warp Gate" but some others do this too, most notably Eyes.  Killing a Warp Gate costs 5 AIP.  In exchange for that, if you've killed all Warp Gates adjacent to one of your planets, the AI can't send waves against it!  Your goal is to Gate Raid until the AI has only one choice, and that's the planet you put most of your defenses on.  You may hear the team "Whipping Boy" or "Choke Point".  Both are names for the system that has been set up to receive all the AI waves.

However, should you accidentally kill all Warp Gates next to your planets, the AI becomes free to pick nearby Warp Gate, which suddenly makes you much more vulnerable.  To make sure you aren't missing anything, in the Galaxy Map, press W and you will see how many Warp Gates are pointing to each system you control.  Red numbers on more than one planet are generally bad .

[zoutzakje]

I've always seen neutering as killing guard posts and warp gates, because it's not something I make a difference about. If I attack to neuter, warp gates and such will die too. If I attack to kill warp gates, guard posts will die too. If I have to leave a warp gate alone for wave control, I still kill warp gate guardians and eyes. I've noticed that leaving any of that stuff alone can only cause problems later in your game.

to mmobley: You don't have to worry about Wormhole Guard Posts just yet. "Do I destroy them or not?" is a question you will only have to ask yourself on the higher difficulties (8-9+), if you want to create a highway through AI systems.
For example, you got a bunch of planets clumped together and for some reason you decide you want a target 3 jumps away, leaving the AI systems in between alive (though neutered). You send your fleet there with a mobile builder, slowly neutering stuff along the way and you set up camp at your target. You start expanding from there with your fleet and after an hour a wave warning shows up. A wave warning for a planet all the way back, meaning your fleet has to go back through those AI worlds (if a mobile fleet is your main defense against waves). But those AI worlds have been on alert and have gained a whole bunch of reinforcements at the wormhole guard posts. Your ships already move slow through AI territory and with ships attacking you at the wormholes it will take even slower, thus possibly being too late for the wave. Or your fleet could actually get wiped out on a high dif mk IV world. That's when you should have destroyed the wormhole guard posts on the route you want to take. Just put your ships in transports after and watch them fly by without being noticed by other AI ships.
Usually only reinforcements of high difs get challenging enough for that tactic to be useful. On lower difs you can just fly through with your whole fleet and still be in time for the wave.
I remember destroying nearly all wromhole guard posts at surrounding planets once on a 9/9 game. Simply because I prefered to see enemy ships only at the command station. Made it easier for me to kill them if I had to.

[mmobley]

Thanks for all the advice. I ended up giving up on that particular game (decided I had made too much of a mess) and tried again on a slightly lesser difficulty. That was much more successful, but was too easy (4/5, same parameters). I beat both homeworlds by creating a moderate homefront of about 12 planets set up on 3 chokepoints, then using my fleet to run through and neuter about 20 other planets, then deep struck the harder homeworld, followed by the easier one. It wasn't too hard, and I found that speed does matter to a certain degree. Nevertheless I won my first game .

But, being a computer programmer who loves to figure out how game AI algorithms work I needed to up my difficulty. Soooooo....

Current game is a 6/5 with same params (50 planets, no auto AP, no expansions, lvl 4 human resistance, others default). I picked a crazy map setting, and chose Shield Bearer for my special type. Everything was going good, until I was making the move to strike the first homeworld...... Once again the AI has gotten carrier happy and tho far less than my 7/7 game (200 instead in most carriers) they are still annoying...... And here are some other thoughts. Feedback is appreciated.

1. My AIP is currently at 560 something. Yes yes I know that is high, but the map (and the game rolls - see below) kinda prompted my taking of 25 planets to defend at chokepoints. The only two mark 4 factories were BOTH relatively close to the AI homeworlds, and far from any viably defensible position I could have without doing what I did. I did take out most of the data centers to offset this a bit, but keeping it lower wasn't working for me. I know that AI tech thresholds and ship types are influenced by AIP, but what else is, and at what points?

2. Gate raiding isn't working really well for me. I am finding that the border aggression is still becoming a huge problem even without the warp gates (carriers now right through neighboring wormholes). However, my research into the game does indicate a potential strategy... How about taking out ALL of the warp gates except one (as opposed to just the ones on bordering wormholes)? My analysis indicates that these gates actually warp IN reinforcements from the other galaxy, not out into my planets. So, by that logic cutting off all warp gates except one (on a mark 1 planet) should bottleneck most AI reinforcements (and limit them to mark 1 ships) to that one particular region..... right?

3. I need some advice on proper turret defense. I did do the HBC/Grav turret setups, and they are quite effective in many cases. However, if the AI sends forces in excess of 500+ ships they fail as well simply due to the fact that the grav turrets get destroyed before the HBCs finish killing them. To make matters worse it seems that most of the other turret types are rather useless except as cannon fodder for the AI ships. Even sniper turrets don't hold up well against a force of 1000 ships. Any ideas where I am going wrong here?

4. Here's another thought... so, the way the game works it seems is that every AI planet has ships based on its mark (mark 1 has mark 1 ships, mark 2, mark 3, etc...). I have noticed that mark 4 planets also send out mark 5 ships (which are really really really bad news...). I have read on these forums that planet neutering is effective on border worlds, and it is.... to a point. But those mark 4 planets are downright annoying ESPECIALLY the core world ones that spawn a seemingly endless supply of mark 5 ships while alerted and/or fighting. I started a strategy of targeting mark 4 and the core worlds for full neutering, and it does appear to be effective (the one coreworld has been almost completely inert with about 460 something ships in it, while the bordering mark 1 planet continues to mass produce carriers full of mark 2 ships, easy targets ). I am having trouble neutering the other 2 coreworlds near this homeworld, but I think that may be due to the AIP being high, and think that if I had gotten to them sooner then I may have succeeded. Is this a good strategy just prior to hitting the homeworld? And no, I have not alerted any of the coreworlds or homeworlds permanently (made that mistake one - never again ).

My strategy is based off of the following facts that I believe to be true (unless you experts tell me otherwise ):
1. Guard posts determine how many reinforcements an AI planet gets, and destroying all of these posts effectively limits their forces to a number around 500 in said planet and effectively neutralizes said planets ability to MOBILIZE (not prevent passing through forces tho) any attack/carrier force.
2. Destroying guard posts also curtails the AI's ability to keep their number of ships in this galaxy ACTIVE, which prompts carriers for rapid deployment of massive forces without having to activate the smaller ship until they are deployed. This seems to be the counter for the "neuter everything you can" tactic, with the following point also making it worse...
3. Destroying guard posts and warp gates does NOT limit the reinforcements coming from the other galaxy into this one at all. In fact it just makes the reinforcements come from the untouched planets (like the coreworlds and homeworlds, which only get mark 5 ships....). This with point two makes the "neuter all" tactic a rather bad idea .
4. Taking into account the 3 above mentioned facts going after the mark 4 planets for neutering (as well as border worlds) would effectively limit the number of more powerful ships while still allowing the AI to seem happy with reinforcing with less powerful ones (mark 1 and 2 ships are so squishy LOL). This should in theory make any game much easier especially when preparing for the end game run on the homeworlds, with their pesky coreworlds effectively out of commission, the AIP rather low (having neutered but not wiped out the command posts), and only briefly alerting said planets when attacking to neuter and to finish off the homeworld they protect.

Am I thinking on the right track here? Anyone have any feedback to offer? Thanks!!!

Saved game attached.

[zoutzakje]

First thing to keep in mind is that not all Carrier forces will attack you. I suspect a lot of it is from Special Forces, which move around the galaxy to protect planets the AI considers important.

AIP influences wave size, wave tech, enemy world reinforcement size or speed, or maybe even both. I think that covers the most important things of what AIP does. To keep AIP low you could only take the most valuable targets (ARS, Factory IV, expansions bring more goodies on worlds too), neutering worlds along the way. Just make sure you don't attack anything further away than 3 or 4 hops from one of your planets or you will cause deepstrike threat. Having a fragmented empire can be hard to defend, but the low AIP can work in your advantage there. Defending with turrets will be easy, because you'll have plenty, but defending with a mobile fleet will be difficult. Warp Jammers would make this scenario a lot easier.
Or you could do what I always do, and take everything along the way to a nice a target, placing logistics commands everywhere and relying heavily on my fleet to both defend and attack. Economy won't be much of a problem, but AIP will get high enough to send powerful waves to you and worlds reinforce quickly, forcing you to keep enemy numbers on bordering worlds low many times. Still, destroying all but 1 bordering warp gates and massively defending the one left alive proves to be working quite good for me at least. Regular waves usually won't be much of an issue (unless you're up against teleporting units, but I don't know if the base game has any). If your grav turrets there die to fast for your HBC's to handle the AI ships, simply build more. Build more and more until your turretball can handle the waves.

Your idea of gate raiding all gates in a galaxy except one won't work. Reinforcements still happen regardless of warp gates, and wave compositions are influenced by AIP, not warp gates. Doing this will just give you more AIP from destroying all those warp gates and it'll bring you little advantage. It does help slowing down ExoGalactic Strikeforce waves (since those are the only waves that actually spawn from the warp gates), but those are expansion stuff.

I'm still sticking to my idea of grav turrets and HBC's for defense. Put tractor turrets around the wormhole too. And if your defense can't hold, just build more and more. If you use the leave one bordering warp gate alive tactic, you're going to have to keep upgrading your turretball there over time. Put a bunch of Lightning Turrets there too. They don't do as much as the HBC's, but they'll soften the enemy ships up for sure.

Here's the thing about coreworlds... neutering them is pretty much pointless, because they reinforce fast and with mk V ships too. What I usually do is take a planet that is 2 hops away from a coreworld (so no alert) and wait till I'm ready to attack. Once I am, I destroy the planet in between, then take on the Coreworld and destroy it completely (maybe place a warp jammer) and then move on to the homeworld. I don't even bother with any of the other coreworlds, I just attack one of them and then move to the homeworld. Of course this could alert other coreworlds if you don't use a warp jammer, but at this stage it shouldn't matter. You're at the endgame and it's an all or nothing battle that usually doesn't last long enough for the other coreworlds to reinforce badly. I never even stick my head into a coreworld if I'm not planning to beat the game very soon.
So basically, when you are ready, destroy one coreworld, ignore the others, then keep sending attack after attack to the homeworld until it is down, leave everything there alone and move on to the next homeworld. Make sure you have already captured a planet 3 hops away from the second homeworld before attacking the first, so you can quickly move to the next once you're done with the first HW.
Also, always start with the toughest homeworld. That way the easiest one can be finished quickly as second. If you do it the other way around, AIP might be so high that it's impossible to take out the tough homeworld.


About your strategy.
Destroying guard posts reduces the total number of reinforcements allowed for that planet, but only to a certain degree. I believe I heard Keith once saying that it's no longer effective to destroy guard posts once the guard post count for that planet is below 10 (wormhole guardposts included). After that it simply means the command station gets more reinforcements than it would otherwise. This means most of it's force will be concentrated at the command station, but given long enough time, it can still become overcrowded and release ships as threat.
The only reason Carriers are created is when there are too many ships on one planet. This is to prevent massive lag people could get otherwise. Keep numbers low on enemy planets and you won't see many Carriers.
Coreworlds and homeworlds already do get reinforcements, whether you do try to do something about it or not. When the AI picks it's reinforcements locations, homeworlds, coreworlds and worlds bordering yours are always their priorities. Whether destroying guard posts increases reinforcements elsewhere I do not know. But even if it does, neutering is still often your key to survival. How you deal with the core/homeworlds is a concern for later. Having at least 4 of the 5 ARS before your assault on the homeworlds is generally a good idea. And if you manage to get mk IV of at least a few ships, then you're really ready to go.

Hope any of this makes any sense to you. I haven't looked at your save yet, because I'm hoping this will help ^^ If it doesn't, I'll take a look at it.

[Hearteater]

First, there are Warp Gates and Reinforcement Gates.  Reinforcement Gates allow reinforcements to appear.  The AI Command Stations are all Reinforcement Gates.  So to stop reinforcements from appearing, you'd need to kill all the AI Command Stations.

Being attacked by threat ships (ships you've woken up) is mitigated by killing those ships before there are too many of them.  I like to think of it as cleaning up after myself.  I like to see the number 0 under Threat in the top bar.  I can be 20 hours into a 8/8 and have threat still be 0.  It creeps up from time to time, but using the galaxy map you can switch to the threat filter to see where they are all located and go hunting (provided you have scouts on those planets, the threat filter doesn't show threat ships on planets you don't have scouts on).

Different from Threat is Border Aggression.  This is when a system gets so full of ships some of them just spill into an impromptu attack.  You can mitigate this by occasionally clearing out all the AI ships in neighboring systems.  Note that the cap is partly determined by the number of Guard Posts alive, so if you kill all the Guard Posts you actually increase the frequency of Border Aggression.  I generally kill problem Guard Posts and leave the rest alone if I'm not planning on regularly moving through the system.  Then I clear it out every 2-4 hours depending.  I find it particularly amusing to clear them out with Leech Starships and Shield Bearers as you get to collect a ton of ships and get them out of there safely.  Mark IV planets are great to farm up high Mark ships.

[mmobley]

Thanks for the feedback. It always helps to learn from the veterans, as this game is as complicated as it is fun (no offense to the developers whatsoever, fun is usually associated with complexity in strategy games ) and every piece of information helps me become a better player to this awesome strategy game.

To both of you - I think my tactics ideas were slightly misunderstood. I am not intending to STOP reinforcements from coming altogether (this is rather impossible no matter how you go about it, as by the time you wreck half the AI the other half has exponentially grown in size to compensate). My intent is to control/slow/redistribute the flow of reinforcements as best as I can, so that I can align my defenses better to face the consequences of smashing the AI where it matters.

Think of the concept of trying to protect a town from a raging river. Obviously you can't stop the flow of water, but you can divert or dam it up. Both are effective ways to save the town, and both are far more possible than trying to stop the flow altogether. I am trying to accomplish this in the game as well, as many times I am finding that hopping into systems to take vital planets is often problematic, and expanding to said systems also has its downfalls. I am simply trying to come up with more creative and alternative strategies to mitigate these consistent issues during gameplay.

Having said that, both of you gave me some valuable input! Hearteater - I had no idea precisely how the threat counter worked, and you cleared that up quite well! That really helps in determining where I need to clean out forces before it is too late. The warps were never a serious problem for me - quite the opposite actually. The problem was border aggression. I was getting hit on my border worlds without warning, which is quite annoying. Staying ahead of that has been a serious challenge.

zoutzakje - You gave me quite a bit of feedback on how the AI reacts to certain things. Again, I am trying to slow the AI down with little AIP gain as possible, hence the idea of gate raiding as opposed to killing a bunch of command posts to bolster a decent field of defense. I have major issues trying to maneuver my fleet from planet to planet to defend on the fly, so I rely heavily on turrets. Unfortunately they fail me a lot, so I need to come up with better turret defense strategies.

IF you can take a look at that game please do. IF you don't think it is winnable I can send an earlier save for strategy suggestions (I save incrementally for now so that I can try different tactics without having to start from scratch). As always I welcome constructive suggestions .

[Hearteater]

On Reinforcements: The game randomly selects AI worlds to reinforce from all the currently alerted AI worlds (and sometimes homeworlds, even if they aren't alerted).  If a core world or home world are alerted, they will always be selected for reinforcements.  So you don't have a ton of control on where reinforcements happen.

However, the AI only gets a limited number of worlds "picks" for reinforcements.  It gets roughly one pick per world it doesn't control on the map, with a cap of around 12-15ish worlds (it depends on difficulty).  So if the AI has more worlds on alert than it has reinforcement "picks" then not all alerted worlds get reinforced every cycle.  So, one strategy you can use is to go put some AI worlds in a corner of the galaxy you don't care about on alert.  Then those planets will random get ships added that you'll never need to deal with.  There are many ways to put such a cluster on alert, the easiest being to kill an AI Command Station since a neutral world automatically puts all adjacent AI planets on alert.  But you can be sneaker about it using certain cloaked ships to do the same with costing any AIP.

[Lancefighter]

At this point, you are really looking to end the game. You have two more systems that dont have collected knowledge that you should collect first. Spend that into turrets, and build more of them. Everyone loves turrets. (laser and mrls probably).

Anyway, take your entire fleet and go blow up a homeworld. Then go blow up the other homeworld. Make use of liberal amounts of missiles if they help, (not sure what mk5 ship's stance on emps is nowadays) or something else. Overall, you want to be ending the game as soon as possible.

You very likely wont be able to stand against a 500 aip enemy - thats more or less by design. Its supposed to become more difficult to play against as you take more land, until it eventually does become unwinnable.

If you cant win the game very quickly, it is entirely likely you will lose.

[Kahuna]

It's not winnable anymore. I had the same problem when I first started playing AI War. I captured too many planets and the AIP skyrocketed. Then the game was in a stalemate and I had to start a new game. Make it your goal to keep the AI Floor below 250.

EDIT: Oh! I could take a look at an (auto)save with 10-300 AIP if you want to.

[mmobley]

It's not winnable anymore. I had the same problem when I first started playing AI War. I captured too many planets and the AIP skyrocketed. Then the game was in a stalemate and I had to start a new game. Make it your goal to keep the AI Floor below 250.

EDIT: Oh! I could take a look at an (auto)save with 10-300 AIP if you want to.

I am liable to agree with you on this one. It would appear that I pissed off the AI way too much before I was in a position to do something about it . Eh... computers don't like me anyways cause I can make them stand up and beg .

But anyways, I had incremented my progress in saves, and found a decent save from way back that may be better to handle. Attached. See what you think. May be able to go back even more if need be...

[Lancefighter]

I would not recommend doing that.

I would recommend a fresh new game.

[TechSY730]

This does bring up two good points that I think are still major issues with the game:

1. It is too easy to get yourself into a stalemate situation (in part because it is not clear how stalemate situation arise)
2. It is not easy to tell if you are in a stalemate situation (of course, you can't tell for sure, especially on edge cases, but there are some cases where the game could do a much better job of "advertising" that you are in a probably unwinnable state)
3. That stalemates are easier to get into than losing situations on normal difficulties in normal play. (If you "piss off" the AI so much you can't really crack its defenses, it should also be enough such that the AI will start cracking your defenses, instead of both being unable to pass each other)


However, I don't know how to resolve this without introducing some sort of sharp "cliff" (Like, AIP 290 will be a tough but winnable fight, AIP 300 the AI will "ROFL-curbstomp"), which just leads to "stay under that hard line" styles of play which are less fun.

Well, we sort of have that now, but the "line" is rather "blurred". Different playstyles can handle different amounts of AIP before becoming unsustanable. (Often times, the playstyles that can handle more AIP also take much, MUCH more time, and are more likely to fall into a "transparent" stalemate situation)
I would NOT like to see that "line" become "sharp", and keep the variety of playstyles involving AIP management viable.

At the same time, I don't want the AI to build up into a "boring" state of defensive enough to hold off human attacks but not offensive enough to take on the human.

At the same time, I don't want those difficulty "cliffs" described above.

And at the same time, I want the AI to behave as "lore" states and not simply rolling over the humans at the start of the game (or at low AIP) with everything they have because of what AIP stands for (how much of a threat humans pose), in particular, because the AI is managing other things that require attention (the lore reason why the AI does not simply throw everything at you from the start).

Ugh...complicated.
And beyond me to figure out how to tweak "pacing" to meet what I consider issues while not compromising what I listed below (which I consider some of the "selling points" of AI Wars' pacing system)
Maybe this "ideal" system is impossible, and any system would have to sacrifice something, whether it being "stalemate avoidance/opaqueness" or one of the other things.


...Sorry for the long, rambling post. 
Let me know if you need any clarifications to my clunkily presented ideas.