After a pass through all maps in set 2 (bonus pack 1), I'm going back through to try to beat the ones I had skipped, always on the hardest difficulty setting.
Map "Ring" was a PITA but I was finally able to beat it, so I thought I'd relay my experience here in case it helps others. First let me say that I must have restarted this map about 15 times before finally finding a stable foothold, and alas, I can't tell you precisely what steps ended up working. I can say that beating this map felt like cheating. I'm 95% sure the AI plays deterministically, though I've never proven that conclusively. But it is also somewhat chaotic, in the mathematical sense. That is, the AI is deterministic but is extremely sensitive to environmental conditions, such that a minor change in my gameplay could have drastic impacts on the board as a whole.
I tried to recall exactly what I did with my 3 starting towns, but I've already forgotten those details. Suffice it to say that I had to try many different possible starting choices, and they had radically different impacts on the map, before finally finding something that worked. The town hall at 3 o'clock is toast no matter what you do, and it's not clear if it matters if you (a) hold off to try to save up for a castle or (b) take the single hex you're able to take and be stuck with a L1 grunt on a 3-hex territory. Normally I don't like (b) as your taxes grown more slowly and you're more vulnerable for longer, but in this case, I think I did (b) so that the local AI players would have to take longer to kill me and deal with a palm tree after I was dead. The town hall at noon lasts longer but is ultimately doomed as well. I *think* my best first move here was to take the dark green hex NE of my noon town hall. And while there are many possible ways to lose the 9 o'clock town hall too, there are a few ways to get it to survive and it was this town hall that ultimately ended up winning. The rest of my notes are about the 9 o'clock town hall.
In some runs, I was able to hold the 9 o'clock base for a long time with a spearman, but was never able to grow. I tried (a) get a castle before a spearman, (b) get a spearman then save up for a castle, and (c) get a spearman then a grunt sidekick with no castle. I *think* it was (b) that ended up working for me. Specifically, I placed a castle near the northern part of my territory, in the middle of the ring such that there was no way for an enemy to bypass the castle *and* I made sure that my castle had a buffer of my territory around it so that there was no way for a single enemy L3 troop to destroy the castle in one turn. In my experience, the AI on hardest difficulty seems to favor killing as much stuff as possible with every troop placement, so having a castle inset from your border means he would have to "waste" a L3 troop on taking an empty hex adjacent to your castle before he could then destroy your castle, and the enemy AI would prefer to kill something with his L3 troop instead. In any case, I have a fully-blocking castle with buffer zone on the north, and I start pressing south.
I was in this situation more than once and it didn't always work. Specifically, when the enemy AI in the north (diamond part of the ring) takes the entire diamond quickly, it proves disastrous. Something about what I did with my 2 other start bases before they were destroyed matters, but I couldn't tell you precisely what.
On the victorious run, I have a castle in my north and I proceed south, counter-clockwise around the ring. And *2-3 other AI players* also proceed to fight their way around the ring counter-clockwise as well. We're all simultaneously moving around the ring in the same direction -- but I'm the only player not losing ground behind me. Critically important, yellow goes north from the 3 o'clock position and makes big inroads into the diamond at top, and yellow and the AI at top (light brown?) spend all their resources sparring against one another, so light brown never tries to follow behind me and take my castle. Meanwhile, I'm using one L3 troop to slowly, slowly advance around the ring counter-clockwise, making sure to never leave a way to flank and starve my lead troop, even though that means progress is super slow at first. Later, I supplement my L3 troop with a L1 or L2 troop and start making progress slightly faster. Because I leave no way to flank me and my border is protected at L3 strength, my immediate opponent focuses his attacks on his other end, leaving just a L1 or L2 troop or castle (always 1 hex back from the border) to slow me down, but I keep moving forward in that direction. Eventually, I have enough tax base to support 2 L3 troops + 1 L1 troop and now the progress counter-clockise around the ring picks up speed. And then, I can start leapfrogging, leaving behind single safe hexes with no trees in order to make faster progress on the front lines. Eventually, I own the entire ring from 1 o'clock to 11 o'clock (two ends of my territory, not connected) and there is one AI player left who owns the diamond, and of course I've already been offered surrender before then anyway. But I proceed to finish off the last AI player as that's usually the most fun part esp. after a long and brutally difficult map. :^)
(A side note about leapfrogging. I often leave single hexes without trees unconquered so I can advance faster along the front lines. It is possible to use this technique to bypass territories with more than one hex, but I rarely do this. It's dangerous because the enemy territory can suddenly come alive and starve a huge portion of your troops, but you can manage this by carefully tracking how much tax reserves the territory it has built up. The main reason I don't leave territories alive behind me is that it takes more of my resources to conquer them. E.g. a single hex can be taken by a L1 troop at leisure whenever you have one to spare. But a double hex takes a L2 troop and a L1 troop to fully conquer, and that L2 troop costs more to maintain. In any case, leapfrogging can be used to bypass any size of enemy territory, and sometimes a map calls for leaving a large but fully intimidated enemy territory behind while you stave off bigger threats on the front lines, but usually, I use this trick with single hexes. If I have a L1 troop that can take a hex on the front lines and be protected by a L2 or L3 troop adjacent to it, that's a much better use of the L1 troop than going back and taking a safe hex in the rear. OTOH, if my choice is between upgrading that L1 guy to L2 in order to advance on the front lines or taking a safe hex in the rear, I'll usually take all the safe hexes in the rear before upgrading as that, I think, will accelerate the war effort overall and minimize the number of turns to victory. Lastly, on a related strategy, I love using a fleet of L1 troops to attack on the front lines. The AI often leaves borders completely unprotected, assuming that players will be too intimidated to attack him because ooh, he has a L3 troop somewhere. And the AI is too chicken to attack usually. In these cases, I love taking a whole ton of L1 troops and doing a "swiss cheese" attack on his territory, leaving as many single hexes behind as I can and denying him as much tax base as I can. When I run low on L1 troops, I might put a L2 troop near the front line (but not on the front line) so he needs to use 2x L3 troops to take my L2 troop. By this point, the enemy as lost so much tax base that he's toast, and I can fill in the holes of the swiss cheese at leisure or right away, whatever suits my needs.)
Conquer the island
1 post • Page 1 of 1