XBLA Catan Hints & Tips & Strategy

[Tonner Cyn]

Managed to beat Catan on medium last night and moved to hard. I didn't really notice that much of a difference. Not that I won or anything . I think I managed 5 or 6 victory points. I still don't get how the computer can amass so many resource cards. Maybe I just have bad spots? It is better to use the resources as soon as possible, right? Because otherwise I can just see getting nailed by a soldier card coming into play (as has happened a few times).

[dogbert]

I still don't get how the computer can amass so many resource cards. Maybe I just have bad spots? It is better to use the resources as soon as possible, right?

 

Depends. There's a careful balance on when to use resources vs hoarding but it's always dangerous to have more than 7 cards in your hand. All it takes to lose them is a roll of a 7 (one of the most common rolls) or a soldier card, which make up the bulk of the development deck.

 

As for amassing resources, it's all about your initial placement combined with the luck of the draw. I won a game on medium last night despite only having access to three resources initially (brick, stone & sheep) because the sheep was space 6, so it came up again & again, getting me enough resources to keep trading in for the resources I couldn't expand to get. If I'd tried to place my initial settlements closer to the other resources, I wouldn't have gotten as many cards.

 

You know if you hold both triggers down, it shows a point value at each 'corner' to show the relative value of placing a settlement at that position?

[Tonner Cyn]

You know if you hold both triggers down, it shows a point value at each 'corner' to show the relative value of placing a settlement at that position?

 

I did not know about that one. I've used a lot of the other button/trigger commands but that one will be interesting to try out. I wonder if I've been picking good value spots or if I've been putting myself in the hole from the beginning.

 

On a side note, has there been enough discussion (and will there continue to be) to warrant a dedicated thread for Catan?

[Anthony G]

You know if you hold both triggers down, it shows a point value at each 'corner' to show the relative value of placing a settlement at that position?

 

Sweet, nice tip!

[dogbert]

More tips, some of which are "duh":

 

- early in the game, brick + wood are by far the most needed for those roads, but in the 'end game', stone & wheat are much more in demand.

- there's usually much more sheep available than any other resource.

- if you've got a monopoly card, and you have some of the resource you're planning to get the monopoly in, trade them away for other resource types before using the monopoly

- ports are very useful & often game winners in that they allow you to churn through resources to get the ones you need without trading.

- if an opponent's sitting on a development card, expect it to be a victory point.

 

Mark? Got tips?

 

I find the 360 version isn't worth playing on easy - 1:3 trades by the computer just kill the challenge. Hard on a 4 player game is nice & crowded

[Covak]

Some thoughts...

 

- You can hold the left bumper to see the resources in play, and you should always check that before trading.

 

- Don't copy my bad habit of double tapping Y to get to the port trading screen, as this sucks everyone else into the regular trading screen (only to see you leave it immediately).

 

- If your initial settlements have 2 or 3 numbers in common then you may be in for a really rough game. No matter how great the numbers look, if they don't come up early you'll be screwed in no time. While you're looking at the numbers and the dot counts and the resources and all that, also consider which numbers you're already on.

 

- If you're going for largest army, remember that you have to wait a turn to play development cards (except for victory point cards - you win right after buying if you hit 10 with one), and you can only play one each turn, and you'll probably get some other handy ones while you're at. Start early!

 

- If you're not going for largest army, you better be going for longest road!

 

- The ability of the robber to stop a hex from producing is *far* more important/powerful than his ability to steal a random card. If you're trying to win, you're best robber placement is usually where it'll hurt whoever's closest to winning (besides you) the most. Of course there are other considerations (like if you're competing with someone else for longest road or largest army or a settlement spot, or if you'd have good odds on stealing a card that could be huge for you), but much of time you simply want to slow down your biggest threat to give yourself a better chance against them.

 

- Take everything I say with a grain of salt, because I get smoked all the time

 

I find the 360 version isn't worth playing on easy - 1:3 trades by the computer just kill the challenge. Hard on a 4 player game is nice & crowded

 

And make them all Sun Tzu!

[Mark E]

- early in the game, brick + wood are by far the most needed for those roads, but in the 'end game', stone & wheat are much more in demand.

 

This is a maxim of Catan, but I find people follow this way more closely than they ought to. In fact, Catan progresses so quickly that the switch from beginning to end can come out of nowhere.

 

The best suggestion I have for Catan is to stay fluid. This, too, is a "duh", but I've played many games with players who had one strategy and they didn't adapt it for anything and they get clobbered trying to achieve a goal that's made impossible by either the board or the players.

 

Last night, for instance, all the damn brick ended up on with 3's and 4's on them. Because of that, there is no good way to focus on brick early on because you could literally be waiting until the end of the game for it to get rolled once. Still, we all got a hex of brick and dutifully waited it out, but I (and I'm pretty sure a bunch of the others did as well) focused on getting wheat and ore instead to upgrade what I did have to cities.

 

That way, when the brick finally does come around, you're not getting just enough for a road or a settlement, you're able to do both. That's the sort of long term that you have to keep an eye on. Catan's rules are not flexible, but everything else about it is, and you gotta roll with a lot of punches.

 

Brian's point about the ports is a good one too; never underestimate a good port. A port is also practically essential if you're forced to place in such a way that you are missing a resource because most players are going to notice and give you the shaft in trades because it'll pretty much bog you into the back of the pack for most of the game. The 2:1 ports in particular can be a godsend in helping you go around this stumbling block and get you what you want.

 

Oh, and I'd like to throw out there that while Longest Road is worth a lot of victory points, getting into an arms race over it with another player is often (not always, to be fair) going to result in the two of you wasting a lot of resources kicking each other in the balls while your opponents run rampant over the big picture.

[dogbert]

By the way, another reason not to play "easy" - the game will fix dice rolls to give players resources if they haven't had any for a while... There's a great post on Qt3 from Brian Reynolds talking about the randomness of dice in Catan:

 

int Catan::roll_dice (int who)
{
 LEADER.decide_soldier_before_dice();

 if (get_winner() >= 0) return dice; //BR// Don't re-roll dice if game is over

 if (teuber) { //BR// Dice deck
   int roll = teuber_deck.get_dice_roll(teuber);
   die[0] = die_rolls[roll][0];
   die[1] = die_rolls[roll][1];
 } else {

   int reroll;
   do {
     reroll = false;
     die[0] = CATAN_RANDOM.get(6) + 1;
     die[1] = CATAN_RANDOM.get(6) + 1;

     if (learn_as_you_play) {
       BHASSERT(!is_multi());               //kinda sketchy local player checks below, sanity check tutorial is single player only
       if (MYLEADER.roads.len() == 2) {
         //if it's not the local players first turn actually randomize the roll allowing everything but a 7
         if (who == WHO && turn > 2) {
           get_non_robber_roll(); //allow everything but a 7
         } else {
           //use fixed rolls
           die[0] = tutorial_fixed_dice[who][0];
           die[1] = tutorial_fixed_dice[who][1];
         }
       } else if (MYLEADER.settlements.len() == 2) {
         //if it's not the local players first turn actually randomize the roll allowing everything but a 7
         if (turn > 2) {
           get_non_robber_roll();
         } else {
           //use fixed rolls
           die[0] = tutorial_fixed_dice[who][0];
           die[1] = tutorial_fixed_dice[who][1];
         }
         //get_non_robber_roll(); //allow everything but a 7
       } else if ((dice_log[8-2] < dice_log[6-2]) || (dice_log[8-2] < dice_log[7-2]) || (dice_log[8-2] < dice_log[9-2])) {
         //Give them some 8's for brick at reasonable intervals
         die[0] = 3;
         die[1] = 5;
       } else if ((dice_log[2-2] < dice_log[12-2]) || (dice_log[2-2] < dice_log[11-2]) || (dice_log[2-2] < dice_log[3-2])) {
         if (!CATAN_RANDOM.get(2)) {
           die[0] = 1;
           die[1] = 1;
         }
       } else if (dice_log[10-2] < dice_log[4-2]) {
         die[0] = 5;
         die[1] = 5;
       }
     } else if (!CATAN.is_multi()) {
       //BR// On "easy" level human player never goes more than X turns without income
       int who  = 0;
       int whom = 1;
       if (LEADER.is_human() && (CONSTANTS.easy_level_max_turns_without_income >= 0) && (LEADER.turns_without_income >= CONSTANTS.easy_level_max_turns_without_income) && !LEADER2.is_human() && !LEADER2.get_personality_difficulty()) {
         int dice_total = die[0] + die[1];
         int any_income = false;
         for (int scan = 0; !any_income && (scan < LEADER.settlements.len()); scan++) {
           int v = LEADER.settlements[scan];
           for (int scan2 = 0; scan2 < VERTEX_TILE; scan2++) {
             int t = VERTEX.tile[scan2];
             if (t < 0) continue;

             if (TILE.dice != dice_total) continue;
             if (CATAN.robber == t) continue;

             any_income = true;
             break;
           }
         }
         if (!any_income) reroll = true;
       }
     }
   } while (reroll && !QUIT);
 }

 dice = die[0] + die[1];

 //CGN: record recent dice rolls
 if (num_die_rolls < NUM_STORED_RECENT_DICE_ROLLS) {
   recent_die_rolls[num_die_rolls] = dice; //forst 8 rolls just fill the arrays
 } else {
   //after the firat 8 we slide the values down and insert the most recent
   for (int scan = 0; scan < NUM_STORED_RECENT_DICE_ROLLS -1; scan++) {
     recent_die_rolls[scan] = recent_die_rolls[scan+1];
   }
   recent_die_rolls[NUM_STORED_RECENT_DICE_ROLLS -1] = dice;
 }

 dice_log[dice-2]++;
 num_die_rolls++;

 //NEEDED: ANNOUNCE: show & announce the dice

 String string = S("The dice roll number is {$NUM0}.", 61819138);
 string = string.parse(string, dice);
 CATANLOG.add(string);
 CATANLOG.roll(dice);
 CATANLOG.add(EMPTY_STRING);

 if (!QUIT) {
   SOUND.play(SC("DICE_ROLLING", 7122881));
   UI_MAIN.roll_dice(die[0], die[1]);
   wait_for_dice_roll();
 }

 if (learn_as_you_play && turn == 1 && who == WHO) {
   UI_MAIN.tutorial_popup.exec(TUTORIAL_DICE_ROLL);
 }

 return dice;
}

[GreenMonkey]

I played the demo of this last night on learning mode. Seemed pretty cool. A bit simplistic coming from a guy that used to play Romance of the Three Kingdoms III all the time I keed!

 

I liked it. I'll pick it up when I get some spare cash.

[Ron]

- there's usually much more sheep available than any other resource.

 

I played a game last night where the 4 sheep spots ended up being 2, 3, 11, and 12.

 

Sheep were worth their weight in gold!

 

I lost due to betting on ore (on 6 & 9), but then getting boned out of an ore port by one player, then a 3:1 port by another player... :bh

 

Ore for sheep!? ... dang...

[ChrisBardon]

Tried a ranked game last night and other than one player dropping out after two turns (I didn't notice he'd been replaced with an AI until a few turns later), it wasn't that bad. The lag between turns was pretty brutal though, and nobody was overly talkative, but it played better than I expected. How is it that 6 came up 4 times more often than 8 though? Dammit, I should have won that game... You can have the best placement in the world, and crappy dice rolls will get you every time

 

I'll second Mark's suggestion that fluidity is king-I usually try to keep the number of cards in my hand down below 7, so it's often a case of having to build *something* to get rid of them. I'm still getting the hang of the modified rules, and not having to deal with the Commodities race that the expansion introduces. I'm also starting to think about playing more towards the development cards, since there are 5 victory points to be had there, as well as 6 decent event cards, and a bunch of soldiers. I didn't realize until my second game that victory point cards are held until the end (in Cities & Knights, you have to play em right away).

 

Looking forward to getting a game going with some of you guys though.

[ChrisBardon]

Anyone know what the Catan auto update is supposed to do? I notice the one online game I played after applying it was a lot less laggy (there used to be a noticeable delay between turns where you couldn't do anything), but I'd only played one online game before that. Could have just been a fix for that.

 

After playing a few more games against the AI, I've noticed a couple things about it that piss me off. They tend to be really predictable with things like where they'll place the robber, and that they'll always start off going for longest road, and only start mining the development deck when they run out of expansion room. I also don't like that they'll just arbitrarily decide to stop trading with you when you get within 2 points of winning, even if those points are concealed as dev cards. Yeah, you don't want to help someone win, but sometimes you still want to trade with the lead player to make your own gains and pull ahead.

 

Tried out a couple of different strategies yesterday with some mixed results. Playing a wheat baron isn't as effective as I thought it might be (since wheat isn't that essential), and some degree of expansionism is absolutely essential, since with 4 players it's way too easy to get painted into a corner and not have anywhere to expand.

[Covak]

2 unplayed dev cards does not equal "concealed" victory points. If I'd been playing I wouldn't have traded with you, either

[dogbert]

I also don't like that they'll just arbitrarily decide to stop trading with you when you get within 2 points of winning, even if those points are concealed as dev cards.

 

That's a perfectly normal strategy. As Covak says, if a player is sitting on dev cards, it's a damn good guess that they're strategy points. I certainly wouldn't trade with an AI that's sitting on 7 points with a few dev cards in his hand.

[Mark E]

Me neither. I won't trade with players who have 8VP unless I'm in dire need, I dunno why I'd expect the AI to do anything different. Unless it was, you know, retarded .

[ChrisBardon]

Me neither. I won't trade with players who have 8VP unless I'm in dire need, I dunno why I'd expect the AI to do anything different. Unless it was, you know, retarded

 

I won't usually trade with someone who has that many VP, but I also won't arbitrarily refuse to. I suppose I'm just too used to playing with the same group of people-I have friends who will purposely make it so that someone else will win, just to keep another person from doing so (i.e. if they can't be the king, at least the kingmaker)

 

I suppose it's "general rule" or "convention" vs adaptive strategy.

[Ron]

Would you really want to play with an AI that might arbitrarily decide to help another player win the game?

 

It would be like playing with a bunch of women who had become bored with the game... back me up here Tom & Mark.

[ChrisBardon]

Would you really want to play with an AI that might arbitrarily decide to help another player win the game?

 

Not necessarily, but I'd like to play against an AI that tried something a little random now and then. Even if I'm losing, I tend to get bored with games that rely heavily on convention. Maybe it's just that playing on live or with AI, you miss the great social interaction that comes with the board game.

 

I'll keep playing more games against the CPU though-it could prove me wrong. As it is, the CPU players seem pretty good, and are definitely able to kick my ass quite readily on hard.

[Mark E]

It would be like playing with a bunch of women who had become bored with the game... back me up here Tom & Mark.

 

ARGH. And yes, king-making (or, in the case of our Drakon games, queen-making), is one of the lamest things in the world .

[Ron]

... Maybe it's just that playing on live or with AI, you miss the great social interaction that comes with the board game.

 

Agreed. It's a lot harder to subtly turn two AI or remote opponents (v. people you can see) against each other while they ignore your prosperity. And that's the best part of any multi-opponent game, in my book.

[dogbert]

ARGH. And yes, king-making (or, in the case of our Drakon games, queen-making), is one of the lamest things in the world .

 

But it's unfortunately really hard to avoid in most boardgames unless everything is utterly level.

[Anthony G]

Damn this game is fun. It started off nice and friendly. I go up 8 points and then Mark, Tom and Chris get nasty with soldier cards and a trade embargo.

Bastards. Nobody would give me wheat!

[Mark E]

The perils of being the leader, or the perceived leader, Anthony .

[General Zot]

Is it me, or does the AI on Hard seem to cheat like mad? I mean the actual dice rolls dry up every time I'm ahead in a 4 player AI hard game. Others having this experience? Also, they seem to really, really like putting the robber on the player - even if the player is way behind the leaders.

 

Frustrating. I finally beat a full hard game last night, took a good dozen trys though.

[dogbert]

The actual dice rolling code is posted on the forum, Snakefish, no cheating available. Personally, I'm playing with a dicedeck on medium/hard.

 

And yeah, the AI is going to gang up on you & abuse you with the robber if it thinks you're ahead - note that the AI is good at figuring out that you're winning even if your points total isn't 1st.