2013 note: This page is currently a brainfart hurricane and work in progress.
2019 note: A new project.

Thoughts on board games


I like video games with a focus on world building and simulation, and the same is sort of true for board games. When I play a board game, I like to see the world in front of me, and the game mechanics shouldn't be too abstract or "gamey". Back in the '80s I remember enjoying:

I might also enjoy Mage Knight, Runebound, Pandemic... Recently I've played some Small World, Terraforming Mars, Terra Mystica, Arcadia Quest, Blood Rage, Stone Age, Agricola, Power Grid, and Ticket to Ride. Some of these are a bit too abstract for my taste.

Game plays me vs I play the game

Some might say that Monopoly is a terrible game because of the roll and go element leaving the player to the wind of the dice. I partly agree, but without that element I don't think I would have enjoyed the game as a kid. It sort of evens out the players a little. If I lose I can always blame the dice, at least partly, and not feel too stupid. Losing at Go or Chess is very different from losing at Monopoly. Randomness can keep the game lighthearted and eventful. In games like Talisman I'd prefer a bit more control though.

My ideal board game

Skeleton Fortress

Choosing a project

At first I wanted to do a Talisman game with a fun world drawing going around the board, coming up with regional monsters and a melange of heroes. Then I remembered Drakborgen and my own old versions of it and wanted to do that instead. I always liked the general idea of that game more than the implementation, so I'm doing my own thing here rather than trying to clone the original. I'm keeping the general idea of the game with the adventurers moving through the generated maze, drawing cards here and there. The original was very focused on the greed element, get in, grab gold and get out. My version should be more focused on general adventuring where each session has a different objective and the players might also have minor private objectives.

Drakborgen photo from Wikipedia.

Skeleton Lords is a sort of all encompassing (in terms of generic fantasy races) setting that I doodled on back in 2006-2008, and I decided that I might as well build on that to give it some momentum. I have some 40 factions which are spread all over the game world, but for this game I'm thinking that an ancient psychic being has drawn/lured/tempted/invited a selection of them into its fortress for some reason. The brave adventurers have to enter the fortress and vanquish the ancient one before it causes trouble.

Budget concerns

Board game development doesn't seem lucrative, because of the physical aspect. It has to be a passion project. Breaking even is perhaps as good as it gets. A lot of people gets a piece of the cake, like prototype production companies, the manufacturer(s), freighters, promoters, distributors and retailers. Computer games (indie) are different. Setting up a web store to sell digital copies is easy in comparison and you don't need to invest in the initial print run. If there's a bug it can be easily fixed in a computer game with a patch, but with board games you can mess up and a cheap manufacturer far away can mess up as well. At the end you have a product which costs something like 40 to 100 dollars and if the game is ok I guess you might sell a few hundred copies as an indie.

With this project I care a lot about the world that I'm creating so taking the publisher route feels like it could easily compromise what I want to do. I'm guessing my game would cost over 5 dollar to produce, representing an investment of >10k for 2k copies, and a smaller profit since I can not sell it for more than 10 or so to the next guy (distributor). Shipping from the manufacturer would probably eat most of this profit, so I couldn't even pay myself for the development time.

But whatever, this is all theoretical and I'm doing it for fun. Bla bla Kickstarter.

Dices... I've been lusting for a custom D12, but a D6 and a D2 will give me the same result the way I'm doing it. The D2 could be something as simple as a custom D6 with 3 blank "0" sides and 3 "snake eye" sides so it might be pretty cheap as you often pay per side. More about the dice later.

The rule booklet would not be hurt much by being in B/W. I think the important part is to illustrate it properly and make it fun. I've seen a lot of color booklets which have tiny photos and too dark low contrast color images, superfluous little border and layout doodads, etc. With black and white it has to be clear, simple and readable.

Miniatures are probably better reserved for some deluxe kit (along with the special D12). Nimble cardboard stand-ups will work fine (1:28 heroic, 1:72 scale, or game scale 1:144?). Nothing big like in Talisman. Thumb joint sized stand-ups with a distortion-free transparent plastic base might feel okay.

Vacuum formed thin plastic shells are often used to keep the game components from rattling around. Tooling these is quite cheap (can be done at home) compared to plastic injection. It would be cool if these could be used to hold the cards in an elegant way, both during a game (for distribution) and for storage. Some kind of loose trays perhaps. Snaps together like a clam?

One really interesting thing I saw recently was a game which used the clam shell box as a playing area, opened and flipped upside down so the sides of the box formed little walls, which would suit a fortress game. Apparently sturdy boxes are expensive to print so it makes sense to make use of it. I can imagine that printing the inside of the box (or the unorthodox extra pieces which are folded in) costs even more though, so the gain isn't all that big perhaps.

New regions could be put into expansions. A region is basically some 25-40 maze tiles and maybe 50'ish small cards, and a booklet, and a box. Maybe also a new character. I'm thinking small flat boxes. A thick cloth bag for storing the tiles and drawing them from would be great. Color (muddy) coded with a screen print on so the player can keep track of it.

Concept Art

Loose doodles for enemies. Some Orchs with a Bonner'esque feel can be spotted. And that is Cthulhu as designed by Lovecraft.

The euro-mini-size card format is a bit too tight but probably cheaper to produce. I'd prefer 50x70mm but that might require special tooling. The underlined 10 and 8 on the card is the important numbers to look out for with my current battle system. The rest of the numbers are the for special cases and to make the Troll seem more like a playable character and not a piece of card with two numbers on. I included a scale drawing of the Troll and its weapon to liven things up.

Most early enemies give one EXP point, though there might be rare, weaker ones which flee easily but give 3 EXP points or perhaps a nice treasure. (e.g. Metal Slime from Dragon Quest). Larger enemies yield 2.

Dungeon Quest movement and encounters

I forgot how things worked in the original Drakborgen, but I'm going my own path anyways:

There are several methods for laying down the maze tiles in a game like this. They can be placed on the grid board in advance and the players simply flip them over as they go. Once flipped a tile has to be oriented so the marked entryway faces the direction that the player came from. Some people put the maze tiles in a bowl or bag and pulls them at random. Others keep them in small piles with the anonymous side up.

Moving into an unexplored tile consumes all movement and is quite likely to trigger an encounter at the new tile. You decide where to move before placing the tile (I don't like being able to change the game physically by choosing after having drawn the tile). Encounters can be monsters, traps or blatant treasure. After moving into a new tile, a die roll of 5+ results in an encounter (draw a card). In the inner maze the chance is greater with a 4+ and you draw from the more difficult encounter card deck.

Moving through already explored corridors and rooms is quicker and safer. I'm thinking character movement should be around 2 to 4 + D2 depending on character. Encounters are less likely. Monsters are assumed to roam around littering the place with trinkets, and different players see different things, explaining why encounters can still happen on explored tiles.

Searching a room is done instead of movement and digging to the next room takes 3 turns (place a bridge tile to the next room). Searching a room could also reveal hidden paths (similar to digging), or trigger encounters because the player is being too noisy. Maybe there needs to be a marker placed on searched rooms. Hidden paths are rolled after the compass table on the table.

Certain rooms have a special appearance, showing lots of garbage, a corpse, or a treasure. These rooms are riskier to search because perhaps the corpse is a living skeleton, and the treasure can be trapped, but the rewards are also greater. Different room types would need different room entry-direction symbols. Some rooms are dark, hazardous (pit, lava, acid, trapped (roll to escape or take dmg)) or partly blocked. Moving through these consume all movement (or a lot). Combat in these rooms is also more random (both combatants add a D6). Some enemies might be born to fight in these environments and gets to add a 6 instead (dark vision, small and nimble in tight quarters, etc). Some rooms are only partially unsafe. In this case combat is played as normal but movement through the unsafe area is tougher.

Monster encounters could have some variation to them. Perhaps a 1 on a D6 means you're ambushed (unless the monster is cowardice), and a 6 means you can choose to ambush or sneak by the monster. 5-6 is combat as normal.

Overkill> Number up and write a little story about / description of each tile, to be read upon entry (optional).


I don't like stressful timers which magically end the game, but I can understand the need for a time limit. In Drakborgen the plot was that after turn 20 the castle would "wake up" and the heroes would get suddenly overpowered. I think an escalation and upheaval is more appropriate. Some computer games have a terrible enemy which appears more often if the player stalls or lingers (Evil Otto in Berserk, Ghost in Spelunker), and I think I like this solution better. So, the players could start drawing more often from the difficult enemy card deck, or there could be a special deck with nasties which just replaces normal encounters (e.g. Not-Shoggoths).

Cards and Decks

When there are only 3 cards left to draw, add the discards, shuffle and refresh the deck. This way there's a chance that you never get that card you're waiting or dreading for, and I like the feel of that.

I'm unsure how to divide up the decks. I'd like to see some tiers so the player can take risks by moving into areas of higher difficulties. Each region would then need:

Also, a few global decks:

Some regions might have special decks for certain terrain features. Shrine cards? Ponds? Doors? I don't think traps need to be in a special deck.


Whilst the terrain is sort of randomly generated, I think the completely blank/open landscape of the original game is a bit dull. Often you sort of zig-zag towards the dragon and move out again, leaving large parts of the board untouched. Here I'm thinking there could be little attractions for the player to be drawn to, such as imp nests with little treasure stashes, shops, block-offs and dangerous shortcuts.

I also wanted to support differently themed regions and different starting conditions for the 4 player entry points. I suppose the board could do this using loose pieces, but they'd fly around and doing the contained regions is difficult this way.

First color art test. The green walls are a homage to the card art of Drakborgen. This kind of green has an old too it. The walls on the actual maze tiles in Drakborgen had a more gray tone though. Unsure about resolution. These are currently 512 pixels which them 512 ppi if 1 inch big. I want the ink to be crisp, so I might go higher. Might be some trick to separating the black channel too.

Note that the board background has little dots of light which serve to smoothen the look of the tile edges. Similarly, each tile has dead ends in all 4 directions. The tiles will have to be arranged in print so open meets open and closed meets closed since the color values are slightly different on the floor.

There will be some sort of symbol marking the entry hallway and also the contents of the room or corridors, though it should be obvious from the look of the tile what's going on. A skeleton marks corpse, a tight corridor or scattered rubble marks difficult to pass. A pit is... a pit to fall into, and furniture may contain loot.

I think DungeonQuest has a rather monotonous game world compared to Talisman with its many regions and special locations (and higher difficulty at the center). One way to implement regions into a DungeonQuest type of game is to divide the board up and place a regional tile in each special region. Once the region is visited the tile is revealed and the appropriate tile and card stacks are then used to populate the region. I'm thinking the game could have little expansion boxes containing tiles and cards for new regions. The maze tiles for each region will of course look different, but they could also build different kind of mazes (open, curvy, lots of debris, traps or doors). On top of this the encounter cards of each region would offer new enemies, artifacts and special rules (right on the cards).

Hmm, maybe the region types should be revealed at the start so the player can plan ahead a bit.

The board has 9 regions and is 15x15 tiles (=225), but a small portion of the tiles are occupied by rubble and otherwise occupied locations (50'ish) so each region is about 20 tiles at the moment. While it's unlikely that the player will ever visit every tile on the board, I like the idea of having a buffer of extra material to add variation and replayability. The player is bound to recognize rooms too easily otherwise, and the game will feel fake. The tiles for each region could be printed on a 6x6 tile and inch sheet (36 tiles). The standard region actually encompasses the area of 4 regions, so it would need 4 sheets (144 tiles). Some of those would be special tiles for bosses (perhaps 9), danger passages, good spots, maybe minibosses.


I think a D6 is a bit tight and linearly random for some things. With a system where I compare values to see who wins, it can be either impossible to win, or it's just sort of random in a way that's a bit frustrating to me (poor rolls are very common with a linear D6). The nonlinear 2D6 has a wider range, but perhaps too much resolution and constantly adding numbers together slows down the game. 2D4 has a nonlinear range of 2-8 but I find that they are a little hard to roll and read.

I'd really like a custom D12 with a range of 1-8 arranged like 1 2 33 44 55 66 7 8, thus making the mid range more likely (I don't care if it's flat). I'd put fumble and criticals on 1 and 8 respectively, so 1/12 chance each, compared to 1/6 or 1/32 for the D6 solution. A single die can be read immediately and if I have it custom made (I'll most likely have to), then I can make icons for fumble and critical.

As for color, opaque speckled sea with gold numbers would fit the theme (copper green is for old and gold is what you seek).

I suppose I could simulate it by forcing a single reroll of a D8's 1,2,7,8, but this gives different probability values. A D6 with its 1 and 6 ends modified by a D2 gives a range of 0-7 with the same probabilities as my custom D12->D8, and the end re-rolls will only happen a third of the rolls. You might think that having a 16.67% chance to roll a 6 isn't much different from having an 8.33% chance to roll a 7, but in the cases where you need to roll perfect and the monster needs to fumble, the probability is very much smaller and the larger range offers these rare chances.

Let's call this extended D6 die... ED6. It has a range from 0 to 7. On a roll of 1, reroll. If you roll 1-3, you get a 0 otherwise you get a 1. Do the same for a 6, but 1-3 counts as 6, otherwise you get a 7. Here a result of 0 or 7 might be a fumble or critical. If two opponents both roll a critical they may cancel out since the bonuses are the same.

Character card

Remember-this-number counters are pretty hard to do. Dices, cubes, sliders, number tokens are all common. Since immersion is important here, I opted for themed wheels with little imp knights managing them, just for the fun factor of turning the wheels and looking at the figures. I don't like when card decoration is just random stuff to fill the space. Should be either nothing (quiet) or fun mayhem. Cut-outs are a bit expensive to tool and I need some kind of hub too, but on the other hand I'm not using a lot of counters with approach. Dice, Cubes and lots of cut-out tokens cost money too.

Peripheral stuff is placed around the character card just like in Talisman. The wheels have a default value (character reset), either at the top or color coded. This way people can modify the base character and play their own version of it.

A sort of example character


Weapons use the Grace and/or Brutality of the user to resolve combat and damage values. These only need to be calculated once when the user levels up or changes weapon (and will likely be memorized if used often). The active weapon is kept close to the character card. Half-stats are rounded up (so half of 5 is 3, just like 6).


A Warhammer has a sharp spike for armor penetration and blunt hammer head for shocking and crushing blows.

Combat value: 		ED6 + Brutality + ½Grace
Forehand Pierce Damage:	ED6 + Brutality + Grace
Backhand Blunt Damage: 	ED6+3 + Brutality 
Stab Blunt Damage: 	ED6 + Brutality + Grace

Assassin's Dagger

Some weapons are more focused on Grace.

Combat value: 		ED6 + 4 + Brutality + ½Grace
F/B.hand Slash Damage: 	ED6 + 2*Grace
Stab Pierce Damage: 	ED6 + 2*Grace

A simpler version could look like this:


Combat value: 		ED6 + Brutality + ½Grace
Damage base:		ED6 + Brutality
Forehand Damage:	+4 P (Pierce)
Backhand Damage:	+3 B (Blunt)
Stab Damage: 		0 B (Blunt)


To resolve combat you first need to figure out the combat value on the card of the currently equipped weapon. The latter is usually a die plus a value plus two character stats of the user. This value can be written up or indicated by counters at the start of the game and whenever any involved variables change.

You also need to look up the combat value of the monster that you're facing. Now roll the dice needed and compare the two values. If yours is equal or higher you roll again for damage. It's possible to best a clumsy armored monster, but then be unable to penetrate its armor. If you won and are able to move to an adjacent space, you can also choose to flee from the monster. Place your figure at the edge of the space to indicate the direction to move next turn. If you lost you can only flee if you roll 6 on a regular D6, but you must resolve damage first (as you're probably getting stabbed in the back).

To resolve damage you need to have your damage values or resistance values ready, and also the respective value of the monster. Some weapons have several edges and can be swung differently, forehand, backhand and stab. If you decide to do piercing damage, roll any dice then compare the resulting value with the piercing resistance value of the monster.

However, this becomes a laborious task of simply finding the highest advantage, so I think it's a good idea to not always give the player a choice in how to strike. The player could be given a choice to either roll a D3 to determine strike type, or to discard an intuition card and choose the ideal strike.

A monster only has one strike to keep things simple and predictable (so a player can skillfully adapt to the probable encounters of his current region). I suppose that in a competitive game the other players could control the monster and the fights could be a little more interesting, but I'll save that for later iterations.

A skeleton will have a higher resistance when being exposed to a piercing weapon because it is kind of hollow. Should the damage value be higher than the resistance value, then one lifepoint of damage is inflicted. A character or monster will die at 0 lifepoints or below.

Design note: Lifepoints are just like rare hit points ("life" makes them sound rare and a little different). With the more rare lifepoints and a binary combat resolution combat should be a bit faster (win or lose) and other players don't have to wait several combat turns since regular monsters only have one lifepoint. The lifepoint value can't be leveled up, making the end game a bit more risky, preventing it from dragging on. Persistence is the physiological and psychological toughness. Brutality is akin to strength and ferocity, and important for determining damage with most weapons. Grace is sometimes also used in damage equations (imagine a well placed assassin's knife. It's akin to agility, dexterity, balance and accuracy.

If the player won the monster is moved to the graveyard. The player gets the amount of EXP indicated on the monster card, and other effects might take place. The player should check his level-up threshold at this point and resolve it.

If the no damage was inflicted then the two parts are locked in combat and its the next player's turn.

Certain weapons could be weaker the second round of combat, such as held spears which provide a good first strike chance but become unwieldy in subsequent close quarters. Let's say -2 (combat roll) if the combat has moved to a second round, unless the opponent also uses a spear. Spears could also get a -2 in blocked areas (tight corridors). This should be clearly written on a weapon card. Some heroes might carry backup weapons but would have to designate which weapon they use before combat.

A few characters have the ability to dual wield one handed weapons. They fight as normal but may choose which held weapon to use for the combat roll and which to use for the damage roll.

Intuition card idea: Before a combat round starts, the player can choose a defensive or offensive stance. An defensive stance gives + 2 in combat but - 2 to damage. An offensive stance works the other way around.


Jayne faces a skeleton and regular combat starts. Jayne looks at her weapon card for the combat value, and also at the skeleton's monster card:

Jayne's Combat value: Brutality[4] + ½Grace[4] + ED6, i.e. ED6 + 6
Skeleton's Combat value: 4 + ED6

Dices roll (another player could roll the monster's die). Jayne gets 8 (2+6) and the skeleton... also a 8 (4+4). Jayne wins because she needs to roll equal or higher. Now it's time to see if the skeleton can be damaged. Skeletons are best killed with a bone-crushing blow, so Jayne wants to the blunt edge of her Warhammer (backhand strike). She fortunately rolls a backhand blow on a D3.

Jayne's Warhammer, Blunt damage: Brutality[3] + 3 + ED6, i.e. ED6 + 7 Skeleton's Blunt resistance: 3 + ED6 (it's unarmored)

The dice doesn't favor Jayne, she ends up with a 8 (1+7) and the skeleton a 10 (7+3). She could remain locked in combat, but she's in a hurry and uses red intuition ability to nudge her die from 2 to 4 (consuming two red intuition cards).

Just a table for reference

Wrote a quick thing in BlitzMax which illustrates the probability of getting equal or larger than a monster with regular D6, D8, ED6, and ED6+D6 for chaotic fight in rubble. Note that the ED6 is more tapered at the min and max than the D8 (as intended). Since the ED6 works a bit like a D12 you see the 12 x 12 = 144 variations. ED6+D6 is more diagonal (vertical), meaning, stats matter less (because of the chaotic terrain). The regular D6 is quite horizontal.

Also do table for simulated combat with Fumble/Crits?


It might be possible to be locked perpetually in combat. In this case, fleeing might be a good option.


If the player is ambushed the monster adds a D6 to its combat value.

Combat value criticals (7) and fumbles (0)

If a combatant rolls a fumble (min on the die) and the other a critical (max on the die), the winner may choose to either deal a wound automatically (1/144 chance for both combatants), or to roll to damage as normal with the chance to deal two wounds.

If both roll a critical they cancel out and the damage is worked out as normal. (1/144 chance).

If both fumble, awkward looks are exchanged and no damage in is inflicted. (1/144 chance).

If only one out of the combatants roll a critical and wins combat, the successful attacker may choose to ignore the defender's armor and shield bonuses (found a chink in the armor) when rolling to damage.

If one combatant fumbles and loses combat, the successful attacker may similarly choose to ignore armor and shield. The fumble die has no effect on damage rolling and damage is dealt as normal (though perhaps more comically).

Critical and fumble values during damage rolling count as regular values. However, some characters might have intuition abilities which can not nudge their own fumble roll or the critical roll of an opponent, so these kind of values still serve a function here.

If the character is wearing "skimpy armor" (this rule would be on the armor card), a fumble during the combat roll would automatically change the opponent's die to a critical. This sounds more like a negative trait an overconfident character could have though (and nude characters wouldn't have this disadvantage). Perhaps skimpy armor only counts as armor on a 4+ roll instead? It's an extra die to roll though.


Each item (weapon, treasure, armour) has a burden value. Once a character carries more than its Brutality value in burden points, its Grace will reduce by the overflow value (e.g. 6 Brutality - 8 Burden). If Grace is 0 or less, the character is overburdened and will have to dump an item. If so, return it to its native deck and shuffle. Perhaps imps ran off with it.

Since some values need to be recalculated, I limited the levels to 4 and I'm thinking a player should only find 0-3 stat changing items per game. These items give significant boosts, so no common "+1 swords" or rings laying all over the place.

Intuition cards

Six of these cards are drawn at the start of the game. They can also be acquired during the course of the game. Some cards just fuel a characters intuition abilities whilst other cards contain abilities of their own.

Perhaps the intuition cards have the intuition color (R, G or B) at the back of the card ("anonymous" side), and they can be used as intuition fuel this way. The player can also gamble and flip them over, revealing a specific intuition ability. They can also prove to be junk/discard or even occasionally the immediate bad effect. This solution saves card space in a way since otherwise I'd mix in the fuel cards with the more defined intuition ability cards. I suppose those could have an ability AND a color, but I think the flip-choice mechanic might be interesting.


Meditation can be done instead of movement (just like a room search). Draw an intuition card.

Game setup


Special things

Moving slowly though a maze and fighting random monsters and getting random treasures will probably get boring pretty quick. When I was a kid I used to play video games more than they deserved to be played because I thought there were hidden stuff in the game. Maybe I hadn't seen everything?

For this reason I'd like to include more rooms/cards than which are likely to be used in a single game. There's of course a limit to the amount of cards which can be printed, but some cards could have special rules such as a table of rare happenings. First you must get the rare card, then there's further branching with a D6 roll. The rare happenings should not be insignificant either. Perhaps the player gets a pet? An overpowered one-use item? Teleportation to random square? Clubbed and thrown in jail with the gear placed in a vault? There could be rooms which challenge the player in some unusual way (a Sphinx/Oracle?).

You could also get cursed, giving you a quest of some sort to rid yourself of the cruse. Get to a place (a pool?), sacrifice gold or intuition cards. There could be shrines like in Diablo 1 which have a random effect. One of your items is replaced with another, your stats are altered, etc. The catacomb mechanic could be replaced by a teleportation event or system of some sort.

Personal quest cards drawn at the beginning of the game could tempt the player with extra EXP and gold. Maybe a village elder has tasked you with finding a particular gem, or vanquish 10 skeletons, or visit a certain room. Each play session could have a different objective for all of the players, such as steal 1000 gold per player to help the poor, or explore 20 tiles per player (map the fortress), or vanquish monsters worth 15 exp per player in a fit of revenge. The game is perhaps automatically won if the boss is killed. There could also be competitive objectives.

Boss runs

There could be different boss cards at the center/end, just like in .. Relic (40K Talisman), I think. It could be revealed at the start, allowing all the players to adapt and make a boss killer character. It could also be revealed gradually to some players throughout the game, or only at the end making it a total surprise to the first player reaching the end (acting as a slight catchup mechanic for the other players?). Maybe all three. Collecting clues in a completive game is pretty fun (3 coordinates in a cube, each a minor clue such as "fire", "beast", "many small" or "frost", "undead", "illusion"). These 3 clues could be put into some treasure deck (and be player-tradable), or just placed in the boss space for people to peek at (no one can squat a clue). Once the first player arrives at the boss,it is finally revealed using the coordinates of the cards. If the player is strictly cooperative, players could attempt to meet up to share their clue (they're not telepathic). This is risky since it uses up valuable time and moving through the fortress is dangerous. Alternatively, they could be telepathic since no other actions are hidden. It already takes time and movement for them to search for the clues.

Before the game starts, a dice roll could determine if the boss is (1-2) already revealed, (3-5) a mystery (clues), or (6) unknowable (until encountered). This could make each game a bit different tactically. For this random boss mechanic to make any sense, it must be possible for the players to adapt their characters to the opposition. Maybe there's a shop which sells items (such as a dragon slayer sword), or locations where certain items are likely to drop (such as a fire resistance ring).

An alternative simpler solution could be to simply place a random boss card at the center, but if the player finds an altar and meditates there s/he can have a vision, revealing the boss.

Art by Arne Niklas Jansson