Prototype: A New Approach to RPG Adventure Mechanics

Read This Design Philosophy!

I’m making a new kind of RPG* and that means doing things different than any other RPG before.  That’s hard.  It’s risky.  Oh, and involves a LOT of exploratory rework.  It’s much easier just to copy what has been done before because it is proven/known.  But the results of this approach should be worth it.

(* Technically I’m making a Mage Simulator but since no one really knows what that means, I stick with the less accurate but more broadly known term “role playing game”)

No [common RPG] feature is being put into Archmage Rises without first being analyzed, taken apart, and put together in different configurations to see how it could work better.  I know of no other way to innovate.

I also have some lofty/ambitious/ridiculous goals: not just to revolutionize RPGs by putting in more of the good stuff from pen & paper, but to also make it play FAST so a deeply satisfactory role play experience can be had in 4-8 hours. 

I have three small kids under three.  I’m lucky if I get an hour each night to play any game.  That game better be quick to get to the awesome while at the same time being deeply satisfying.

FTL, Hearthstone, and Card Hunter show this is possible.  I can have a deeply engaging gaming experience, then pop out to change a dirty bum or put the stuffed ducky back into the crib from whence he was thrown.

All the Flavor, Now Less Filling!

If you have ever cooked with a broth, you know about bouillon cubes.  They are small cubes that look like a d6 but taste like chicken, vegetable, or beef.  These little cubes started life as a big vat of broth, but were boiled down and boiled down to just their purest essence.  You can then buy 1 Liter of taste in about 2cms of packaging.

When Blizzard crated Hearthstone they spent a lot of time analyzing Magic: The Gathering, and other collectible card games.  They wanted much of what makes Magic Magic (the taste), but they wanted it to be faster, more accessible, and more applicable to the digital play space (less filling).  So they did some radical things:

  • No mana/resource cards.  Heretical!
  • Cut the deck from 60 to 30 cards.  Outrageous!
  • No counter spells* during your opponents turn.  Inconceivable!
  • In competitive magic play, you generally have four copies of 15 cards making a 60 card deck.  So Blizzard said “Hey, let’s just make it two copies of 15 cards”.

I’m not going to argue if Hearthstone is better than Magic.  As a lapsed magic player, and a game store owner who sells magic singles to many adoring fans, I think playing Hearthstone feels just as good as playing Magic.  If it isn’t identical, it is close enough.  Let’s not argue which piece of chicken tastes more like chicken.

*I said earlier there are no counter spells in Hearthstone.  That is both true and untrue. 

Magic allows you to interrupt your opponent on their turn by playing instants or counter spells.  They can then in turn respond to your instant with their own instant.  And you can keep interrupting each other someone runs out of cards.  It can get pretty thorny in the middle of someone’s turn to figure out the exact sequence in which all these things happen.  It’s even harder to implement on the distributed internet.  How much time do you give a player to play a counter? 0.5s?  5s?  Boy that would bog down the game!

So Blizzard created something new they call “Secrets”.  These are cards played on your turn that trigger on your opponent’s turn and only on a specific kind of action.  Actions like: casting a spell, bringing out a minion, attacking a minion, or attacking the hero.  The opponent doesn’t know what the trigger is, only that there are one, maybe two or even three secrets lying in wait.  This gives the threat and emotion of counter spells, without all that nonsense of timing and sequence because it is boiled down to one-to-one. 

Just because something has always been done that way, doesn’t mean it is right.

Hearthstone proves it is possible to go into an established game genre, turn it upside down, cut the fat, and make something that tastes amazing.

What I’m Doing

I’m making a game out of RPG bouillon.  I’m boiling down RPG tropes to their essence and then putting them into a fast accessible deep game.  And one of those areas I have to get right is in the Dungeon Exploration.  I call it Adventure mode.

Before D&D had 1” squares and pre-painted minis, you had a Dungeon Master who simply told the players what they saw, and the players made decisions on what to do.

Now here is what I have concluded:  Whether you are playing Skyrim or Pathfinder with scale maps and miniatures, however you move in a dungeon you are simply going from encounter to encounter.  This is an important concept to understand, so let me break it down with examples:


The party moves to a T intersection.  The GM puts down the next two pieces of the map, one to the east, one to the west.  The players deliberate, maybe cast a spell or make a skill check, then choose to go west.  At the end of the corridor is a locked door.  They use spells, skills, items, or brute force, and they open the door.  They enter a room.


The player walks down a corridor to a T intersection.  Looking to the east or west they see it goes off into the darkness.  They choose to go west.  They walk for a bit, maybe some height changes or the corridor weaves, but ultimately they end up at a locked door.  They use magic or lock picks to enter a mini-game and unlock the door.  They enter a room.

In both games the players did the same thing.  One probably played faster than the other before getting into the room.

Here is what I notice:

1) Walking around is just white space.  Walking around is not terribly interesting, in fact, it is a necessary evil to getting from something interesting to something else interesting. 

2) Both games play linearly.  No matter what choices you make, you encounter “interesting stuff” in a sequence, one at a time. 

3) Fun is in making interesting choices, not walking in a straight line.  I call an interesting choice in a dungeon an “encounter”.  There were three encounters in my little example: the T intersection, the locked door, and the room.

So what if a game just doled out the encounters and skipped all that walking around?

Now, I hope you understand what I’m trying to do with Adventure Mode in Archmage Rises.  Skip all that navigating whitespace and get to the good stuff: decision points.  Encounters.

Adventure Mechanics Prototype

I will try to describe the Adventure mechanics concept as clearly as I can:

You have a dungeon.  It is a container for interesting encounters.  How many encounters?  That depends on the size and difficulty of the dungeon. 

Let’s think of an encounter as a card.  A card can be overcome (completed) by interesting choices.  Some cards are simple and require few choices, some are complex and have many choices. 

If an encounter is a card then that means it could come from a source of many cards called a deck.  A deck is for cards that share a similar theme.  What kind of theme? Underground, ruined temple, under water, fire caves, spooky forest, whatever we desire. 

This means that a dungeon can be created by picking a theme deck, shuffling it, then drawing the required number of encounter cards.  The more cards, the longer and more difficult the dungeon.  The less, the easier.

This also means that no two dungeons are ever alike.  If you don’t know what to do in a particular dungeon you can’t call and wake me at home because I don’t know either.

Now the makers of Skyrim tried this approach with Daggerfall.  Unfortunately, they were dealing with 3d geography from a set of templates and it all got very repetitive.  So the lesson here is that if you have to navigate a virtual space, the random shuffle of cards is difficult to pull off unless you have a ton of cards.  And this is precisely what Bethesda has been doing up to Skyrim: making smaller and smaller cards that fit better and better together to not look repetitive.  Because I’m using text instead of 3d modelling, I think I can have more cards than Bethesda.

Now unlike a board game where you really are limited to physical cards, I’m using virtual ones.  Which means a card can have something randomly interesting.  Instead of being attacked by one orc, it can be 1-6 orcs.  Instead of a door of difficulty 5, it can be from 1-10.

I can push this further by adding modifiers to the cards.  If I were to take just three decorators, like say, a bubbling magical fountain, a flower with noxious fumes, and a strangely colored potion, they can be added to any card.  So instead of it being a 10ftx10ft room, it’s now a 10ftx10ft room WITH a flower of noxious fumes.  Without the flower it’s a pretty obvious choice to search the room for loot.  But now, do you risk it?  Will the flower effect you in some adverse way?  You may now see that with 10 base cards and 4 modifiers suddenly we have 40 possible cards.  Granted, finding a noxious flower in 25% of the encounters is pretty boring, but I hope you can see the beauty and possibility of a properly stocked system.

So this is what I need your help doing: testing this theory in practice.  It’s a big one, because if the adventures are boring and suck, then no one will play Archmage Rises beyond the character creation screens.  But if it is engaging and wonderful in the way ROGUE, ADOM, and NETHACK are, then people will play Archmage Rises for the rest of time.  Or at least until the next Blizzard, Bioware, Bethesda, or some other company name starting with B is released. :-)

This Prototype Sucks!

This is as early access as it gets folks!

Now that you understand what I’m trying to do, lets make sure your expectations are sufficiently low.  A prototype is built to answer a question:

  • Does a mechanic work or not? 
  • And if not, why not? 
  • And what could we do to fix it? 

Whether a prototype leads to including the mechanic or completely abandoning it, it has succeeded because the question has been answered.

I whipped up this prototype in the tool I work fastest in: C# WinForms.  It’s not the right place to make a game, but for prototypes it’s ok.  Inherent drawbacks are there is no “game loop” so I cannot animate things or display things in gamey ways.  There is also no “juice” or pizazz either.  Things are not as clear or exciting as they will be in the real game.

Very Important Instructions for the Prototype

Early playtesters are reporting the game is crazy hard!  I'm sorry! It isn't balanced and the procedural generation is "swing-y".  All i can say at this point is don't play the prototype to try and win.  You probably won't.  Like a pure Rogue-like, just try and see how far you can get.

The size of the Dungeon determines it's difficulty (number of cards to get through).  Start with the smallest ones first and work up to the larger ones!

When your hitpoints reach 0 you are dead.  You can recover hitpoints at the healer in town.

Stamina represents your energy available for spell casting.  If you reach 0 stamina while in a fight you immediately die because you cannot cast.  If you reach 0 stamina adventuring and run out of light you cannot cast more so you die lost in the darkness.

Light is how long you can dimly see.  When exploring caves, dungeons, or ruins your ability to see well, remember your location, and move safely over rough terrain is paramount.  You cannot move in the dark, therefore if you run out of light you will be told you cannot proceed until you cast Light.

Time Left represents the amount of time you can stay awake and explore before collapsing.  Everything you choose to do in a dungeon takes time.  Returning back to the entrance takes roughly half the time it took to traverse it the first time.  There is a point of no return where turning back and heading to the entrance will simply take more time than you can remain awake.  At this point, you must press on to complete the dungeon and find another exit.

The bottom of the screen shows an event log telling you what happened and usually why.  The latest info is at the bottom (last line). You can scroll up with cursor keys to see what happened, say in a combat that took all your stamina and health.

All skill tests are done by rolling 2 6-sided dice.  So if a trap has a difficulty of 9, that means you have to roll 9 or more on 2d6.  Some spells add bonuses to skill checks.  All skill checks are automated and called as needed.  All skills are equal in the prototype.

Spells are how you do cool things.  You are a mage after all!  You either know a spell, or you don’t.  If you do, it starts at level 1.  If you find another spellbook of a spell you already know, the level will increase by 1.  Higher level spells have either a better effect or lower casting cost.  I only designed spell effects up to level 3.

Some encounters are nearly impossible to overcome without the appropriate spell.  If you are at a cliff and don’t have Fly or Featherfall, the time wasted and damage taken from failed climbs will probably kill you.  Turn around, go home, and try another dungeon until you get that spell.

As you explore a dungeon, the map is displayed on the right.  Here are the map colors:
White – current location
Gray - Unexplored
Purple – Room
Brown – Door
Dark Gray – Cliff or Gap
Blue – Pool
Reddish Pink – Trap
Yellow – Entrance/Exit

Many of the helpful spells (detect traps, detect items, light) have a specific duration.  To see which spells are currently active look at the Active Spells area of the screen, it says what is active and how long in hours its benefits will last.

Poison is a status effect that will drain your HP over time.  I can only be cured in town.

Searching rooms results in treasure, spells, or both.

You have one chance to run from a fight.  If it works, you'll be in your previous location.  If not, you have no choice but to fight.  Gotta love permadeath!

Town Screen


You begin the game knowing two spells (arcane bolt for fighting and light for seeing in the dark).  You have some spending money from which you can buy more spells.  Don’t blow all your money as you may need some of it for healing and resting if your first adventure is unsuccessful.  You absolutely want a Knock spell as these are the only thing that can get you through doors and many dungeons have doors.

Here you can spend your hard won gold on new spell books, healing, and recovering stamina by sleeping at the inn.  The stock of the magic store changes every time you return from a dungeon.  Duplicate spell books can be sold in the magic store for differing prices.  They have the same effect, you simply found one at a discount.  There is a mouse over tooltip for spells in the magic store to see what they do.  There is no need to cast a spell in town.

There are 5 dungeons created at the start of the game.  Try and complete all 5 before dying.  You can resume a dungeon you ran away from at any time.

Deficiencies and Important Release Notes:

I’ve abstracted away combat into the simplest form that I could think of.  So when you are attacked the game simulates a certain number of rounds until you win or lose, spending stamina and losing hitpoints.  The game will automatically use your best combat spell (arcane bolt or fireball) when attacking.  If you have Arcane Shield, it will automatically absorb incoming damage, you do not need to cast it manually.  In the real game combat will break out into full tactical battle.  You can see what happened in combat by scrolling the event log at the bottom of the screen.

In the real game you determine how “strongly” you cast a spell and its effect is determined by your power level.  The stronger you cast a spell the higher the chance of a miscast.  This just wasn’t worth recreating in the prototype so I made the concept of spell levels: the higher the level the better the effect or lower the stamina casting cost.  No miscasts.

There is no saving of progress: win or lose in one play through.

The game is unbalanced: it will range from impossibly hard to far too easy.  But most of the time it is challenging for a while.

The dungeon creation algorithm needs plenty of improvement.  Sometimes it thinks it is making a T intersection but only goes one direction so really it is just a corner.  Sometimes there is a door or a pool at a dead end; disappointing to overcome the obstacle just to be face first in stone wall.

(Thanks to the testers for bringing all these frustrations to my attention.  Sorry my game caused you so much pain!)

So you want to be a Hero?

You can download the Windows only prototype here: (works for sure on 64bit Win10, probably 8, 7 and others).  Requires .NET 4.5.2+

Until I decide to take it down.

(Note: Your browser may give you some warning like below.  This is because I haven't signed the EXE because I don't want to pay for a certificate just for this.  So click the Keep option.)

1. Play it a few times and see if you can complete all 5 dungeons. 

I don’t know if anyone will actually do this, but the best is if you can record a Let’s Play video of you playing while you talk through your thoughts as you go.  Upload it privately to youtube and send me a link through the contact form.  That gives me the best view into your experience which I will use to make decisions on how go forward.

2. Then go to this short 5 question survey and give me your honest feedback:


Thanks for your participation you rock!