Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Designing around a core mechanic
by Charmie Kim on 06/12/12 03:04:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


It's easy to plug game mechanics into your game design, but it's not always obvious whether those mechanics suit your game. Here's a simple framework to help you understand and evaluate your design from the core out.

Reposted from blog

In our wild wild game design world every designer is likely to have their own shade of design methodology, lack thereof being a kind of methodology of its own. I'd like to share a bit of mine.

When I was still a game design student in Vancouver, I was taught this life-changing design tool by a mentor (giving credit where it's due!) over lunch at a White Spot. At least one mind was blown in that generic family-style restaurant that day. I've been whipping this tool out to evaluate every bit of game design I do ever since. Trust  me, it's amazing.
The tool is a deceptively simple diagram that I call the 'Core Diagram':

A concentric diagram with 'Core Mechanic' at the center circle, 'Secondary Mechanics' on the second circle, 'Progression' on the third circle and 'Narrive or Hero Myth' on the outer most circle.

In this model, the core mechanic is at the very center and forms a nucleus for your game. The other mechanics form layers around the core, with the narrative forming the very outer layer.

Theory-crafting game designers love to define words, as do I, so let's not skip that bit! By mechanic I mean a system that facilitates interaction, and by interaction I mean a kind of conversation between the player and the game. Neither of these words actually amount to what games actually are, because a game is the experience generated by those words when they get put in a disco with the player's brain and circumstance. However, until we invent neuro-technology that can transfer experiences directly from one brain to another, what us game designers can control within this dance are the mechanics. The mechanics are the paint and paintbrush, the nail and hammer, the two girls and cup of our art!

But still, it's probably not very clear what exactly is a 'Core' mechanic in a game.  Easiest way to understand it, I think, is in relation to time.

  • The core mechanic in a game will usually be the purposeful interaction that occurs the most frequently. In a platforming game, this is usually jumping. In a shooter, it is usually shooting. In a racing game, it will be driving. Another way to determine the core mechanic is, if without it, you wouldn't be able to play the game at all.
  • The secondary mechanics are the interactions that happen less frequently. They could even be layered out from more frequent to least frequent.
  • Progression systems form the mechanical envelope of the game, being the source of change within the game system at a holistic level.
  • The Narrative layer is the outer most layer that puts all the inner layers within it into context.

Gameplay and Innovation

Now that you understand the model, could you guess which games each of these core diagrams represent?

4 core mechanic diagrams that map the mechanics of 4 games

The answers are:
A. Super Mario Bros.
B. Portal
C. Flower
D. Every fantasy RPG ever made :)

There are some qualitative observations that can be made immediately, just from looking at these examples.

  • The best games usually have a very strong core mechanic that is easy to grasp but provides room to expand upon. It also helps if the mechanic has a powerful meaning to it of its own - there's a good reason why shooting is such a popular core mechanic in our field.
  • The most effective games are ones where each layer compliments the other. You can test the relationship between the layers by seeing what effect each layer has on the other. i.e. "In order to remove enemies, I must jump, and in order to progress through levels, I must remove enemies." If your layers don't have this kind of gating relationship going outward, and contextual relationship going inward, you may want to re-consider your design!
  • Truly fresh experiences often result from innovations at the core of the game. For example, Flower is to this day one of my most memorable game experiences because I'd never played a game that made me feel so much like I was flying in the wind. It had an unusual core mechanic, and it did that mechanic extremely well.
  • Sometimes innovation comes from having an unusual combination of layers, for example, the shooting core mechanic won't normally be paired with solving puzzles. But Portal did it, and did it well.
    Also consider how Portal differs from shooter games that have puzzles on the side (puzzles that do not use the shooting mechanic in order to solve them), and how effective those experiences are in comparison.
  • Some combinations of mechanics are truly timeless, such as D. It's like a classic dish in French cuisine - it tastes good, and it's hard to mess with.

Social and Mobile

In the last year or so I started looking at social and mobile games in this light, and again, it's really fascinating to see how they map.

Mechanical mappings of 2 games

Let's play guess the game again! Ready?
A. Angry Birds
B. CityVille
Now some more observations!

  • The biggest shift in design caused by new platforms and audiences are in the Core and the Narrative layers. Removing pigs is no different from removing mushrooms, and completion or unlocking mechanics have always been staples in progression design. This is really interesting to me because I understand it to mean that the Core shifts mostly with new interfaces or platforms, like the touch screen, while Narratives shift because of the different players that the games target. But otherwise, game design is still game design!
  • Angry Birds is an awkwardly designed game. Flinging relates to removing pigs, but the relationship is indirect, and sometimes feels arbitrary, even. This also makes relating flinging to completing levels rather awkward. You know that strange feeling you get in an Angry Birds level where you have that one pig off to the side that you can't seem to get, and you're madly playing fling trial-and-error to get it? Yea, it gets a little awkward, and not fun! Something else that gets left out in this diagram is the points system, it just doesn't fit very well with the other layers. Removing pigs gets you points but you have to remove them anyway so it's redundant, and the points are needed to complete the levels but in an entirely arbitrary way!  Every game designer in the world has their own opinion on how Angry Birds got to be so big, but I think I have proof here that it ain't the design ;D
  • In comparison, CityVille is amazingly elegant within the inner 3 layers. Look how tightly collecting currency weaves into buying buildings, and collecting XP weaves into unlocking buildings, which weaves back into buying buildings, and then again, weaves back into collecting from them. Beautiful! But, there is still a weakness, and it's a big one. Exactly how does clicking buildings to collect from them (it's not even made very clear that they are supposed to be taxes) and unlocking buildings (again, messaged in a very 'game-y' way with buildings unlocking at every level) make you a better mayor? CityVille could do well with some tweaks in how it integrates its overall narrative.
  • I don't have a diagram here for all the Zynga games but my biggest beef with them is that almost every one of their virtual world games (other than their newer Indiana Jones game and the hidden object game) have the same 3 inner layers - collect/harvest, buy stuff, unlock stuff. It's like they know how good it is and they wanted to explore that same design until noone wanted to play it anymore. :)

I haven't tackled how social/multiplayer fits into all this, that would be a post of its own. But a good measure to go by is, a truly social game would require more than one player involved at each of the layers. If I were to make a Zynga game more social, for example, I would make collecting an activity done with friends (this is already the case), buying buildings would be in relation to friends (for example, if I buy the Fashion Design Studio building and you buy the Clothing Boutique building, I could supply you with clothes for your building and we could split profits, right?) etc. Is it any wonder MMO games like WoW are so powerful? They take the classic RPG formula and apply social dynamics every step of the way.

Strategy Games a.k.a. The Slow Core

The core mechanics I've looked at in other games so far have a physical 'fun' to it on its own. A good designer working on a platformer would pay a lot of attention to the physics of a single jump so that the core activity feels good even without the secondary mechanics or progression. Yet, it would be a mistake to think that every core mechanic needs to have such a twitchy tight singular loop. Looking at strategic games, for example, the core mechanic is often 'unit placement'. Physically speaking, there's nothing innately joyful about placing a unit in a strategy game, but look at it as a cerebral activity and it sheds light on how deep and meaningful this core mechanic can be and why strategy games are so much fun. Note also that with strategy games, the core mechanic is far more complex and involves lots of different feedback loops within it. In other words, there's a lot more information being processed within the interaction right in the core!

Multiple Cores and Modal Shifts

I would also add a caveat here and say that not all games fit this mould so well, and those are some of the most fun. Many successful games do modal shifts where you go from one core diagram to another. This works really well, I think, if one set of mechanics is more twitch and the other more relaxed, and the modal shift is used for pacing. A great example of this is one of my favourite game franchises of all time, Mass Effect!

I hope this tool is as inspiring for you as it has been for me, at the very least I hope you find the musings interesting. Try mapping some of your favourite games and see what the diagram can teach you through them. Are there any games that really don't map at all? Let me know!

Charmie is a game designer currently working independently at Funstorm Games
You can follow Funstorm Games on twitter here.

Related Jobs

Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer
DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer


Altug Isigan
profile image
Very structured approach, really a great tool for analyses. I have a few reservations though: First, in some games it is very difficult to speak of a core, or to make a clear distinction between core and secondary mechanics. In mind come games like The Sims or Civilization. So maybe it is useful to keep in mind that the distinctions in the model are analytical distinctions. In the end, games are larger than analytical models, and I hope that this model will develope over time to be more inclusive and to possess more explanatory power when used for more complex games. Second, I would refrain of thinking of the narrative layer as something that is being "wrapped" around gameplay. A core mechanics is already a core function or event in the narrative sense, and I would like to see this model evolving into a direction that recognizes this dual sense of player activity.

Besides these, I can only say thanks a lot for sharing, I wish I would have had a mentor like that.

A side note: When I saw the diagrams in your article, I immediately remembered how they resemble diagrams in regard to character growth and development. Please check out this article and have a look at the diagram towards the end:
Might be an aspect that would be interesting to connect to the diagram that you present here.

Charmie Kim
profile image
Thanks for the comment :)
Calling the narrative layer a wrapper was probably a poor choice of words, I should probably edit that. I think the ideal design has every layer form a relationship with the other 3 layers, the integration of the layers is key.

Yes you are right there are definitely some genres that ill-fit. Civ I've managed to feel good about abstracting the core as 'unit placement', with unit creation and unit action as secondary mechanics. The Sims is trickier but I think 'fulfilling needs' would constitute a core mechanic there. The genres that I've been having the most trouble with actually is the time-management genre and the point-and-click adventure genre, and weirdly enough, Uncharted 2 and 3. I definitely hope to see a greater model that explains Uncharted 2 for me some day.

To interject, I think I'm just too green at the game theory scene, I'm not sure what you mean when you say that the distinctions are analytical. What is the alternative?

I find your diagram in your referred post very interesting, thanks for linking it. I've recently started thinking of the player as sort of an actor within the stage that is the game, and how everything we build seems to be the system equivalent of 'what would this character do given his/her situation' in a story, except it would be more like 'what should this character be allowed to do given his/her situation'. So to see your distinction of 'to be' and 'to do' is something to think about. Also the idea of having sort of a left-brain and right-brain to the core diagram.

Joe Cooper
profile image
Reducing games is a hobby of mine and I think I can add to this thread.

Firstoff, the Sims is one of my favorite games. I think fulfilling needs is a good description of the core Sims dynamic and it can be done with two actions; "go" and "start task". Everything is built around tasks which consume time and have to be done at certain points. The game becomes most interesting when one is coordinating multiple units to perform tasks that benefit all. All this together makes up the core game, around which you can add leveling and other perks, all of which affect Sims' performance at tasks. (As in, how likely they are to succeed and how effective they are at fulfilling needs.)

Really, I write this to illustrate that it -can- be reduced to its elements and that it -has- a "tetristic", mechanical design side.

The Sims Online actually drops the above mechanics and you can see how much it suffers without them. I enjoyed Sims 2 for months and even enjoyed Farmville for a few weeks, but Sims Online just was dead to me in five minutes.

"The genres that I've been having the most trouble with actually is the time-management genre"

Do we mean like Farmville?

I've called Farmville a "clock game"; you are essentially setting timers and it can engage players when they must think about their real-life schedule and weigh this against aesthetic\resonance choices. (e.g. "I like pumpkins more than corn, but when will I be back at the computer?")

"the point-and-click adventure genre"

I would say they're closely related to academic tests.

Phoenix Wright my favorite example. Its core is extremely similar to SAT-style reading comprehension tests.

You're presented with massive loads of texts (cleverly disguised as engaging stories about lawyers and ghosts!) and tested on your ability to spot inconsistencies between characters' claims.

Myst is very similar but, even though it has lots of texts (presented in "books"), the tests are very lateral and abstract...

But the upshot is that they're something like academic tests wrapped in art and story.

People have trouble with them under many models because they don't have situations emerge dynamically from systems like Tetris, Sims or football. But IMHO they fit nicely under your model with the "core" being that they're scripted intellectual tests.

Charmie Kim
profile image
Good point Joe re: Sims vs. Sims Social (I think you meant to say Sims Social the facebook game not Sims Online the MMO :D I always make that mistake too)

Actually I meant time-management as in games like Diner Dash. It's ironic because I'm currently working on one and I eventually decided that the core is simply... 'management actions' which is not entirely satisfactory for me but it seems to work ok.
I came to a similar conclusion as you with point-and-click adventure games as well, that it had to do with 'trying' or 'guessing' options.

Altug Isigan
profile image
I completely agree with integration between all layers being key. And that's actually what I tried to point out by saying that we must be careful not to forget that the distinctions we make here are analytical. We make these useful distinctions so that we can digest what's going on when we make our analysis, but in reality none of the layers is in a vacuum, and their relations will often span many layers at once, because there is a structural relationship between them. The risk in forgetting about the distinctions being analytical is that at some point we may start to think of the categories we invented for analytical purposes as facts, which could cause our approach as designers become more rigid, making us "slaves" of the tools we created and causing mental or creative barriers, because the concepts we rely on have an impact on our perception, interpretations and the way we "work".

Let me just say againg that I find the model very useful and I think it works exceptionally well for most of the classic arcade games and todays casual games, and your examples illustrate that wonderfully. More complex games would require some parallel or overlapping core diagrams, showing how several cores are related to each other. Some years ago, Ernest Adams wrote a very nice article here on gamasutra about multi-layer games, and it looks to me that in some games one could come up with a core diagram for each of the various layers of a complex game. One could then further analyse how these multiple cores "feed" each other with the values that they generate. For example when you think of The Sims, managing the characters in the game, and building/altering the home seem to rely on completely different cores, yet action on one layer would ultimately also affect the other layer, and only when both are related to each other in that way can we really experience the narrative "wrapper".

In regard to the character being an actor within the stage: I agree with that notion, and personally I even draw a line between the player as a real person and the player as an "actant" who is represented within the game world through a variety of audio-visual clues. And I think that adding the left-brain right-brain thing is a great idea, something I couldn't think of until you brought it up, because I'm really not familiar with the cognitive science aspect of games :)

The issue with the point and click genre is its capacity to put many different cores under the hub of a single type of interaction. The prime example for me here is Diablo with its wonderfully crafted interface. As a real player you basically do just that, pointing and clicking, but the meaning of the clicking would change depending on what the cursor rests on at the time of clicking. So the design manages to place three core "narrative" actions (walking, slaying, collecting) under a single hub, and it is therefore incredibly fluent and immersive. I find it also a genius act on the part of the designers to divide the player's representation into two entities, cursor and avatar. Yet a player always feels "whole", not even noticing the divide.

Jerome Goomba
profile image
This is very interesting, yet I agree with Altug "in some games it is very difficult to speak of a core, or to make a clear distinction between core and secondary mechanics".

A lot of game are mixing genre thus it is difficult to extract one core mechanic.

Jitesh Panchal
profile image
I like the approach of evaluating and making design based on diagrams like these. Especially, I love your take on Angry Birds and the inherent slow-core usually associated with Strategy games.

Tjien Twijnstra
profile image
Thanks for this article. I'm a firm believer of designing around a core mechanic, and love the clear approach you outlined here. Thanks!