Recommend
1 
 Thumb up
 Hide
8 Posts

BoardGameGeek» Forums » Board Game Design » Board Game Design

Subject: Modular Testing rss

Your Tags: Add tags
Popular Tags: [View All]
Andrew van Laar
Canada
flag msg tools
Avatar
mbmbmb
Hey All,

So I was reading a post (See Post Here) about how to go about play testing and balancing a game.

Some of the posters mentioned that they break down the game into the individual components and rules and test them on their own. If they work, they then incorporate them into a full and complete game to test alongside all the different components and rules. The posters refer to this as Modular Testing.

I am new to board game design and would really appreciate if anyone would be able to explain this type of testing further because I just can't grasp how one would test a game in this manner. It seems impossible to me to break up a game and test parts of it outside the whole. If possible, maybe an example of a commonly know game (Monopoly, Risk, Settlers etc) could be used to show how this approach would have been done with that particular game.

I really appreciate any feedback on this and I am sure there might be other new designers (or old ones who don't know much about this!) who could glean a wealth of information. If I am the only one who doesn't understand this method, then I shall sit here in blissful ignorance until I am enlightened by a fellow BGG!

Thanks so much!
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Arias
United States
Sanford
FLORIDA
flag msg tools
Avatar
mbmbmbmbmb
One designer's approach (can't recall blogger's user name here) was to iteratively develop the game.

e.g. If he decided the game was to have a modular tile map, he would playtest JUST that feature over and over until it "felt" right mechanically and entertainment-wise. Validate that modular tiles was the best approach, and shake out any contingencies that would create weird rules exceptions or other complexity to optionally design out.

Then if there was some kind of card draw, add that feature next, but only with a few cards and a few card types. And test that feature + its interaction if any with the map mechanic (e.g. If there were specific tile/card interactions).

Vs. for a reasonably complex game putting in all the features and testing all of them at once, and trying to tune things so it "feels right" regarding pacing, tension (or whatever emotion you want the game to create), fun factor, etc.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Andrew van Laar
Canada
flag msg tools
Avatar
mbmbmb
Hey James,

Thanks for your response!

Using your example of a modular tile map, what exactly are you testing? This is where I get stumped. How do you test a board type/style when the other components that get played into the game have a big affect on what type of board you use.

For example, if I am designing a monopoly clone and just try to test the board, without knowing that I am going to be able to purchase properties on the board, how would I test the board out? Without knowing all the game components and ways players interact with the board, it would be impossible to test the board out no?

I must be missing something... or be terribly dense.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Arias
United States
Sanford
FLORIDA
flag msg tools
Avatar
mbmbmbmbmb
A practical example...

Like almost every other Geek here, I am designing my own dungeon crawler.

I wanted a "one-off" game type with no overlord/GM where the map was randomly generated, not scenario-based.

After looking at a gazillion other crawlers (many of which pointed to as exemplars by the community) I decided I wanted a minimal stack of room/corridor tiles of various sizes and some kind of card deck (not rolling dice on tables) to determine the next Tile.

Playtesting just this feature exposed a lot of issues
- ensuring good looking maps ... e.g. No corridor-corridor-corridor or room-room-room layouts.

- preventing "map wanders off the table"

- keeping the set of tiles to a minimum (SWIA: looking at you) while having just enough for a good looking map.

- what is the right size of map anyway? How many rooms? How many straights, turns, T/Y/+ intersections


All of the above was via look at other games' approaches, ponder on what would I do differently vs. copy, work out an approach, then test (trial and error) to see how it felt.

And this was before I added Heroes, Items, Hazards, Traps, Monsters, Treasure, etc.

Where I landed was a "deckbuilding" approach similar to 90s Warhammer Quest, but where each card has some tiles selection logic to prevent goofy layouts, with the "right" tile set and mix arrived at via experimentation. Has worked very well thru lots and lots of playtests.

Now if I was doing a basic roll & move game, where the board was static, I would have done NONE of the above.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Corsaire
United States
North Carolina
flag msg tools
Avatar
mbmbmbmbmb
Imagine Catan...

Three main subsystems:
Resource distribution
Building
Card collecting

You could test resource distribution by placing towns and rolling the dice many times while adding new towns. Then review the final resource distribution. You can "what if" various board positions. You could do this mechanically or simulated.

Test building by taking turns laying roads and towns and cities and upgrading.

Similar for cards.


It's slightly stranger for light games. Imagine your game has a hand of cards and an auction system and board control. Each element could be a light game itself.

P.s. This is playtesting you do yourself or with a design or development partner, not public playtesting.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Andrew van Laar
Canada
flag msg tools
Avatar
mbmbmb
With both your examples I THINK that I am beginning to understand better.

So this method is for more heavy games rather than really light ones first off. This makes sense as some games have so little intricacy, to strip it down to smaller systems really doesn't make sense.

The point you made Corsaire, makes sense in that you play each component as an individual game by itself and see if it works. If it does it goes into the larger game to be tested with the other components.

I think the problem is I am a very visual person and seeing a video or someone do this in person would help me understand what they are actually doing during this testing. I don't expect anyone to make a video, haha, but maybe that is where I am struggling the most to grasp this.

Thank you both for your explanations
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
marc lecours
Canada
ottawa
ontario
flag msg tools
mbmbmbmb
A lot depends if you can split off one aspect of the game from the others. For some games it is easier than others.

Another way is to just test the beginning of a game or the end of a game. In a current project, I am just testing the set up and the first 3 rounds (out of 12). I am looking at how the base mechanics work, whether the game is unfair after 3 turns, how fast the game plays, whether I can simplify any aspect of the game, etc.

If I was testing just the end of a game, I would start from a position and just look at the last round or two. I would be looking at whether someone who is behind can catch up (there is no point playing the last turn if no one can catch up to the leader), whether the game is immune to bashing the leader and kingmaking.

Even in a game like Monopoly, you can roll the dice 1000s of times to see which areas are more likely to be landed upon (I think it is the Orange properties). You can distribute the properties with hotels to different players and roll dice to see how long it takes someone to run out of money. Or you can test out how many properties are bought after twice around the board (and how long it takes). Etc.

With "Risk" you could certainly test combat with various numbers of armies fighting each other.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Michael Love
United States
Florida
flag msg tools
Hi, Andrew. I'll expand a bit on the type of modular testing we're performing with our 6 mm war game.

When we first began developing this game there was no consideration of publication; we just decided to develop a game we liked to play. And so our testing reflected that. We tested by playing the game. Though we didn't call it testing and we didn't target specific rules or situations to validate. We just played and as a consequence, many rules we scribbled down were barely tested.

When we decided on publication we knew we had to adhere to more comprehensive testing. To do so we created a thorough checklist of every rule and situation we could think of that would come up in a game. For example, we performed standalone tests with engines of war, formations, individual spells, terrain, shock charges, and much more. As we tested each individual rule and situation we determined if the result met our design goals and if we were happy with it. If not, we either continued testing, modified the rule, or removed it completely.

Some of these situations we are testing may never come up. Others rarely. Still, they had to be tested and simply testing by playing a "real" game wouldn't necessarily do that.

Of course you are correct in that you have to account for the rules as a whole, which is why we perform the modular tests first and then move on to testing the game as a whole. Our theory is that modular testing provides a good foundation from which to build the rest of the game.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.