Designer James Baillie first published this article on his online community Exilian.
I’ve been modding and designing computer and tabletop games for years in various forms, and whilst I’m certainly an amateur at both arts I thought it’d be interesting to share some of my thoughts on just one or two ideas of how these areas can link together.
Many of the features of good board and tabletop game design are pretty similar – indeed a good boardgame should be fun to play on the computer against a good enough AI (with the caveat that it often won’t be possible to build one). The reverse is less true, because computer games can sometimes use a level of calculation that you can’t reasonably mimic at tabletop level – a computer can, analogously, roll hundreds of dice and use the results every second without breaking a sweat!
Ultimately, both board and tabletop games can have similar sorts of goals, along sets of categories common to both – they can be games of skill, games of strategy, games of chance, games of story & evocation, or (most usually) some mixture thereof. I’m just going to pull out one or two thoughts on each of these elements, though they all deserve articles in their own right (and may get them in future if people are interested)! Elements of skill we’ll skip – these generally have the hardest parallels to draw, because they rely on an often very specific physical action or timing technique that can’t be replicated in a different game let alone a different system.
Strategy in computer games is usually played against an AI rather than another player – and this, oddly enough, is one of the areas where observing tabletop play can be most helpful. Artificial intelligences for playing games, to state the obvious, aren’t human. That often means that it’s really easy to make an AI that’s more than capable of beating humans, in most games – well designed AIs can see everything that’s happening in the game, calculate what to do hundreds of moves ahead, and control more things far more simultaneously than a human could manage. Such an AI is miserable to play against. We want AIs that act like humans, and that goes beyond simply ensuring that the AI makes blunders sometimes or is “slow” enough for the human to face.
Humans, for example, tend to have particular styles – perhaps preferred moves to use or units to train – that they will use even if not optimally calculated for the situation. They will also have particular sub-goals that they set themselves and try to achieve, often ending up in tunnel vision situations rather than recalculating every turn. Perhaps most importantly, they interact with other players in a human way – for an AI that can only see the scope of an individual instance of a game, it may not make sense to over-punish betrayals, whereas for humans used to the idea that there will always be another game, vengeance to teach a lesson is strategically common. Equally, friendships and alliances have lives of their own beyond mere calculation among humans, and that “stickiness” is similarly something that can be observed and then, potentially, replicated.
Moving on, there are numerous ways to present chance in a physical game. The most common tend to be dice and card decks, though spinners and other randomisation methods can be used. Dice and card decks are quite different – a die represents the closest analogy to a random number generator in a computer game, which is the usual mechanism for adding randomness there. Die-roll results are independent; that is to say, every single time you roll there’s the same outcome of getting, say, a 5. Computers, as mentioned earlier, have a lot more random number “power” than a human rolling a die, as they can roll dice with arbitrary numbers of sides and in as great a number as their designer chooses, but the systems are essentially similar.
It’s easy to assume that the same is true of a shuffled deck of cards – you’ll pick cards out and there’ll be a certain probability of any given type of card depending on what was shuffled in to begin with – but actually, if you know what cards have been drawn previously and you know what’s in the pack, you can subtract those from your mental list. If I have five cards – two green, two red, and one blue – I have a 40% chance of drawing a green at random. If I draw a green card first, assuming I haven’t shuffled all the cards again, the chance of me getting a second green card is now down to 25%. Now, this feature of a card deck doesn’t make it an inferior system, but it makes it a different one. The advantage and disadvantage of cards is that you know that after X goes, where X is the number of cards, you will get a certain type. Imagine the above scenario, and imagine that I win when I pull a blue card. If I pull a green, then a red, I as a player can actually be more hopeful about the third result, because I know my chances have improved to one in three, which wouldn’t be the case if I just used random numbers. As such, deck-type probability systems if the player knows that’s what’s being used, can be psychologically helpful for players. If you’re designing a computer game, medium you can mimic this kind of system easily by creating an array and “shuffling” it with a random number generator.
Finally, let’s think for a moment about storytelling in board and computer games, because it’s here where I think the two often have most to learn from one another. Boardgames often have to be far more minimalist in the way they tell stories – you have a limited range of pieces and options for players, and yet get those to immerse the players in a compelling way. This is worth thinking about for computer game designers, because it lets you strip down to the things that most evoke the theme you’re trying to go for – if you have just a paragraph of text at the start of a rulebook, or the illustration and title on a card, and you’re trying to evoke whole characters or settlements or cultures with that, what are the elements you really need in place and what’s just filler? Looking at how boardgames deal with the problem might help you find the answer.
On the other hand, boardgames can occasionally learn a thing or two about what can work in the more flexible computer-generated world, and think about how to replicate that on the tabletop. Computer game design can often afford to give players more depth of story and world, given the presence of (usually) a single player over a longer timespan. The ability to simulate a full setting in closer detail, rather than “boiling down” specific aspects of human decision-making to abstract rules as boardgames often must, also lends itself to certain levels of depth and consistency. Boardgame designers can perhaps learn from these more fluid environments how a story can be told with the freedom of simulated worlds, and then try and consider some of the reductions of those; as computer game design matures as an art in its own right, it becomes a stronger source of ideas for how players and their worlds can interact and stories can develop, and these things can then be brought into the often faster, sparser, more social worlds of board gaming.
I’ve only hinted in this article at a few ways that board and computer game designers could learn from one another – if you’ve got more thoughts, please do comment below or get in touch and I may try and write a follow-up article with more thoughts on this area. Hope you enjoyed reading!